BARRY MARTIN wrote:

KM> Assuming you can get it to install and run in the VM. You should
KM> hear my VM War Stories....

I don't play all that much but have had a few instances where it just
doesn't play nice. I've also been using Raspberry Pi's for a disposable machine. No, not going to toss the Pi itself, but can overwrite the SD
card and so remove all traces of the failed experiment.

If I'm just experimenting, I use real hardware and a stack of small hard drives... much more reliable results. One PC, fifty OSs. <g>

But when I need to run something that the host OS doesn't like... that's
why I do VMs.

The two main use cases being:

Host: any linux
VM:
-- WinXP, for a handful of WinApps I can't replace and for reliable
networking

Host: Win8/10
VMs:
-- WinXP so my eyes don't bleed when I need to work there all day

And the later experiments, in search of this interface between usability
and compatibility that is so absent in Certain Later Windholes...

Host: Win11
Still trying to come up with a combo that works on the Win11 netbook. XP
runs so slowly... 15 minutes later I was still waiting for the desktop
to finish loading.

Win9x/NT4 cannot use the Guest Additions, so can only run at 800x600
(which is awfully small on a 1920x1080 monitor) and does not see shared folders. That's not useful.

So far Win2k installer keeps looping back on installing hardware, so
that didn't work.

Host: XP64, which limits it to VirtualBox v4.3.12
The main object is to get a few things working that need a newer OS.
VMs:
-- WinXP32 (for the DOS app I can't live without) -- works fine, tho I
had to install from scratch; it would not import my existing "appliance"
even as a naked vmdk (you can un7zip an .OVA file to extract this).
-- Linux Mint -- runs fine but only at 1024x768.
-- PCLinuxOS -- a couple of last-year's-models will install, but throw
up all over the Guest Additions, reducing the screen to 800x600!! May
try setting it up as Live only, and save the machine state instead of installing it. The installed version does not like the virtual video
driver, judging by where it likes to stall.
-- Win8.1 -- runs fine, tho it makes my eyes bleed. However, now I have
a working FTP client without having to schlep files to a Win1x box
first. Chrome itself won't install, but Supermium (current Chrome for
old Windows) runs fine. (I'd forgotten how braindead IE was back then...)
-- Win10 won't install.
-- ReactOS just for giggles...runs okay but only at small screen size,
and I think there have been a lot of usability regressions since the
previous version (which was actually pretty good).

Did you see the crazy thing (I think it was) MJD did, with VMs inside of
VMs until it went all the way from Newest Windows to Oldest Windows??

> Yes, the 'rolling upgrade process' can be a litte detail-filled!
KM> "Rolling" is supposed to refer to the software, not the server
KM> cabinet!

Only when forget to set the brakes on the cabinet's wheels!

OUT OF THE WAY, BERTHA IS ROLLING!!!!!

KM> I've come to greatly prefer a rolling distro for an everyday
KM> desktop. There are NO UPGRADES and therefore NO REINSTALLS.
KM> Someone took the oldest extant PCLOS from 2010 to 2021 (I think
KM> it was) using only the rolling update process via Synaptic, and
KM> only twice did he have to stop and twiddle something at the
KM> command line. But normally it's just click-the-usuals, go away
KM> for a while, and it's done.
KM> The other advantage is that problems get fixed NOW, not when the
KM> next major version rolls around.

That option has advantages and disadvantages. I do like the 'fixed
now', assuming it doesn't break something else. (Of course the latter
is a possibility any time.) Ubuntu does have LivePatch, which is
probably your fix-it-now.

Regressions have been pretty rare. They all do some degree of updating between, but the real point is there is no such thing as a release
version, because it just keeps trundling to the future without the need
of silly version numbers.

Around here I'd prefer manual version upgrades, meaning to go from
version 22 to version 23, not the more-minor updates. Have had old
computers no longer work properly with upgrades: IMO not a fault of the
OS, though one could say it didn't check for compatibility.

Yeah, that is a problem. And yes, it should do a compatibility check,
but that's not sexy programming, you should just do a reinstall and stop bothering us.

KM> The downside is you have to do your updates regularly, because it
KM> can get out of sync if you let things go too long. (Tho Synaptic
KM> at least is pretty good at squaring things up.) But it's fast
KM> enough, given it's usually small, to do 'em often.

Agree: though with the limited experience I have it seems Ubuntu does a reasonable job: it seems one can take a 'basic' version (say 22.04.03),
which does the installaion and original versions, then at the end of the installation ask for the updates and it will d/l umpteen files, od the
magic, and now 22.04.39. (Making up the numbers.)

Well, my Fedora setup started at v32, and is now v40, with KDE upgraded
from v5 to v6, tho that's still early enough that there are minor
oddities. OTOH it did fix a longstanding visual bug in my preferred theme.

KM> Speaking therewhich, I need to do the twice-yearly version
KM> upgrade on the Fedora system, which I keep forgetting because
KM> it's an overnight job (sometimes two nights) and takes some
KM> babysitting before it gets going. Grrr.

And if you're like me the babysitting is the major part of the delay!
Good reason to have two monitors so can have one on the upgrading system
and the other for while-I'm-sitting-here-do-e-mail, etc.

Fortunately it's not a system I'm trying to use while it's doing this
(in fact I hardly use it at all) but I still had to wait for the
repository list (which was about 100mb and dozens of files) then two
hours for the download, then half an hour for the update, then repeat
this for the full upgrade, except the download took the rest of the
night and the upgrade itself about an hour.

I have no idea how one upgrades Debian, if it can even be done...
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

Boo!

Boo who?!!

...Thought I figured soming out but now makes less sense. If a 'boo' is
a boyfriend/girfriend/lover ("he/she is my boo") and 'boo' is said by a
ghost or to scare someone, why is it good to be a 'boo'?

KM> It's the description of the fault that tells me it's most likely
KM> bad video RAM.
KM> That's quite distinctive -- you get a solid mass of random ASCII
KM> characters. We used to see that a lot in the Olden Tymes.
KM> When the card is loose, or the GPU is failing, you get random
KM> lines.
Ahhh! I don't recall the random characters part. The make-and-break of
the data stream would cause the card to try to make sense of it and output what it thought it was told.

I think it happens when the amount of usable RAM is so
constrained (probably to just the first address bank) that it
can't process more than the first few pixels, and the result is
Random ASCII Characters in 25-line textmode (usually the bottom
32 bytes of the ASCII table at that, so it's hearts and spades
and such), having displayed all it managed to scrape up.

That makes sense. I'm thinking it could also mean a faulty bit which is causing a bad instruction and so garbage output. That one based on
years ago when I built a Heathkit TV and the overlay display (time,
channel, etc.) was sometimes scrambled. A little analysis and only
certain letters: let's say "C" so "Channel" might display as "Thannel"
and "MAY" as "MAR".

Figured out via ASCII Chart the problem was a certain bit level was
being set wrong: let's say 3rd LSB (out of the 8). Figured out where
the problem was on the schematic (between encoder and decoder) but which
was LSB and MSB? (I'm not that good and pre-web access to get the chip specs.) So ended up making four reair solders: two on each end of the
trace and two since I wasn't sure which was LSB and MSB. Ta-dah! No
more misspellings!

KM> When the cable is loose, you'll lose some color (and have frex a
KM> pink screen).
Right.

Why it's usually pink, I dunno, tho I have seen blue and green,
just not as often. Probably depends on the layout of the cable
head, and which side is pulling loose, and how it sits on the
board, that sort of thing. Ie. which pins are being pulled out,
more likely on the side typically away from gravity.
Gravity always wins!

At least I don't have to go looking for dropped parts on the ceiling too!

KM> This isn't absolute but it's a pretty good guideline.
Good starting points which usually work.

Usually good enough, if it's hardware. If it's software, can be
any damn thing.

Though even software follows rules. I remember one script I was
creating, and since just self-trained a lot of pluck some code from
here, some other code from there.... Something wasn't working so put in
an echo statement to display the output at a certain level of the
script. That helped figure out the problem, so now getting the correct
result per the echo statement but the next step was always displaying
"0". Used to semi-work before! Ends up the following statement was
taking the output of the echo statement: echo finished its job properly
so outputted a '0' for 'good out'. Comment the test echo and now
worked!

> Doesn't eliminate video addressing statement as a possibility. I don't
> recall how to look up; guessing "all f's" in the range end-point would
> mean it's at the far end.
KM> Addressing shouldn't cause this kind of visual screw up; it would
KM> cause a lockup or no video at all due to the conflict. And if
KM> you're not a modern gamer, or rendering video, you're not gonna
KM> fill up that video RAM anyway.
Probably so: this is another of my Black Box areas and so stuff in, something happens, stuff out. On the -- guess could call it
'clarification' area know there are maximum resolutions so I would guess there is some sort of upper limit. Doesn't seem to be pertinent as the video card and monitor seem to always adjust to each other.

Modern OSs probe the vidcard and the monitor, which proceeds to
tell the OS what it is capable of, producing a range of options
that if wrong will do no worse than look goofy, but with neither
damage the hardware nor leave you with a blanked-out display that
needs a reinstall or CLI magic to fix.

...Trying to piece that together with a problem I have here: took an old computer out of service here as a Frontend (view recorded show via
MythTV) because of a problem with the video, which right now not
recalling exactly what -- there's a note on the case. "Everything"
pointed to the video card so swapped that out with something current.
Nope! Same problem! ...Just left the Raspberry Pi in it's place.

Modern OSs now understand this stuff, and have done so pretty
reliably for about 25 years. Part of the longstanding problem
with linux was that until about 10 years ago, or a bit longer for
some of the more advanced distros, you had to know your hardware parameters and sometimes input them manually. It seems to have
finally got this right. When you're doing it for free, as has
mostly been the case, hardware programming is not near as sexy as
cute apps and games, so it gets way less attention.

Yes, I vaguely recall having to initially configure the video settings.
Now pretty much does it itself, though that occassionally creates a
problem when using a 4K TV and so viewing from ten feet away! <g>
...Even the resolution selectrion GUI is teeny-tiny when standing right
in front of the TV! <g>

So if I remember correctly Mike's problem was after a while his monitor would go black and need a reboot to get things going again. I don't
recall if he stated a time but seemed he implied weeks or months. I
want to get in on the thread because I have a similar problem with one essentially headless system which tends to not talk to a monitor after
some time -- I'm thinking around 1.5 to 2 months.

Fedora used to do that, tho the problem seems to have gone away
as of v39. Except it would decide to not speak to the outside
world after about a week, unless regularly rebooted. Meanwhile
PCLOS had been running for several months, and WinXP for about a
year and a half, with no such issues.

Almost seems like 'something runs out of room'. I have some computers
around here rebooting themselves on a weekly basis because of that kind
of issue. Not necessarily an OS issue -- may have been years ago, but
now more like a problem with some utility causing the lockup/overrun. (Thinking of a couple years back when had three RPi4's running a previous version of Motion and other RPi4 running same OS but Motion not
installed. The Motion ones needed frequent rebooting, the others not
running Motion would run for months. [Important note: Motion has been corrected since then!])

As we skeap ?? speak Fedora is sitting there upgrading to v40.
It's been at it since 11pm last night, tho a lot of the time was
5GB of downloads on a 3Mbps connection. Had to do a full update
first, then run the upgrade (fortunately the commands are still
handy in the console buffer... it started life as v32). It is now
running the final step and will be done in about an hour. You can
see why I find rolling a lot less trouble.

Better hope that connection maintains!! ...The long time to do an update/upgrade was one reason I only did one machine at a time: besides multiple connections slowing down the limited bandwidth if something
went wrong like an extended power failure I have only one computer to
recover.

And I get my first look at KDE Plasma v6. Given the newness
thereof (just released) for the next while there will be a lot of
updates, possibly to the point of a Whole New Monkey. But they
are usually pretty good about getting the major bugs out (not
least because unlike say Xfce, which is one guy and change, it's
a team of a hundred-plus folks.)

More people seems to be better as a couple might be experts in video,
others, audio, etc. Plus if only if one or two doing the whole job and something happens.....

http://www.wholenewmonkey.com/

Huh: under maintainence -- guess the upgrade and update data is stil
lrolling in!

¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... A group of flamingos is called a stand.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

> Oops! That's what I get for not being overly familiar with the stuff!
KM> I have become painfully familiar. <g>

That was also part of your job. Same as back in the olden days when I
was selling computers and they were still more of a novelty I resisted
using some utilities even though they were clearly better than what
Windows was providing: I was sort of providing customer support and
figured I'd better be intimately familiar with what the customer's
system had.

Tho my clients were all Windows...

> KM> Whine whine whine!
> It's 1700 somewhere! <bseg>
KM> <tips tricorn hat>

(Hmm: did Ky slide over to the year or did he miss the 24-hour version
of 5 o'clock somewhere?)

You need to be specific, otherwise I take the one that's a better
fashion statement.

KM> Generally if I'm using linux I'm using linux, and I expect things
KM> to behave like linux.

The problem I'm finding is some utilties only use Windows, and some

And there's my problem...

KM> But yeah, there are some things linux still doesn't do well, and
KM> may never do well, or even at all.

Agree, and while could be an inconvenience not necessarily a bad thing.
Using my thumbdrive repair example, Linux might "never" be able to repair/recover and now (time - 2024) it may be cheaper and easier to
toss a failed thumbdrive. OTOH if something really important on it
that needs to be recovered then worth the cost to send to a recovery
service.

And given it's a flash drive... unless it's purely a filesystem error,
AFAIK it's not recoverable.

KM> Frex, there is NO dedicated RTF editor for linux. I've looked.
KM> No, LibreOffice is not satisfactory (insert rant about clean
KM> formatting code vs printer-defined formatting and how the latter
KM> needs to be stripped out for publication, and also to avoid
KM> random screwups). So... XP in a VM, and my dedicated RTF editor
KM> to the rescue.

Haven't fiddle but thinking gedit and mousepad. ... I haven't used a dot-matrix printer in years so zero experience. Only thing I do with
plain text is fiddle with a little coding.

Nope. Gedit and the like are plaintext editors. In that realm I like
KATE (KDE Another Text Editor, I assume) and it can do all sorts of
programmer formatting.... but the one set of tags it DOES NOT DO is RTF.

I expect in the Early Daze this was rejection of a standard from
Microsoft (there was a lot of that -- if MSFT does it, we won't, or will
do the opposite, even if it's the only sane method) but now it's just...
not sexy programming, so no one does it.

KM> It's a nuisance, but... better than fighting with WINE. And XP
KM> will always speak to the whole network, which linux never will.
KM> (Does not like random other linux boxen, never mind random
KM> Windows. And cannot be trusted to write files without fragmenting
KM> them all over the place.)

Could be. (Not disagreeing with you, just insufficient mackground on my end.) I do agree it does seem odd Linux doesn't have a defregment
option -- hey: good reason to do a fresh install as that puts the files
back together! <g>

The claim is that linux doesn't fragment, because it just keeps moving
on out to the largest available single block (which of course wastes a
LOT of drive space).

I beg to differ. The red blocks are a single file, written by linux; the
blue blocks are files of similar (rather large) size written by Windows.

http://doomgold.com/images/linux/fragmented.jpg

The file was damaged beyond use (fortunately not the only copy).

But yes, for ages the only way to defragment linux was the same way we
did it in the DOS 5 era and before -- copy everything off, format the
drive, copy everything back. And that's still really the typical method,
if you're using a HDD. (Not so relevant with SSDs.) Or use huge drives
and expect to waste about a third of the space.

Well, there's this, but apparently it's not very good and there's debate
over whether the drive should be mounted or not. And by default it only
does one file.

https://www.howtoforge.com/tutorial/linux-filesystem-defrag/

Read also the comments.

btrfs filesystem reportedly has its own stability issues....

I'll admit to having problems 'seeing' other computers around here but
the correction usually was to fix the ssh connection and then Remmina or TigerVNC would run correctly. Both of those do have some sort of a

I think it must be a more general linux thing. Someone I know sent me a tutorial on how to get Mint to behave on a network and it was a lot of
bother, so I haven't tried it yet. Need to look on old system for the email....
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> Assuming you can get it to install and run in the VM. You should
KM> hear my VM War Stories....
I don't play all that much but have had a few instances where it just doesn't play nice. I've also been using Raspberry Pi's for a disposable machine. No, not going to toss the Pi itself, but can overwrite the SD
card and so remove all traces of the failed experiment.

If I'm just experimenting, I use real hardware and a stack of
small hard drives... much more reliable results. One PC, fifty
OSs. <g>

Probably better to use real hardware than 'pretend' hardware but again a
matter of "all depends". I don't test/try stuff all that often and the
expense of getting hot swap enclosures would probably really make a 'no'.
Plus I'm running out of room: another computer plus storage of those
hard drives!

But when I need to run something that the host OS doesn't like...
that's why I do VMs.

The two main use cases being:

Host: any linux
VM:
-- WinXP, for a handful of WinApps I can't replace and for
reliable networking

Host: Win8/10
VMs:
-- WinXP so my eyes don't bleed when I need to work there all day

And the later experiments, in search of this interface between
usability and compatibility that is so absent in Certain Later Windholes...

Host: Win11
Still trying to come up with a combo that works on the Win11
netbook. XP runs so slowly... 15 minutes later I was still
waiting for the desktop to finish loading.

Win9x/NT4 cannot use the Guest Additions, so can only run at
800x600 (which is awfully small on a 1920x1080 monitor) and does
not see shared folders. That's not useful.

So far Win2k installer keeps looping back on installing hardware,
so that didn't work.

Host: XP64, which limits it to VirtualBox v4.3.12
The main object is to get a few things working that need a newer
OS. VMs:
-- WinXP32 (for the DOS app I can't live without) -- works fine,
tho I had to install from scratch; it would not import my
existing "appliance" even as a naked vmdk (you can un7zip an .OVA
file to extract this). -- Linux Mint -- runs fine but only at
1024x768. -- PCLinuxOS -- a couple of last-year's-models will
install, but throw up all over the Guest Additions, reducing the
screen to 800x600!! May try setting it up as Live only, and save
the machine state instead of installing it. The installed version
does not like the virtual video driver, judging by where it likes
to stall. -- Win8.1 -- runs fine, tho it makes my eyes bleed.
However, now I have a working FTP client without having to schlep
files to a Win1x box first. Chrome itself won't install, but
Supermium (current Chrome for old Windows) runs fine. (I'd
forgotten how braindead IE was back then...) -- Win10 won't
install. -- ReactOS just for giggles...runs okay but only at
small screen size, and I think there have been a lot of usability regressions since the previous version (which was actually pretty
good).

You do a heck of a lot more experimenting than I do!

Did you see the crazy thing (I think it was) MJD did, with VMs
inside of VMs until it went all the way from Newest Windows to
Oldest Windows??

No but sounds like an interesting project! Wonder if considered going
back to MS-DOS?! ..Wonder how much storage it takes? Presume on a
NVMe just for a reasonable speed to load the most current version, then
the next from that, and the next from that one....

> Yes, the 'rolling upgrade process' can be a litte detail-filled!
KM> "Rolling" is supposed to refer to the software, not the server
KM> cabinet!
Only when forget to set the brakes on the cabinet's wheels!

OUT OF THE WAY, BERTHA IS ROLLING!!!!!

When I was growing up a strip mall was created on a swampy area. They
filled in with the the usual dirt (guess from other construction sites)
but also trees. ...You know what happens when trees rot or even the
branches collapse? Yup! So the store floors are supported by what's underneath, which sometimes wasn't there. ...Eventually if one didn't
hold on to shopping cart it would roll away!

KM> I've come to greatly prefer a rolling distro for an everyday
KM> desktop. There are NO UPGRADES and therefore NO REINSTALLS.
KM> Someone took the oldest extant PCLOS from 2010 to 2021 (I think
KM> it was) using only the rolling update process via Synaptic, and
KM> only twice did he have to stop and twiddle something at the
KM> command line. But normally it's just click-the-usuals, go away
KM> for a while, and it's done.
KM> The other advantage is that problems get fixed NOW, not when the
KM> next major version rolls around.
That option has advantages and disadvantages. I do like the 'fixed
now', assuming it doesn't break something else. (Of course the latter
is a possibility any time.) Ubuntu does have LivePatch, which is
probably your fix-it-now.

Regressions have been pretty rare. They all do some degree of
updating between, but the real point is there is no such thing as
a release version, because it just keeps trundling to the future
without the need of silly version numbers.

"Beeg number good! Impressive!" Personally I don't use the version
numbers other than to keep track: right now I need '31' for MythTV and
it pretty much doesn't matter which version of the OS it runs on. As
for the OS in general, I'm not a latest-and-greatest type of guy as for
the OS in general but do sort of like to keep the thing up-to-date for security and other patches.

Around here I'd prefer manual version upgrades, meaning to go from
version 22 to version 23, not the more-minor updates. Have had old computers no longer work properly with upgrades: IMO not a fault of the
OS, though one could say it didn't check for compatibility.

Yeah, that is a problem. And yes, it should do a compatibility
check, but that's not sexy programming, you should just do a
reinstall and stop bothering us.

I've noticed Mozilla appears o do a full re-install instead of an update
and I think LibreOffice does too.

KM> The downside is you have to do your updates regularly, because it
KM> can get out of sync if you let things go too long. (Tho Synaptic
KM> at least is pretty good at squaring things up.) But it's fast
KM> enough, given it's usually small, to do 'em often.
Agree: though with the limited experience I have it seems Ubuntu does a reasonable job: it seems one can take a 'basic' version (say 22.04.03), which does the installaion and original versions, then at the end of the installation ask for the updates and it will d/l umpteen files, od the magic, and now 22.04.39. (Making up the numbers.)

Well, my Fedora setup started at v32, and is now v40, with KDE
upgraded from v5 to v6, tho that's still early enough that there
are minor oddities. OTOH it did fix a longstanding visual bug in
my preferred theme.

That's good!

KM> Speaking therewhich, I need to do the twice-yearly version
KM> upgrade on the Fedora system, which I keep forgetting because
KM> it's an overnight job (sometimes two nights) and takes some
KM> babysitting before it gets going. Grrr.
And if you're like me the babysitting is the major part of the delay!
Good reason to have two monitors so can have one on the upgrading system
and the other for while-I'm-sitting-here-do-e-mail, etc.

Fortunately it's not a system I'm trying to use while it's doing
this (in fact I hardly use it at all) but I still had to wait for
the repository list (which was about 100mb and dozens of files)
then two hours for the download, then half an hour for the
update, then repeat this for the full upgrade, except the
download took the rest of the night and the upgrade itself about
an hour.

That's one nice thing about living in a small city with fiber-optic
Internet access! Of course it still depends on the hardware on my side:
I've had the other side of the desk tied up for hours/overnight as the installation proceeds.

I have no idea how one upgrades Debian, if it can even be done...

I've done upgrades on Ubuntu (so by extension Debian) but on smaller,
more basic machines. (By 'more basic' I'm meaning not too much has been added/modified from the original.) OTOH I do tend to go for a
from-scratch upgrade install because something major has been changed in
the machine, or a setting is wrong (old example: root set too small) and
while something like the root partition can be expanded (I've done it)
from my past experiences (Linux and Windows) it's just overall better to
start with a clean slate.



¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... "Please insert CD-ROM into Drive A."
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

> Oops! That's what I get for not being overly familiar with the stuff!
KM> I have become painfully familiar. <g>
That was also part of your job. Same as back in the olden days when I
was selling computers and they were still more of a novelty I resisted
using some utilities even though they were clearly better than what
Windows was providing: I was sort of providing customer support and
figured I'd better be intimately familiar with what the customer's
system had.

Tho my clients were all Windows...

Though your magic was plugging in the USB thumbdrive with your favourite
Linux flavour and fixing the Windows problem that way! <g>

> KM> Whine whine whine!
> It's 1700 somewhere! <bseg>
KM> <tips tricorn hat>
(Hmm: did Ky slide over to the year or did he miss the 24-hour version
of 5 o'clock somewhere?)

You need to be specific, otherwise I take the one that's a better
fashion statement.

So now tip your tricorn at 5 o'clock!

KM> Generally if I'm using linux I'm using linux, and I expect things
KM> to behave like linux.
The problem I'm finding is some utilties only use Windows, and some

And there's my problem...

I have a bit of a suspicion there's some sort of under-the-table
agreements for the 'Windows exclusively', otherwise why would companies manufacture for only part of the market?

KM> But yeah, there are some things linux still doesn't do well, and
KM> may never do well, or even at all.
Agree, and while could be an inconvenience not necessarily a bad thing. Using my thumbdrive repair example, Linux might "never" be able to repair/recover and now (time - 2024) it may be cheaper and easier to
toss a failed thumbdrive. OTOH if something really important on it
that needs to be recovered then worth the cost to send to a recovery service.

And given it's a flash drive... unless it's purely a filesystem
error, AFAIK it's not recoverable.

Highly limited personal experience here so probably yes. Most of the
problem cases are off-brands, though I had a quirk with all of the black
and yellow ADATA thumbdrives eventually failed yet all of the black and
blues ones are fine. Same seller, same capacity, purchased about the
same time. <shrub>

KM> Frex, there is NO dedicated RTF editor for linux. I've looked.
KM> No, LibreOffice is not satisfactory (insert rant about clean
KM> formatting code vs printer-defined formatting and how the latter
KM> needs to be stripped out for publication, and also to avoid
KM> random screwups). So... XP in a VM, and my dedicated RTF editor
KM> to the rescue.
Haven't fiddle but thinking gedit and mousepad. ... I haven't used a dot-matrix printer in years so zero experience. Only thing I do with
plain text is fiddle with a little coding.

Nope. Gedit and the like are plaintext editors. In that realm I
like KATE (KDE Another Text Editor, I assume) and it can do all
sorts of programmer formatting.... but the one set of tags it
DOES NOT DO is RTF.

I scanned though some articles and didn't find a reason; was
half-expecting when I saw 'Microsoft' and 'proprietary' to also see a reference to a copyright or whatever legal "this is mine alone"
document. Half-figured something along those lines would be a suitable reason.

I expect in the Early Daze this was rejection of a standard from
Microsoft (there was a lot of that -- if MSFT does it, we won't,
or will do the opposite, even if it's the only sane method) but
now it's just... not sexy programming, so no one does it.

Usually pretty much if there is a need then we will do it, otherwise not
worth it.

KM> It's a nuisance, but... better than fighting with WINE. And XP
KM> will always speak to the whole network, which linux never will.
KM> (Does not like random other linux boxen, never mind random
KM> Windows. And cannot be trusted to write files without fragmenting
KM> them all over the place.)
Could be. (Not disagreeing with you, just insufficient mackground on my end.) I do agree it does seem odd Linux doesn't have a defregment
option -- hey: good reason to do a fresh install as that puts the files
back together! <g>

The claim is that linux doesn't fragment, because it just keeps
moving on out to the largest available single block (which of
course wastes a LOT of drive space).
I beg to differ. The red blocks are a single file, written by
linux; the blue blocks are files of similar (rather large) size
written by Windows.
http://doomgold.com/images/linux/fragmented.jpg

To my thinking the only way to keep a file from fragmenting is to have a
rule it needs to be laid down in one contiguous section -- and that
might a bit of added difficulty with those bad segments.

The fragmentation discussion almost points to a very good reason to
start with a fresh system rather than upgrade. I don't recall reading
about it specifically as they always seem to say 'files' which sort of
gets thought of as the data -- stuff added by humans. Seems to me the Operating System files could just as easily get fragmented and
eventually have problems.

The file was damaged beyond use (fortunately not the only copy).

Backups are good! Multiple backups are better! Multiple backups in
different formats are even better!

But yes, for ages the only way to defragment linux was the same
way we did it in the DOS 5 era and before -- copy everything off,
format the drive, copy everything back. And that's still really
the typical method, if you're using a HDD. (Not so relevant with
SSDs.) Or use huge drives and expect to waste about a third of
the space.

Which is essentially creating a new system. ...Over the years I've sort
of learned to buy/use a lot larger computer hardware than I think I'll
need because something will come along to use all of what I have now.
First XT had a 20 MB hard drive -- I'll never run out of space! Nine
months later space is running low, the sister XT becomes available with
a 40 MB hard drive (and EGA card -- woo-hoo!! [Original was CGA.]
Network the two, so 60 MB available. I'LL NEVER RUN OUT OF SPACE!!!

Well, there's this, but apparently it's not very good and there's
debate over whether the drive should be mounted or not. And by
default it only does one file. https://www.howtoforge.com/tutorial/linux-filesystem-defrag/
Read also the comments.
btrfs filesystem reportedly has its own stability issues....

I'm not so sure there is any one great (or even good) file defragger. It
would seem each formatting style has it's own rules which would need to
be specifically addressed by the defragger. Would also seem each hard
drive manufacturer and version/family would also need to be address as
well -- why would Western Digital create Black series for regular use
and Red for a NAS? Data is data, but one group has faster writes while
another faster reads. ...Just seems to be too many variables.

And I'll bring up another tangent to file placement: I've noticed after
a reboot if I ask for a GUI list of files in a subdirectory the first
time it takes a while -- the more files the longer I'm waiting. But
generally any time after the first inquiry it is an instantaneous
response. Seems there's something working to counteract any slow
response due to fragmentation.

I'll admit to having problems 'seeing' other computers around here but
the correction usually was to fix the ssh connection and then Remmina or TigerVNC would run correctly. Both of those do have some sort of a

I think it must be a more general linux thing. Someone I know
sent me a tutorial on how to get Mint to behave on a network and
it was a lot of bother, so I haven't tried it yet. Need to look
on old system for the email....

You need a larger hard drive to conveniently store the trivia! <g>

Some of the issues on my side are caused by the way I do things: I'll
use the same physical Raspberry Pi but with a different (base ^*) SD card
the certificates won't match and the easiest and quickest way, plus the
way I usually find the problem, is to SSH into the remote unit:
"certificates don't match" error pops up, contains the fix command line
-- copy and paste, done!

"Base" card up there because I can't think of a better word. I can make
a duplicate of the card and swap it in -- everything will work properly.
Format the card or put a different card in, even with the same OS, and
the unit will now be 'different' to the network: someone else is impersonanting that IP! The certificate doesn't match!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Ignorant?! Ha! I don't even know the meaning of the word!
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

KM> Tho my clients were all Windows...

Though your magic was plugging in the USB thumbdrive with your favourite Linux flavour and fixing the Windows problem that way! <g>

That's my magic for recovering data off drives with busted passwords...

> > KM> Whine whine whine!
> > It's 1700 somewhere! <bseg>
> KM> <tips tricorn hat>
> (Hmm: did Ky slide over to the year or did he miss the 24-hour version
> of 5 o'clock somewhere?)
KM> You need to be specific, otherwise I take the one that's a better
KM> fashion statement.
So now tip your tricorn at 5 o'clock!

I shall eagerly await this turn of the clock!

> KM> Generally if I'm using linux I'm using linux, and I expect things
> KM> to behave like linux.
> The problem I'm finding is some utilties only use Windows, and some
KM> And there's my problem...

I have a bit of a suspicion there's some sort of under-the-table
agreements for the 'Windows exclusively', otherwise why would companies manufacture for only part of the market?

Linux desktops are something like 4% of the desktop market, and most of
that is adjunct to the used-hardware market, and the stats probably
includes Android tablets (Android is essentially linux). No one in their
right financial mind would spend resources on a linux desktop version
that requires more than setting a target flag in the compiler. This is
much easier with modern compilers targeting modern OSs, hence a lot of
major apps (eg. LibreOffice, Softmaker Office, the KDE stable of desktop
apps) are often available across platforms.

However, hardware-level utilities are generally coded in Assembly
Language, which does not port easily to another OS, precisely because it
does its own hardware access. Same reason drivers aren't portable as such.

KM> And given it's a flash drive... unless it's purely a filesystem
KM> error, AFAIK it's not recoverable.

Highly limited personal experience here so probably yes. Most of the

It becomes like recovering data off a regular memory chip: it just ain't there.

problem cases are off-brands, though I had a quirk with all of the black
and yellow ADATA thumbdrives eventually failed yet all of the black and
blues ones are fine. Same seller, same capacity, purchased about the
same time. <shrub>

I remember that. Likely different supplier for the chips and/or logic
boards. Plus ADATA had a longstanding repute as crap, so I was not
surprised. <g>

I'd only buy Samsung, Sandisk/WD, and PNY if it's a scratch drive.

Otherwise, I wouldn't pay money for a flash drive of any description.

> KM> Frex, there is NO dedicated RTF editor for linux. I've looked.
> KM> No, LibreOffice is not satisfactory (insert rant about clean
> KM> formatting code vs printer-defined formatting and how the latter
> KM> needs to be stripped out for publication, and also to avoid
> KM> random screwups). So... XP in a VM, and my dedicated RTF editor
> KM> to the rescue.
> Haven't fiddle but thinking gedit and mousepad. ... I haven't used a
> dot-matrix printer in years so zero experience. Only thing I do with
> plain text is fiddle with a little coding.
KM> Nope. Gedit and the like are plaintext editors. In that realm I
KM> like KATE (KDE Another Text Editor, I assume) and it can do all
KM> sorts of programmer formatting.... but the one set of tags it
KM> DOES NOT DO is RTF.

I scanned though some articles and didn't find a reason; was
half-expecting when I saw 'Microsoft' and 'proprietary' to also see a reference to a copyright or whatever legal "this is mine alone"
document. Half-figured something along those lines would be a suitable reason.

As I said below....

KM> I expect in the Early Daze this was rejection of a standard from
KM> Microsoft (there was a lot of that -- if MSFT does it, we won't,
KM> or will do the opposite, even if it's the only sane method) but
KM> now it's just... not sexy programming, so no one does it.

Usually pretty much if there is a need then we will do it, otherwise not worth it.

Sometimes we won't do it even if there's a need, to thumb our noses at
the Man. Or why for a long time linux had the crap audio format off OGG,
but Would Not Do MP3, because it was Encumbered by Patents. In the early
days, forget if it was GIF or JPG but one was still under patent so we
won't touch it, you will use PNG and like it.

There's a point where Principle becomes Cutting Off Your Own Nose.
Debian did a lot of that; until recently you couldn't run "non-free"
drivers, so if you had X Y or Z hardware that hadn't given their driver
source code to linux for free, you were SOL. I remember when that nixed
all of NVidia as vidcards, despite being the most common in linux's
largest target market (PC builder enthusiasts).

> KM> It's a nuisance, but... better than fighting with WINE. And XP
> KM> will always speak to the whole network, which linux never will.
> KM> (Does not like random other linux boxen, never mind random
> KM> Windows. And cannot be trusted to write files without fragmenting
> KM> them all over the place.)
> Could be. (Not disagreeing with you, just insufficient mackground on my
> end.) I do agree it does seem odd Linux doesn't have a defregment
> option -- hey: good reason to do a fresh install as that puts the files
> back together! <g>
KM> The claim is that linux doesn't fragment, because it just keeps
KM> moving on out to the largest available single block (which of
KM> course wastes a LOT of drive space).
KM> I beg to differ. The red blocks are a single file, written by
KM> linux; the blue blocks are files of similar (rather large) size
KM> written by Windows.
KM> http://doomgold.com/images/linux/fragmented.jpg

To my thinking the only way to keep a file from fragmenting is to have a
rule it needs to be laid down in one contiguous section -- and that
might a bit of added difficulty with those bad segments.

That is in fact how DOS does it, if there's room. That's supposedly how
linux does it. Except the evidence is that linux does anything but. One suspects if there is ONE fragment, it just blows the rule away entirely
and writes segments any damn place.

This is why I do not trust linux to write archival files, and now always either have Windows-via-the-network or a sacrificial external drive
(that I don't mind periodically reformatting) act as the middleman.

And my observation is that the linux filesystem is much more fragile in
the event of an insult such as an ungraceful shutdown (and in the event,
fsck will often just delete affected files, sucks to be you.)

I have also had linux delete whole swaths of the sacrificial drive
because ONE file failed to copy TO that drive.

Cloud enterprise gets away with it because they have honkin' big tape
backups, not because the linux filesystems are so reliable.

The fragmentation discussion almost points to a very good reason to
start with a fresh system rather than upgrade. I don't recall reading
about it specifically as they always seem to say 'files' which sort of
gets thought of as the data -- stuff added by humans. Seems to me the Operating System files could just as easily get fragmented and
eventually have problems.

Really, it means only use SSDs and assume that periodically the system
will still need a copy-and-back defragging. And never, ever write user
data (especially cache or swap) to the same partition as the OS files.

KM> The file was damaged beyond use (fortunately not the only copy).

Backups are good! Multiple backups are better! Multiple backups in different formats are even better!

Ya think??!

KM> But yes, for ages the only way to defragment linux was the same
KM> way we did it in the DOS 5 era and before -- copy everything off,
KM> format the drive, copy everything back. And that's still really
KM> the typical method, if you're using a HDD. (Not so relevant with
KM> SSDs.) Or use huge drives and expect to waste about a third of
KM> the space.

Which is essentially creating a new system. ...Over the years I've sort
of learned to buy/use a lot larger computer hardware than I think I'll
need because something will come along to use all of what I have now.
First XT had a 20 MB hard drive -- I'll never run out of space! Nine
months later space is running low, the sister XT becomes available with
a 40 MB hard drive (and EGA card -- woo-hoo!! [Original was CGA.]
Network the two, so 60 MB available. I'LL NEVER RUN OUT OF SPACE!!!

Along come 12TB drives, and.....

KM> Well, there's this, but apparently it's not very good and there's
KM> debate over whether the drive should be mounted or not. And by
KM> default it only does one file.
KM> https://www.howtoforge.com/tutorial/linux-filesystem-defrag/
KM> Read also the comments.
KM> btrfs filesystem reportedly has its own stability issues....

I'm not so sure there is any one great (or even good) file defragger. It would seem each formatting style has it's own rules which would need to
be specifically addressed by the defragger. Would also seem each hard

That is true -- it has to know the filesystem and be thoroughly aware of
any inbuilt defects. Same as you can't use a FAT file utility on an NTFS
file system; it doesn't know how to handle it and will just mark it as Bad!

Had a client manage to do this. Used a FAT16 disk maintenance utility on
a FAT32 drive... ooops. Fortunately, I still had their old drive (just replaced) as backup...

drive manufacturer and version/family would also need to be address as
well -- why would Western Digital create Black series for regular use
and Red for a NAS? Data is data, but one group has faster writes while another faster reads. ...Just seems to be too many variables.

NAS is expected to be more archival, doesn't need the speed, but needs
max data reliability (which may mean more platters per TB). Black is a
much faster drive, intended for desktop and gaming use. I have long
thought Blue and Green were QC grades rather than different drives.

And some of it is just marketing.

And I'll bring up another tangent to file placement: I've noticed after
a reboot if I ask for a GUI list of files in a subdirectory the first
time it takes a while -- the more files the longer I'm waiting. But generally any time after the first inquiry it is an instantaneous
response. Seems there's something working to counteract any slow
response due to fragmentation.

Has to rebuild the file cache, and apparently in Ubuntu that cache is
not persistent.

Also if you have spinning rust drives set to sleep after X-long, it
takes about 30 seconds for the drive to wake back up.

> I'll admit to having problems 'seeing' other computers around here but
> the correction usually was to fix the ssh connection and then Remmina or
> TigerVNC would run correctly. Both of those do have some sort of a
KM> I think it must be a more general linux thing. Someone I know
KM> sent me a tutorial on how to get Mint to behave on a network and
KM> it was a lot of bother, so I haven't tried it yet. Need to look
KM> on old system for the email....

You need a larger hard drive to conveniently store the trivia! <g>

Why is the trivia the size of the universe??!

Some of the issues on my side are caused by the way I do things: I'll
use the same physical Raspberry Pi but with a different (base ^*) SD card
the certificates won't match and the easiest and quickest way, plus the
way I usually find the problem, is to SSH into the remote unit: "certificates don't match" error pops up, contains the fix command line
-- copy and paste, done!

That's using a server function (do I know each piece of hardware by
serial number?) on a desktop, aka that's just dumb. But linux sometimes
does that. Once killed a Debian install by cloning it to a larger drive.

And the current Debian install (which hadn't been up in a year or so) is
going "servers? what servers??" meaning it's gonna be a full reinstall
instead of an upgrade. Good thing for it that it's just experimental and
not an everyday-use setup.

"Base" card up there because I can't think of a better word. I can make
a duplicate of the card and swap it in -- everything will work properly. Format the card or put a different card in, even with the same OS, and
the unit will now be 'different' to the network: someone else is impersonanting that IP! The certificate doesn't match!

Yeah, it's written what amounts to a serial number to the disk, and YOU CHANGED IT!!! Windows servers do the same thing and then some. You can't
just swap a drive, it's assumed you murdered the first one and replaced
it with a golem. On the Giant Server (now being gradually dysmangled for parts) I heard about it when I put RAM sticks back in a different order!!
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

KM> If I'm just experimenting, I use real hardware and a stack of
KM> small hard drives... much more reliable results. One PC, fifty
KM> OSs. <g>

Probably better to use real hardware than 'pretend' hardware but again a matter of "all depends". I don't test/try stuff all that often and the expense of getting hot swap enclosures would probably really make a 'no'. Plus I'm running out of room: another computer plus storage of those
hard drives!

My experience is that LiveCD, Virtual Machine, and Real Hardware are not equivalent, in behavior or performance. Many a Linux LiveCD can see the
other PCs on the network, but the same one INSTALLED cannot.

KM> But when I need to run something that the host OS doesn't like...
KM> that's why I do VMs.

You do a heck of a lot more experimenting than I do!

A whole lot. <g>

Still preferably on real hardware, but am hunting for a VM that will
work on Roadkill... my regular XP VM wouldn't even finish loading.

KM> Did you see the crazy thing (I think it was) MJD did, with VMs
KM> inside of VMs until it went all the way from Newest Windows to
KM> Oldest Windows??

No but sounds like an interesting project! Wonder if considered going
back to MS-DOS?! ..Wonder how much storage it takes? Presume on a
NVMe just for a reasonable speed to load the most current version, then
the next from that, and the next from that one....

LOL, you can do that, if you have enough RAM!

> > Yes, the 'rolling upgrade process' can be a litte detail-filled!
> KM> "Rolling" is supposed to refer to the software, not the server
> KM> cabinet!
> Only when forget to set the brakes on the cabinet's wheels!
KM> OUT OF THE WAY, BERTHA IS ROLLING!!!!!

When I was growing up a strip mall was created on a swampy area. They
filled in with the the usual dirt (guess from other construction sites)
but also trees. ...You know what happens when trees rot or even the
branches collapse? Yup! So the store floors are supported by what's underneath, which sometimes wasn't there. ...Eventually if one didn't
hold on to shopping cart it would roll away!

LOL, the old Costco here (they've moved) had that issue. The parking lot
was atop what used to be a dump. Flat when first paved, but a few
decades later it was up hill and down dale in every direction, tho with
the largest dent toward the middle. And I mean a serious slope, not just
a little dip!

> KM> I've come to greatly prefer a rolling distro for an everyday
> KM> desktop. There are NO UPGRADES and therefore NO REINSTALLS.
KM> Regressions have been pretty rare. They all do some degree of
KM> updating between, but the real point is there is no such thing as
KM> a release version, because it just keeps trundling to the future
KM> without the need of silly version numbers.

"Beeg number good! Impressive!" Personally I don't use the version

LOL, there is that. Then again, numbers like 0.29 are not impressive.

numbers other than to keep track: right now I need '31' for MythTV and
it pretty much doesn't matter which version of the OS it runs on. As
for the OS in general, I'm not a latest-and-greatest type of guy as for
the OS in general but do sort of like to keep the thing up-to-date for security and other patches.

Came across some idiot on Youtube "demonstrating how unsafe XP is
online" .... first thing he does is DISABLE THE FIREWALL, and naturally
it immediately collected every circulating network worm or virus. Uh,
stupid, do that with ANY OS and it'll have the same thing happen!!

And then he says, "I don't think the firewall is much good" ... <headdesk>

> Around here I'd prefer manual version upgrades, meaning to go from
> version 22 to version 23, not the more-minor updates. Have had old
> computers no longer work properly with upgrades: IMO not a fault of the
> OS, though one could say it didn't check for compatibility.
KM> Yeah, that is a problem. And yes, it should do a compatibility
KM> check, but that's not sexy programming, you should just do a
KM> reinstall and stop bothering us.

I've noticed Mozilla appears o do a full re-install instead of an update
and I think LibreOffice does too.

A lot of these monolithic programs do that. However, IIRC Ubuntu is now
all containerized (or at least I heard it was going to be), which means
you always replace the whole thing.

KM> I have no idea how one upgrades Debian, if it can even be done...

A: Presently, it can't, if your install is over ... about a year since
it was last updated?? anyway, it whines that I must authorize a
different server, but zero information on how that's to be done. Screw
it, in the same time I can just install a new one. Not like I want to
save anything but the wallpaper.

I've done upgrades on Ubuntu (so by extension Debian) but on smaller,
more basic machines. (By 'more basic' I'm meaning not too much has been added/modified from the original.) OTOH I do tend to go for a
from-scratch upgrade install because something major has been changed in
the machine, or a setting is wrong (old example: root set too small) and while something like the root partition can be expanded (I've done it)
from my past experiences (Linux and Windows) it's just overall better to start with a clean slate.

Generally, but there's another advantage of Rolling...

If I have to reinstall with every version upgrade, I won't use it. That simple.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

KM> Boo!

Boo who?!!

No crying until you spill the milk!

..Thought I figured soming out but now makes less sense. If a 'boo' is
a boyfriend/girfriend/lover ("he/she is my boo") and 'boo' is said by a
ghost or to scare someone, why is it good to be a 'boo'?

Um. I have no good answer to this. Maybe I'll think up a silly one instead.

That makes sense. I'm thinking it could also mean a faulty bit which is causing a bad instruction and so garbage output. That one based on
years ago when I built a Heathkit TV and the overlay display (time,
channel, etc.) was sometimes scrambled. A little analysis and only
certain letters: let's say "C" so "Channel" might display as "Thannel"
and "MAY" as "MAR".

That's... interesting!!

Figured out via ASCII Chart the problem was a certain bit level was
being set wrong: let's say 3rd LSB (out of the 8). Figured out where
the problem was on the schematic (between encoder and decoder) but which
was LSB and MSB? (I'm not that good and pre-web access to get the chip specs.) So ended up making four reair solders: two on each end of the
trace and two since I wasn't sure which was LSB and MSB. Ta-dah! No
more misspellings!

Oh! Clever solution. But when you've got a schematic to work from....

KM> Gravity always wins!

At least I don't have to go looking for dropped parts on the ceiling too!

This is a very good point.

Well, unless you're on the space station.

> KM> This isn't absolute but it's a pretty good guideline.
> Good starting points which usually work.
KM> Usually good enough, if it's hardware. If it's software, can be
KM> any damn thing.

Though even software follows rules. I remember one script I was

It's supposed to. When there are not rules it makes up its own.

Someone discovered that the value of Pi used in DOOM was wrong (but it
was a lookup table, so the values were fixed). So he tried lots of other values. Correct Pi was close enough. Others went from strange to nonfunctional.

https://media.ccc.de/v/mch2022-236-non-euclidean-doom-what-happens-to-a-game-when-pi-is-not-3-14159-

creating, and since just self-trained a lot of pluck some code from
here, some other code from there.... Something wasn't working so put in
an echo statement to display the output at a certain level of the
script. That helped figure out the problem, so now getting the correct result per the echo statement but the next step was always displaying
"0". Used to semi-work before! Ends up the following statement was
taking the output of the echo statement: echo finished its job properly
so outputted a '0' for 'good out'. Comment the test echo and now
worked!

Ah yes, the old step-through method... usually a Good Idea when you can.


> KM> Modern OSs now understand this stuff, and have done so pretty

KM> reliably for about 25 years. Part of the longstanding problem
KM> with linux was that until about 10 years ago, or a bit longer for
KM> some of the more advanced distros, you had to know your hardware
KM> parameters and sometimes input them manually. It seems to have
KM> finally got this right. When you're doing it for free, as has
KM> mostly been the case, hardware programming is not near as sexy as
KM> cute apps and games, so it gets way less attention.

Yes, I vaguely recall having to initially configure the video settings.

I remember it vividly, having installed RedHat 6 which at the time you
had to set up X entirely yourself. Make a good guess and get a screen.
Make a wrong guess and start over.

Now pretty much does it itself, though that occassionally creates a
problem when using a 4K TV and so viewing from ten feet away! <g>
..Even the resolution selectrion GUI is teeny-tiny when standing right
in front of the TV! <g>

LOL. Some of my failed Virtual Machines have been the size of a postage stamp....

Did get Win2K VM installed on Roadkill tho, so now I have a restful grey workspace when I need it. Win11 lacks this.

> So if I remember correctly Mike's problem was after a while his monitor
> would go black and need a reboot to get things going again. I don't
> recall if he stated a time but seemed he implied weeks or months. I
> want to get in on the thread because I have a similar problem with one
> essentially headless system which tends to not talk to a monitor after
> some time -- I'm thinking around 1.5 to 2 months.
KM> Fedora used to do that, tho the problem seems to have gone away
KM> as of v39. Except it would decide to not speak to the outside
KM> world after about a week, unless regularly rebooted. Meanwhile
KM> PCLOS had been running for several months, and WinXP for about a
KM> year and a half, with no such issues.

Almost seems like 'something runs out of room'. I have some computers

Yeah, that is usually why a system goes goofy over time -- some cache or
data stack gets full or corrupted or wraps around, but the effect is the
same.

around here rebooting themselves on a weekly basis because of that kind
of issue. Not necessarily an OS issue -- may have been years ago, but
now more like a problem with some utility causing the lockup/overrun.

That was often the issue with Windows. Not a durn thing wrong with
Windows; it was the crappy "fix something" utility that broke it.

(Thinking of a couple years back when had three RPi4's running a previous version of Motion and other RPi4 running same OS but Motion not
installed. The Motion ones needed frequent rebooting, the others not
running Motion would run for months. [Important note: Motion has been corrected since then!])

That's good! Cuz frequent rebooting... not my idea of Quality. Been
ruined by software that runs for months on end. RoughDraft and WinAmp
have been up since... probably last October, when the system was last rebooted. SeaMonkey needs to be restarted every few weeks, tho, or it
gets sluggish.

LibreOffice has been up since March, that being when I started work on
the current going-very-slow paid edit (it's not on a deadline, and it
makes my brain hurt).

KM> As we skeap ?? speak Fedora is sitting there upgrading to v40.
KM> It's been at it since 11pm last night, tho a lot of the time was
KM> 5GB of downloads on a 3Mbps connection. Had to do a full update
KM> first, then run the upgrade (fortunately the commands are still
KM> handy in the console buffer... it started life as v32). It is now
KM> running the final step and will be done in about an hour. You can
KM> see why I find rolling a lot less trouble.

Better hope that connection maintains!! ...The long time to do an

Fortunately it caches downloads so you don't have to start over.

That's one thing Windows Update gets wrong. Anything interrupted starts
over.

update/upgrade was one reason I only did one machine at a time: besides multiple connections slowing down the limited bandwidth if something
went wrong like an extended power failure I have only one computer to recover.

Yeah, good policy.

KM> And I get my first look at KDE Plasma v6. Given the newness
KM> thereof (just released) for the next while there will be a lot of
KM> updates, possibly to the point of a Whole New Monkey. But they
KM> are usually pretty good about getting the major bugs out (not
KM> least because unlike say Xfce, which is one guy and change, it's
KM> a team of a hundred-plus folks.)

More people seems to be better as a couple might be experts in video,
others, audio, etc. Plus if only if one or two doing the whole job and something happens.....

KDE has about a hundred people involved.

KM> http://www.wholenewmonkey.com/

Huh: under maintainence -- guess the upgrade and update data is stil
lrolling in!

Better not be under maintenance, it hasn't been changed in years!

(Yes, I got this domain solely for this joke. Cuz it's always funny.)

.. A group of flamingos is called a stand.

I thought it was called a lawn ornament!
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> Tho my clients were all Windows...
Though your magic was plugging in the USB thumbdrive with your favourite Linux flavour and fixing the Windows problem that way! <g>

That's my magic for recovering data off drives with busted
passwords...

Apparently he encryptio isn't where I expected.

> > KM> Whine whine whine!
> > It's 1700 somewhere! <bseg>
> KM> <tips tricorn hat>
> (Hmm: did Ky slide over to the year or did he miss the 24-hour version
> of 5 o'clock somewhere?)
KM> You need to be specific, otherwise I take the one that's a better
KM> fashion statement.
So now tip your tricorn at 5 o'clock!

I shall eagerly await this turn of the clock!

Bad news: the batteries in the analog clock died!

> KM> Generally if I'm using linux I'm using linux, and I expect things
> KM> to behave like linux.
> The problem I'm finding is some utilties only use Windows, and some
KM> And there's my problem...
I have a bit of a suspicion there's some sort of under-the-table
agreements for the 'Windows exclusively', otherwise why would companies manufacture for only part of the market?

Linux desktops are something like 4% of the desktop market, and
most of that is adjunct to the used-hardware market, and the
stats probably includes Android tablets (Android is essentially
linux).

I would guess part of the raeson for the only 4% is Linux wasn't really marketed until recently, and if one can save money by using the old
equipment then let's use the old equipment.

No one in their right financial mind would spend
resources on a linux desktop version that requires more than
setting a target flag in the compiler. This is much easier with
modern compilers targeting modern OSs, hence a lot of major apps
(eg. LibreOffice, Softmaker Office, the KDE stable of desktop
apps) are often available across platforms.

Makes sense.

However, hardware-level utilities are generally coded in Assembly Language, which does not port easily to another OS, precisely
because it does its own hardware access. Same reason drivers
aren't portable as such.

So might have several 'barriers'. First is the compiled for the
original OS. The human creating the code is trained and familiar with
that OS, so to create code for a second OS needs to learn that also.
...A lot more details in there, of course.

KM> And given it's a flash drive... unless it's purely a filesystem
KM> error, AFAIK it's not recoverable.
Highly limited personal experience here so probably yes. Most of the

It becomes like recovering data off a regular memory chip: it
just ain't there.

Solid state drives (of whatever format) have numerous good points but
also some major bad points. Backups are even better now!

problem cases are off-brands, though I had a quirk with all of the black
and yellow ADATA thumbdrives eventually failed yet all of the black and blues ones are fine. Same seller, same capacity, purchased about the
same time. <shrug>

I remember that. Likely different supplier for the chips and/or
logic boards. Plus ADATA had a longstanding repute as crap, so I
was not surprised. <g>

The good news was I didn't loose anything other than time.

Otherwise, I wouldn't pay money for a flash drive of any
description.

I've mainly used them for SneakerNet and Live Boot.

> KM> Frex, there is NO dedicated RTF editor for linux. I've looked.
> KM> No, LibreOffice is not satisfactory (insert rant about clean
> KM> formatting code vs printer-defined formatting and how the latter
> KM> needs to be stripped out for publication, and also to avoid
> KM> random screwups). So... XP in a VM, and my dedicated RTF editor
> KM> to the rescue.
> Haven't fiddle but thinking gedit and mousepad. ... I haven't used a
> dot-matrix printer in years so zero experience. Only thing I do with
> plain text is fiddle with a little coding.
KM> Nope. Gedit and the like are plaintext editors. In that realm I
KM> like KATE (KDE Another Text Editor, I assume) and it can do all
KM> sorts of programmer formatting.... but the one set of tags it
KM> DOES NOT DO is RTF.
I scanned though some articles and didn't find a reason; was
half-expecting when I saw 'Microsoft' and 'proprietary' to also see a reference to a copyright or whatever legal "this is mine alone"
document. Half-figured something along those lines would be a suitable reason.

As I said below....

KM> I expect in the Early Daze this was rejection of a standard from
KM> Microsoft (there was a lot of that -- if MSFT does it, we won't,
KM> or will do the opposite, even if it's the only sane method) but
KM> now it's just... not sexy programming, so no one does it.
Usually pretty much if there is a need then we will do it, otherwise not worth it.

Sometimes we won't do it even if there's a need, to thumb our
noses at the Man. Or why for a long time linux had the crap audio
format off OGG, but Would Not Do MP3, because it was Encumbered
by Patents. In the early days, forget if it was GIF or JPG but
one was still under patent so we won't touch it, you will use PNG
and like it.

The underlying explanation to the why's.

There's a point where Principle becomes Cutting Off Your Own
Nose. Debian did a lot of that; until recently you couldn't run
"non-free" drivers, so if you had X Y or Z hardware that hadn't
given their driver source code to linux for free, you were SOL.
I remember when that nixed all of NVidia as vidcards, despite
being the most common in linux's largest target market (PC
builder enthusiasts).

Right, and that probably explained in my old notes there are flip-flops: "Nvidia won't work, use AMD" "Nvidia is better but check
compatability".

> KM> It's a nuisance, but... better than fighting with WINE. And XP
> KM> will always speak to the whole network, which linux never will.
> KM> (Does not like random other linux boxen, never mind random
> KM> Windows. And cannot be trusted to write files without fragmenting
> KM> them all over the place.)
> Could be. (Not disagreeing with you, just insufficient mackground on my
> end.) I do agree it does seem odd Linux doesn't have a defregment
> option -- hey: good reason to do a fresh install as that puts the files
> back together! <g>
KM> The claim is that linux doesn't fragment, because it just keeps
KM> moving on out to the largest available single block (which of
KM> course wastes a LOT of drive space).
KM> I beg to differ. The red blocks are a single file, written by
KM> linux; the blue blocks are files of similar (rather large) size
KM> written by Windows.
KM> http://doomgold.com/images/linux/fragmented.jpg
To my thinking the only way to keep a file from fragmenting is to have a rule it needs to be laid down in one contiguous section -- and that
might a bit of added difficulty with those bad segments.

That is in fact how DOS does it, if there's room. That's
supposedly how linux does it. Except the evidence is that linux
does anything but. One suspects if there is ONE fragment, it just
blows the rule away entirely and writes segments any damn place.

So may as well say it writes in segments; just appears like it doesn't
at the beginning because there is so much available room and 'easier'
not to figure out where to fragment.

This is why I do not trust linux to write archival files, and now
always either have Windows-via-the-network or a sacrificial
external drive (that I don't mind periodically reformatting) act
as the middleman.

Here it seems better/easier to use a combination of backup to two
devices if important, one device if not so important. ...And every so
often check the backups are actually being done!

And my observation is that the linux filesystem is much more
fragile in the event of an insult such as an ungraceful shutdown
(and in the event, fsck will often just delete affected files,
sucks to be you.)

I don't recall having that problem -- ungraceful shutdowns, yes; loss of
files, no. Have lost some work because I wasn't able to save/update a
file before the system freeze, but that's not what we're talking about.

I have also had linux delete whole swaths of the sacrificial
drive because ONE file failed to copy TO that drive.

Linux senses your dislike and retaliates!!

Cloud enterprise gets away with it because they have honkin' big
tape backups, not because the linux filesystems are so reliable.

Backups are good!!

The fragmentation discussion almost points to a very good reason to
start with a fresh system rather than upgrade. I don't recall reading
about it specifically as they always seem to say 'files' which sort of
gets thought of as the data -- stuff added by humans. Seems to me the Operating System files could just as easily get fragmented and
eventually have problems.

Really, it means only use SSDs and assume that periodically the
system will still need a copy-and-back defragging. And never,
ever write user data (especially cache or swap) to the same
partition as the OS files.

Here the "big systems" have a SSD for the OS and hard drive for the
data. Off-hand not sure where the swap goes (which storage device).

KM> The file was damaged beyond use (fortunately not the only copy). Backups are good! Multiple backups are better! Multiple backups in different formats are even better!

Ya think??!

Yup! <g>

KM> But yes, for ages the only way to defragment linux was the same
KM> way we did it in the DOS 5 era and before -- copy everything off,
KM> format the drive, copy everything back. And that's still really
KM> the typical method, if you're using a HDD. (Not so relevant with
KM> SSDs.) Or use huge drives and expect to waste about a third of
KM> the space.
Which is essentially creating a new system. ...Over the years I've sort
of learned to buy/use a lot larger computer hardware than I think I'll
need because something will come along to use all of what I have now.
First XT had a 20 MB hard drive -- I'll never run out of space! Nine
months later space is running low, the sister XT becomes available with
a 40 MB hard drive (and EGA card -- woo-hoo!! [Original was CGA.]
Network the two, so 60 MB available. I'LL NEVER RUN OUT OF SPACE!!!

Along come 12TB drives, and.....

Right now it appears I won't need anything that large, though a couple
of systems here are using 4 TB drives.

KM> Well, there's this, but apparently it's not very good and there's
KM> debate over whether the drive should be mounted or not. And by
KM> default it only does one file.
KM> https://www.howtoforge.com/tutorial/linux-filesystem-defrag/
KM> Read also the comments.
KM> btrfs filesystem reportedly has its own stability issues....
I'm not so sure there is any one great (or even good) file defragger. It would seem each formatting style has it's own rules which would need to
be specifically addressed by the defragger. Would also seem each hard

That is true -- it has to know the filesystem and be thoroughly
aware of any inbuilt defects. Same as you can't use a FAT file
utility on an NTFS file system; it doesn't know how to handle it
and will just mark it as Bad!

Makes sense: "if I don't know what to do with it then it doesn't meet my parameters so must be bad".

Had a client manage to do this. Used a FAT16 disk maintenance
utility on a FAT32 drive... ooops. Fortunately, I still had their
old drive (just replaced) as backup...

That was fortunate!

drive manufacturer and version/family would also need to be address as
well -- why would Western Digital create Black series for regular use
and Red for a NAS? Data is data, but one group has faster writes while another faster reads. ...Just seems to be too many variables.

NAS is expected to be more archival, doesn't need the speed, but
needs max data reliability (which may mean more platters per TB).
Black is a much faster drive, intended for desktop and gaming
use. I have long thought Blue and Green were QC grades rather
than different drives.

The lower grades make sense: why discard a good unit just because it
doens't meet the upper standard? Lots of people out there who would be
happy to pay a little less for a slightly slower drive.

And some of it is just marketing.

Ooo! Pretty lights and flashy colours!! ...Some time back I bought some
(to me) off colour RAM. The specs wre identical, just (say) $100 for
black and $90 for red. The computer didn't care, I'd only see the RAM
when I go inside. ...Eventually spent the $10 on something else.

And I'll bring up another tangent to file placement: I've noticed after
a reboot if I ask for a GUI list of files in a subdirectory the first
time it takes a while -- the more files the longer I'm waiting. But generally any time after the first inquiry it is an instantaneous
response. Seems there's something working to counteract any slow
response due to fragmentation.

Has to rebuild the file cache, and apparently in Ubuntu that
cache is not persistent.

Appears so. ...Maybe written to RAM for speedy access?

Also if you have spinning rust drives set to sleep after X-long,
it takes about 30 seconds for the drive to wake back up.

Right. I think this one is one all the time; would seem if sleeping
then I'd have the wait issue daily.

> I'll admit to having problems 'seeing' other computers around here but
> the correction usually was to fix the ssh connection and then Remmina or
> TigerVNC would run correctly. Both of those do have some sort of a
KM> I think it must be a more general linux thing. Someone I know
KM> sent me a tutorial on how to get Mint to behave on a network and
KM> it was a lot of bother, so I haven't tried it yet. Need to look
KM> on old system for the email....
You need a larger hard drive to conveniently store the trivia! <g>

Why is the trivia the size of the universe??!

That's a piece of trivia unto itself!

Some of the issues on my side are caused by the way I do things: I'll
use the same physical Raspberry Pi but with a different (base ^*) SD card the certificates won't match and the easiest and quickest way, plus the
way I usually find the problem, is to SSH into the remote unit: "certificates don't match" error pops up, contains the fix command line
-- copy and paste, done!

That's using a server function (do I know each piece of hardware
by serial number?) on a desktop, aka that's just dumb. But linux
sometimes does that. Once killed a Debian install by cloning it
to a larger drive.

I've done a couple of moves of a Raspberry Pi OS (Buster, I don't think
as recent as Bullseye but possibly) successfully, but probably 16 GB to
32 GB, and isn't that in the same 'section' for FAT?

And the current Debian install (which hadn't been up in a year or
so) is going "servers? what servers??" meaning it's gonna be a
full reinstall instead of an upgrade. Good thing for it that it's
just experimental and not an everyday-use setup.

May be a sneaky way to get around the upgrade problem!

"Base" card up there because I can't think of a better word. I can make
a duplicate of the card and swap it in -- everything will work properly. Format the card or put a different card in, even with the same OS, and
the unit will now be 'different' to the network: someone else is impersonanting that IP! The certificate doesn't match!

Yeah, it's written what amounts to a serial number to the disk,
and YOU CHANGED IT!!! Windows servers do the same thing and then
some. You can't just swap a drive, it's assumed you murdered the
first one and replaced it with a golem. On the Giant Server (now
being gradually dysmangled for parts) I heard about it when I put
RAM sticks back in a different order!!

Wasn't the same sequence therefore different!

¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Little known facts: cow farts come from the dairy air.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> If I'm just experimenting, I use real hardware and a stack of
KM> small hard drives... much more reliable results. One PC, fifty
KM> OSs. <g>
Probably better to use real hardware than 'pretend' hardware but again a matter of "all depends". I don't test/try stuff all that often and the expense of getting hot swap enclosures would probably really make a 'no'. Plus I'm running out of room: another computer plus storage of those
hard drives!

My experience is that LiveCD, Virtual Machine, and Real Hardware
are not equivalent, in behavior or performance. Many a Linux
LiveCD can see the other PCs on the network, but the same one
INSTALLED cannot.

My experiementation and recollections aren't sufficient to verify or
dispute. Here usually LiveCD is to do an installation and at that point
I don't care if that computer sees the other computers or not.

Virtual Machine.... know it sometimes has problems even transferring
data to the host machine even with that scratchpad function set. (Ever
do SneakerNet from one machine to the same physical machine?!)

KM> But when I need to run something that the host OS doesn't like...
KM> that's why I do VMs.
You do a heck of a lot more experimenting than I do!

A whole lot. <g>

But still easier and safer in the long run to experiment on a disposable machine, even if doesn't always work the same.

Still preferably on real hardware, but am hunting for a VM that
will work on Roadkill... my regular XP VM wouldn't even finish
loading.

First thing that comes to mind is insufficent space. ...Checking
mine..... 32-bit XP, Motherboard tab has 'Enable I/O APIC' tic'd.
Everything else seems relatively normal.

KM> Did you see the crazy thing (I think it was) MJD did, with VMs
KM> inside of VMs until it went all the way from Newest Windows to
KM> Oldest Windows??
No but sounds like an interesting project! Wonder if considered going
back to MS-DOS?! ..Wonder how much storage it takes? Presume on a
NVMe just for a reasonable speed to load the most current version, then
the next from that, and the next from that one....

LOL, you can do that, if you have enough RAM!

Can that be increased with Virtual RAM?!

> > Yes, the 'rolling upgrade process' can be a litte detail-filled!
> KM> "Rolling" is supposed to refer to the software, not the server
> KM> cabinet!
> Only when forget to set the brakes on the cabinet's wheels!
KM> OUT OF THE WAY, BERTHA IS ROLLING!!!!!
When I was growing up a strip mall was created on a swampy area. They filled in with the the usual dirt (guess from other construction sites)
but also trees. ...You know what happens when trees rot or even the branches collapse? Yup! So the store floors are supported by what's underneath, which sometimes wasn't there. ...Eventually if one didn't
hold on to shopping cart it would roll away!

LOL, the old Costco here (they've moved) had that issue. The
parking lot was atop what used to be a dump. Flat when first
paved, but a few decades later it was up hill and down dale in
every direction, tho with the largest dent toward the middle. And
I mean a serious slope, not just a little dip!

Mine might have yours beat: I recall there were some sinkholes in the
stores (probably also the parking lot but I was too young to drive so
didn't pay attention) which were cordoned off. We didn't do it but I
remember Dad commented on running with the xcart would be like a
rollercoaster.

> KM> I've come to greatly prefer a rolling distro for an everyday
> KM> desktop. There are NO UPGRADES and therefore NO REINSTALLS.
KM> Regressions have been pretty rare. They all do some degree of
KM> updating between, but the real point is there is no such thing as
KM> a release version, because it just keeps trundling to the future
KM> without the need of silly version numbers.
"Beeg number good! Impressive!" Personally I don't use the version

LOL, there is that. Then again, numbers like 0.29 are not
impressive.

Only when shopping and there's a dollar sign prefixing!

numbers other than to keep track: right now I need '31' for MythTV and
it pretty much doesn't matter which version of the OS it runs on. As
for the OS in general, I'm not a latest-and-greatest type of guy as for
the OS in general but do sort of like to keep the thing up-to-date for security and other patches.

Came across some idiot on Youtube "demonstrating how unsafe XP is
online" .... first thing he does is DISABLE THE FIREWALL, and
naturally it immediately collected every circulating network worm
or virus. Uh, stupid, do that with ANY OS and it'll have the same
thing happen!!

Right! Take any current system, disable firewalls and security stuff,
and see how long it lasts! ...I'd be willing to bet my old DEC Rainbow
100 running DOS 2.11 would be trashed quickly. ...Well, might take a
while: as I recall 4Kbps modem.

And then he says, "I don't think the firewall is much good" ...
<headdesk>

Once disabled it is no good!

> Around here I'd prefer manual version upgrades, meaning to go from
> version 22 to version 23, not the more-minor updates. Have had old
> computers no longer work properly with upgrades: IMO not a fault of the
> OS, though one could say it didn't check for compatibility.
KM> Yeah, that is a problem. And yes, it should do a compatibility
KM> check, but that's not sexy programming, you should just do a
KM> reinstall and stop bothering us.
I've noticed Mozilla appears to do a full re-install instead of an update and I think LibreOffice does too.

A lot of these monolithic programs do that. However, IIRC Ubuntu
is now all containerized (or at least I heard it was going to
be), which means you always replace the whole thing.

Could be. I had thought they would stick with the compartmentalization
so if one thing breaks it doens't take down the whole thing.

A: Presently, it can't, if your install is over ... about a year
since it was last updated?? anyway, it whines that I must
authorize a different server, but zero information on how that's
to be done. Screw it, in the same time I can just install a new
one. Not like I want to save anything but the wallpaper.

I've done upgrades on Ubuntu (so by extension Debian) but on smaller,
more basic machines. (By 'more basic' I'm meaning not too much has been added/modified from the original.) OTOH I do tend to go for a
from-scratch upgrade install because something major has been changed in
the machine, or a setting is wrong (old example: root set too small) and while something like the root partition can be expanded (I've done it)
from my past experiences (Linux and Windows) it's just overall better to start with a clean slate.

Generally, but there's another advantage of Rolling...
If I have to reinstall with every version upgrade, I won't use
it. That simple.

LIS in some other message I tend to do a full upgrade just because it
makes more sense: I usually have a new (updated) machine and so have
changed enough it makes more sense to discover all the hardware new than
to have to new OS look at the old list and make revisions from that. (I
know I'm using human-thinking method.) Plus over the decades with
MS-DOS, Windows and a few flavours of Linux I haven't had the greatest
luck in upgrading and having everything work properly the first day.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Computer Cooking: BACKUP: Leftovers
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

> KM> Tho my clients were all Windows...
> Though your magic was plugging in the USB thumbdrive with your favourite
> Linux flavour and fixing the Windows problem that way! <g>
KM> That's my magic for recovering data off drives with busted
KM> passwords...

Apparently he encryptio isn't where I expected.

Encrypted drives would be different, yeah. Then you need the password.
But these are just login passwords, and that's readily bypassed.

> > > KM> Whine whine whine!
> > > It's 1700 somewhere! <bseg>
> > KM> <tips tricorn hat>
> > (Hmm: did Ky slide over to the year or did he miss the 24-hour version
> > of 5 o'clock somewhere?)
> KM> You need to be specific, otherwise I take the one that's a better
> KM> fashion statement.
> So now tip your tricorn at 5 o'clock!
KM> I shall eagerly await this turn of the clock!

Bad news: the batteries in the analog clock died!

<checking> The battery is nine years old, but the second hand is still moving....

Wait. Wouldn't that be the third hand?
Or maybe the gripping hand??

> > KM> Generally if I'm using linux I'm using linux, and I expect things
> > KM> to behave like linux.
> > The problem I'm finding is some utilties only use Windows, and some
> KM> And there's my problem...
> I have a bit of a suspicion there's some sort of under-the-table
> agreements for the 'Windows exclusively', otherwise why would companies
> manufacture for only part of the market?
KM> Linux desktops are something like 4% of the desktop market, and
KM> most of that is adjunct to the used-hardware market, and the
KM> stats probably includes Android tablets (Android is essentially
KM> linux).

I would guess part of the raeson for the only 4% is Linux wasn't really marketed until recently, and if one can save money by using the old
equipment then let's use the old equipment.

No, it's because the vast majority of the desktop market is enterprise business (you buy PCs by the each and keep 'em til they die; they buy
'em by the pallet or the truckload and replace them every three years,
when they fall out of warranty support), and enterprise business needs *legally* supported software. Only RedHat does that, and only for
servers, and even then linux does not run the Adobe and Microsoft and
AutoDesk products that are most of what business uses.

Linux is really only *legally* supported in the narrow server market,
and ONLY for the base server OS, and ONLY for RedHat and possibly Ubuntu Server and a couple others, none of any interest to the consumer
desktop. And not for ANY desktop hardware.

And there's legal liability. My sister's architecture firm (she's risen
to 2nd in command after the founder and knows of what she speaks) is big enough to have about a hundred offices worldwide, and its own legal
counsel. Who says you will NOT use unsupported or not-industry-standard ANYTHING (not even company vehicles can be out of warranty and support) because if you do and something goes wrong with that building you
designed, YOU ARE LIABLE JUST BECAUSE OF THAT. Doesn't matter if you
screwed up or not, that is how the COURTS and JURIES will see it. So you
will use CURRENT VERSION of MAINSTREAM COMMERCIAL SOFTWARE and LIKE IT.

You can design a building perfectly well using FreeCAD. You also open
yourself up to millions (potentially billions with today's construction
costs) in legal liability that you would NOT be exposed to if you had
instead used industry-standard software, which is the current supported version of AutoCAD, period. There's the relevant phrase: INDUSTRY STANDARD.

And people use at home what they use at work. It's just easier, because
you don't have to learn a second OS and wholly different way of doing
things. Especially not when the linux way of doing things is often
contrary and difficult.

So there's no incentive to use linux on the personal desktop, outside of hobbyists and the perverse, who are a tiny market segment. Frankly I'm surprised it's as high as 4%; at peak Apple with all its fanboys only
had 20%, and now it's a lot less, even tho Apple puts a lot of money and effort into marketing and support. (Back then Apple had the monopoly on graphical apps. As of Win95, Windows ate Apple's lunch.)

Which means outside of a tiny slice of those hobbyists and perverse,
there's no market for linux desktops. Dell tried. It didn't go anywhere (presumably because not only was there no real demand, but also because after-sale support was a nightmare, hardly worth the $5 to $30 per unit
they saved on licensing Windows; likely the same reason they stopped
selling bare-metal desktop systems too.) What there's no real market for
is doing astonishingly well to have 4% of what is logged by internet
servers (where the stats come from).

KM> No one in their right financial mind would spend
KM> resources on a linux desktop version that requires more than
KM> setting a target flag in the compiler. This is much easier with
KM> modern compilers targeting modern OSs, hence a lot of major apps
KM> (eg. LibreOffice, Softmaker Office, the KDE stable of desktop
KM> apps) are often available across platforms.

Makes sense.

KDE even runs a "binary factory" that automates cross-platform builds (Windows, Mac, Android, others), tho that lately migrated to Gitlab and
is no longer the nice handy interface it used to be. *($# if I can
figure it out now. It's just bloody awful.

https://invent.kde.org/explore/groups?sort=name_asc

KM> However, hardware-level utilities are generally coded in Assembly
KM> Language, which does not port easily to another OS, precisely
KM> because it does its own hardware access. Same reason drivers
KM> aren't portable as such.

So might have several 'barriers'. First is the compiled for the
original OS. The human creating the code is trained and familiar with
that OS, so to create code for a second OS needs to learn that also.
..A lot more details in there, of course.

And all that for a market segment that doesn't really exist.

> KM> And given it's a flash drive... unless it's purely a filesystem
> KM> error, AFAIK it's not recoverable.
> Highly limited personal experience here so probably yes. Most of the
KM> It becomes like recovering data off a regular memory chip: it
KM> just ain't there.

Solid state drives (of whatever format) have numerous good points but
also some major bad points. Backups are even better now!

Yeah, that exactly. And some lose data if they sit unpowered for very long.

Which is probably why they also suck a laptop dry just from sitting
there turned off.

> problem cases are off-brands, though I had a quirk with all of the black
> and yellow ADATA thumbdrives eventually failed yet all of the black and
> blues ones are fine. Same seller, same capacity, purchased about the
> same time. <shrug>
KM> I remember that. Likely different supplier for the chips and/or
KM> logic boards. Plus ADATA had a longstanding repute as crap, so I
KM> was not surprised. <g>

The good news was I didn't loose anything other than time.

*whew* !!

KM> Otherwise, I wouldn't pay money for a flash drive of any
KM> description.

I've mainly used them for SneakerNet and Live Boot.

Yeah, that sort of thing where it's disposable, okay. NOT for everyday
or storage.

KM> There's a point where Principle becomes Cutting Off Your Own
KM> Nose. Debian did a lot of that; until recently you couldn't run
KM> "non-free" drivers, so if you had X Y or Z hardware that hadn't
KM> given their driver source code to linux for free, you were SOL.
KM> I remember when that nixed all of NVidia as vidcards, despite
KM> being the most common in linux's largest target market (PC
KM> builder enthusiasts).

Right, and that probably explained in my old notes there are flip-flops: "Nvidia won't work, use AMD" "Nvidia is better but check
compatability".

Yep, there ya go.

> KM> http://doomgold.com/images/linux/fragmented.jpg
> To my thinking the only way to keep a file from fragmenting is to have a
> rule it needs to be laid down in one contiguous section -- and that
> might a bit of added difficulty with those bad segments.
KM> That is in fact how DOS does it, if there's room. That's
KM> supposedly how linux does it. Except the evidence is that linux
KM> does anything but. One suspects if there is ONE fragment, it just
KM> blows the rule away entirely and writes segments any damn place.

So may as well say it writes in segments; just appears like it doesn't
at the beginning because there is so much available room and 'easier'
not to figure out where to fragment.

IOW, it fibs about how it does things, and what you won't admit happens
you can't fix.

KM> This is why I do not trust linux to write archival files, and now
KM> always either have Windows-via-the-network or a sacrificial
KM> external drive (that I don't mind periodically reformatting) act
KM> as the middleman.

Here it seems better/easier to use a combination of backup to two
devices if important, one device if not so important. ...And every so
often check the backups are actually being done!

Backups go on NTFS filesystem here, period. Written by Windows, any Windows.

KM> And my observation is that the linux filesystem is much more
KM> fragile in the event of an insult such as an ungraceful shutdown
KM> (and in the event, fsck will often just delete affected files,
KM> sucks to be you.)

I don't recall having that problem -- ungraceful shutdowns, yes; loss of files, no. Have lost some work because I wasn't able to save/update a
file before the system freeze, but that's not what we're talking about.

Nope. This is "instead of fixing the file table entry, we'll nuke the
file". There is no predicting when it will do that, either, but if it
decides it needs a long session with fsck, you can count on it. You
think the file is still there, and it's not.... fortunately all it's
eaten are Youtube downloads, but some are no longer available.

KM> I have also had linux delete whole swaths of the sacrificial
KM> drive because ONE file failed to copy TO that drive.

Linux senses your dislike and retaliates!!

This is the linux I love! But it cheats on me anyway!!

> The fragmentation discussion almost points to a very good reason to
> start with a fresh system rather than upgrade. I don't recall reading
> about it specifically as they always seem to say 'files' which sort of
> gets thought of as the data -- stuff added by humans. Seems to me the
> Operating System files could just as easily get fragmented and
> eventually have problems.
KM> Really, it means only use SSDs and assume that periodically the
KM> system will still need a copy-and-back defragging. And never,
KM> ever write user data (especially cache or swap) to the same
KM> partition as the OS files.

Here the "big systems" have a SSD for the OS and hard drive for the
data. Off-hand not sure where the swap goes (which storage device).

SSDs and NVMes here for workspace and OS, HDD for storage.

> Network the two, so 60 MB available. I'LL NEVER RUN OUT OF SPACE!!!
KM> Along come 12TB drives, and.....

Right now it appears I won't need anything that large, though a couple
of systems here are using 4 TB drives.

Famous Last Words. <g>

> > drive manufacturer and version/family would also need to be
address as

> well -- why would Western Digital create Black series for regular use
> and Red for a NAS? Data is data, but one group has faster writes while
> another faster reads. ...Just seems to be too many variables.
KM> NAS is expected to be more archival, doesn't need the speed, but
KM> needs max data reliability (which may mean more platters per TB).
KM> Black is a much faster drive, intended for desktop and gaming
KM> use. I have long thought Blue and Green were QC grades rather
KM> than different drives.

The lower grades make sense: why discard a good unit just because it
doens't meet the upper standard? Lots of people out there who would be
happy to pay a little less for a slightly slower drive.

Same reason we have Celeron and lower-speed CPUs to this day. The ones
in the center of the wafer are better quality. Around the edges, to
varying degrees less so. But still perfectly good working chips, just
can't count on 'em being as good. But that's also why so many cheap CPUs
can be drastically overclocked -- they're actually as good, but can't
count on it for sales purposes.

KM> And some of it is just marketing.

Ooo! Pretty lights and flashy colours!! ...Some time back I bought some
(to me) off colour RAM. The specs wre identical, just (say) $100 for
black and $90 for red. The computer didn't care, I'd only see the RAM
when I go inside. ...Eventually spent the $10 on something else.

LOL. I don't want all that bling, it's needless heat and power consumption.

So buy server RAM and no bling. <g>

> And I'll bring up another tangent to file placement: I've noticed after
> a reboot if I ask for a GUI list of files in a subdirectory the first
> time it takes a while -- the more files the longer I'm waiting. But
> generally any time after the first inquiry it is an instantaneous
> response. Seems there's something working to counteract any slow
> response due to fragmentation.
KM> Has to rebuild the file cache, and apparently in Ubuntu that
KM> cache is not persistent.

Appears so. ...Maybe written to RAM for speedy access?

Evidently.

KM> Also if you have spinning rust drives set to sleep after X-long,
KM> it takes about 30 seconds for the drive to wake back up.

Right. I think this one is one all the time; would seem if sleeping
then I'd have the wait issue daily.

Likely so. One of mine silently runs an MP3 all the time because
otherwise it won't stay awake. The downside is it records a spin-up each
time the file plays. It's recorded something like 17,000 spin-ups.

> You need a larger hard drive to conveniently store the trivia! <g>
KM> Why is the trivia the size of the universe??!

That's a piece of trivia unto itself!

It never ends!

> Some of the issues on my side are caused by the way I do things: I'll
> use the same physical Raspberry Pi but with a different (base ^*) SD card
> the certificates won't match and the easiest and quickest way, plus the
> way I usually find the problem, is to SSH into the remote unit:
> "certificates don't match" error pops up, contains the fix command line
> -- copy and paste, done!
KM> That's using a server function (do I know each piece of hardware
KM> by serial number?) on a desktop, aka that's just dumb. But linux
KM> sometimes does that. Once killed a Debian install by cloning it
KM> to a larger drive.

I've done a couple of moves of a Raspberry Pi OS (Buster, I don't think
as recent as Bullseye but possibly) successfully, but probably 16 GB to
32 GB, and isn't that in the same 'section' for FAT?

Uh, no....FAT is DOS filesystem, Pi probably uses ext4 or some such.

KM> And the current Debian install (which hadn't been up in a year or
KM> so) is going "servers? what servers??" meaning it's gonna be a
KM> full reinstall instead of an upgrade. Good thing for it that it's
KM> just experimental and not an everyday-use setup.

May be a sneaky way to get around the upgrade problem!

LOL. Not so sneaky, and of course it wants to do an update during
install. NO NO NO I don't have six hours to babysit the damn thing....
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> Boo!
Boo who?!!

No crying until you spill the milk!

OK, I spilt it into the measuring cup, now what?

..Thought I figured soming out but now makes less sense. If a 'boo' is
a boyfriend/girfriend/lover ("he/she is my boo") and 'boo' is said by a ghost or to scare someone, why is it good to be a 'boo'?

Um. I have no good answer to this. Maybe I'll think up a silly
one instead.

I'm not sure I can tell the difference. <rs!>

That makes sense. I'm thinking it could also mean a faulty bit which is causing a bad instruction and so garbage output. That one based on
years ago when I built a Heathkit TV and the overlay display (time,
channel, etc.) was sometimes scrambled. A little analysis and only
certain letters: let's say "C" so "Channel" might display as "Thannel"
and "MAY" as "MAR".

That's... interesting!!

It became obvious what and almost where the problem was once made a
binary chart:

01000001 A
01000010 B
01000011 C
01000100 D
01000101 E
01000110 F
^
If the 3rd LSB (marked) opens/go low then an "E" displays as an "A" ==>
101 becomes 001.

Figured out via ASCII Chart the problem was a certain bit level was
being set wrong: let's say 3rd LSB (out of the 8). Figured out where
the problem was on the schematic (between encoder and decoder) but which
was LSB and MSB? (I'm not that good and pre-web access to get the chip specs.) So ended up making four reair solders: two on each end of the
trace and two since I wasn't sure which was LSB and MSB. Ta-dah! No
more misspellings!

Oh! Clever solution. But when you've got a schematic to work
from....

Came with the kit manuals from Heathit. And not that I can read the
whole thing and understand but knew enough from previous electronics fiddlings. Actually still have the schematic on the cork board (which I
use as a door to cover some shelves with parts boxes) just because it
looks neat.

KM> Gravity always wins!
At least I don't have to go looking for dropped parts on the ceiling too!

This is a very good point.
Well, unless you're on the space station.

The good news is I think they have minimal gravity so anything dropped
just kind of floats so unless accidentally swatted whould be easy to
find. Bad news is if they don't find it could be catastrophic whereas I
just grumble and get a new one.

> KM> This isn't absolute but it's a pretty good guideline.
> Good starting points which usually work.
KM> Usually good enough, if it's hardware. If it's software, can be
KM> any damn thing.
Though even software follows rules. I remember one script I was

It's supposed to. When there are not rules it makes up its own.

Thought probably based on established rules, though which ones can be
'fun' to figure out.

Someone discovered that the value of Pi used in DOOM was wrong
(but it was a lookup table, so the values were fixed). So he
tried lots of other values. Correct Pi was close enough. Others
went from strange to nonfunctional.

I would suppose there would be some out-of-range occurrences.

https://media.ccc.de/v/mch2022-236-non-euclidean-doom-what-happens -to-a-game-wh
n-pi-is-not-3-14159-

I only watched part of it (never played Doom); did see the demo of the
pi = 3 and the walls didn't act correctly: expanded then contracted.
Also the e-mail prior to that section where some unhappy user suggested
the person who mis-entered the 10th digit of pi in table be fired. heck
I'm lucky to remember the 3.14 part! ...Don't really need pi for
anything -- let's see, last time I had to figure the circumference of a
circle was in nineteen........ There was also a chart pop-up: I thought
the problem was simply typing a 7 for the 4 (miskey on the NumPad)
rather than memory errror.

creating, and since just self-trained a lot of pluck some code from
here, some other code from there.... Something wasn't working so put in
an echo statement to display the output at a certain level of the
script. That helped figure out the problem, so now getting the correct result per the echo statement but the next step was always displaying
"0". Used to semi-work before! Ends up the following statement was
taking the output of the echo statement: echo finished its job properly
so outputted a '0' for 'good out'. Comment the test echo and now
worked!

Ah yes, the old step-through method... usually a Good Idea when
you can.

Probably not the best/most efficient but as I barely know what I'm doing
it worked enough to get me to the problem. And also learned once find
the problem hide the thing that found the problem!

KM> Modern OSs now understand this stuff, and have done so pretty
KM> reliably for about 25 years. Part of the longstanding problem
KM> with linux was that until about 10 years ago, or a bit longer for
KM> some of the more advanced distros, you had to know your hardware
KM> parameters and sometimes input them manually. It seems to have
KM> finally got this right. When you're doing it for free, as has
KM> mostly been the case, hardware programming is not near as sexy as
KM> cute apps and games, so it gets way less attention.
Yes, I vaguely recall having to initially configure the video settings.

I remember it vividly, having installed RedHat 6 which at the
time you had to set up X entirely yourself. Make a good guess and
get a screen. Make a wrong guess and start over.

So back to my "always have a spare system" so have something that works,
and can research why the new system isn't working as expected.

Now pretty much does it itself, though that occassionally creates a
problem when using a 4K TV and so viewing from ten feet away! <g>
..Even the resolution selectrion GUI is teeny-tiny when standing right
in front of the TV! <g>

LOL. Some of my failed Virtual Machines have been the size of a
postage stamp....

So far not quite that small here!

Did get Win2K VM installed on Roadkill tho, so now I have a
restful grey workspace when I need it. Win11 lacks this.

Serenity NOW!!!! <g>

> So if I remember correctly Mike's problem was after a while his monitor
> would go black and need a reboot to get things going again. I don't
> recall if he stated a time but seemed he implied weeks or months. I
> want to get in on the thread because I have a similar problem with one
> essentially headless system which tends to not talk to a monitor after
> some time -- I'm thinking around 1.5 to 2 months.
KM> Fedora used to do that, tho the problem seems to have gone away
KM> as of v39. Except it would decide to not speak to the outside
KM> world after about a week, unless regularly rebooted. Meanwhile
KM> PCLOS had been running for several months, and WinXP for about a
KM> year and a half, with no such issues.
Almost seems like 'something runs out of room'. I have some computers

Yeah, that is usually why a system goes goofy over time -- some
cache or data stack gets full or corrupted or wraps around, but
the effect is the same.

I have some of the computers here configured to reboot weekly to avoid problems. ...A couple of versions of Motion ago on Raspberry Pi's
Bullseye was really 'interesting': definitely had to be rebooted weekly
if not sooner: had Watchdog running but also another script of my own
creation to check for keywords in various error files and either restart
Motion or in a couple of cases do an actual reboot. (The current or
maybe one-off on Bookworm does not have these problems.)

around here rebooting themselves on a weekly basis because of that kind
of issue. Not necessarily an OS issue -- may have been years ago, but
now more like a problem with some utility causing the lockup/overrun.

That was often the issue with Windows. Not a durn thing wrong
with Windows; it was the crappy "fix something" utility that
broke it.

So the thing to fix the break broke the thing when it wasn't broken!

(Thinking of a couple years back when had three RPi4's running a previous version of Motion and other RPi4 running same OS but Motion not
installed. The Motion ones needed frequent rebooting, the others not running Motion would run for months. [Important note: Motion has been corrected since then!])

That's good! Cuz frequent rebooting... not my idea of Quality.
Been ruined by software that runs for months on end. RoughDraft
and WinAmp have been up since... probably last October, when the
system was last rebooted. SeaMonkey needs to be restarted every
few weeks, tho, or it gets sluggish.

Right: improper shutdown is a last resort. With the old Motion/Bullseye
issue above I had tried various repairs/updates, etc., and just didn't
fix. Definitely something with that combination as other RPi/Bullseye
systems would run for weeks.

LibreOffice has been up since March, that being when I started
work on the current going-very-slow paid edit (it's not on a
deadline, and it makes my brain hurt).

Mine hasn't been up nearly that long due to a 'restart needed to finish updates' on May 11. Right now have seven LibreOffice documents open as working on a few projects at once and the others are for semi-active
stuff I want to be reminded of. ...In the mean time have revised other
dox (open, update, save and close) and created new.

...Read somewhere LibreOffice isn't designed to be kept up that long -- overnight starts causing a few quirks. I'm thinking might be more of
the change in the OS's date and 'borrowing' of RAM for maintenance
projects so LibreOffice can't find where it left some of its stuff.
Here I have quite a bit of RAM so might be why it takes a while to see
some of those problems.

KM> As we skeap ?? speak Fedora is sitting there upgrading to v40.
KM> It's been at it since 11pm last night, tho a lot of the time was
KM> 5GB of downloads on a 3Mbps connection. Had to do a full update
KM> first, then run the upgrade (fortunately the commands are still
KM> handy in the console buffer... it started life as v32). It is now
KM> running the final step and will be done in about an hour. You can
KM> see why I find rolling a lot less trouble.
Better hope that connection maintains!! ...The long time to do an

Fortunately it caches downloads so you don't have to start over.

Good!

That's one thing Windows Update gets wrong. Anything interrupted
starts over.

Boo!

update/upgrade was one reason I only did one machine at a time: besides multiple connections slowing down the limited bandwidth if something
went wrong like an extended power failure I have only one computer to recover.

Yeah, good policy.

Over the decades I've had all sorts of quirkies occur so just easier
doing one-at-a-time. Not to say I haven't multi-tasked -- all depends.

KM> And I get my first look at KDE Plasma v6. Given the newness
KM> thereof (just released) for the next while there will be a lot of
KM> updates, possibly to the point of a Whole New Monkey. But they
KM> are usually pretty good about getting the major bugs out (not
KM> least because unlike say Xfce, which is one guy and change, it's
KM> a team of a hundred-plus folks.)
More people seems to be better as a couple might be experts in video, others, audio, etc. Plus if only if one or two doing the whole job and something happens.....

KDE has about a hundred people involved.

Seems like a good number: leader, group leaders, workers who can
collaborate (hopefully!) and work off each others' knowledge and
experiences.

KM> http://www.wholenewmonkey.com/
Huh: under maintainence -- guess the upgrade and update data is stil lrolling in!

Better not be under maintenance, it hasn't been changed in years!

Surgery joke in box, underneath that a line "site maintenance",
underneath that an oval-swirl thing which I have a hard time reading
because of my 'tint blindness' but connects to be able to write an
e-mail. Under that 'too many' monkeys and clicks to the link.

(Yes, I got this domain solely for this joke. Cuz it's always
funny.)

Go for it! :)

.. A group of flamingos is called a stand.

I thought it was called a lawn ornament!

More stuff to cut around!

("Pull out, dummy!")
(That's what she said!)


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... My wife told me stop impersonating a flamingo; I had to put my foot down. --- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

> KM> Tho my clients were all Windows...
> Though your magic was plugging in the USB thumbdrive with your favourite
> Linux flavour and fixing the Windows problem that way! <g>
KM> That's my magic for recovering data off drives with busted
KM> passwords...
Apparently the encryption isn't where I expected.

Encrypted drives would be different, yeah. Then you need the
password. But these are just login passwords, and that's readily
bypassed.

Just look at the Post-It Note on the monitor!

> > > KM> Whine whine whine!
> > > It's 1700 somewhere! <bseg>
> > KM> <tips tricorn hat>
> > (Hmm: did Ky slide over to the year or did he miss the 24-hour

ersio

> > of 5 o'clock somewhere?)
> KM> You need to be specific, otherwise I take the one that's a better
> KM> fashion statement.
> So now tip your tricorn at 5 o'clock!
KM> I shall eagerly await this turn of the clock!
Bad news: the batteries in the analog clock died!

<checking> The battery is nine years old, but the second hand is
still moving....

And this morning found the battery in the bathroom clock needs replacing.
No idea of when it was replaced last but I'd guess two years.

Wait. Wouldn't that be the third hand?
Or maybe the gripping hand??

Time slips by fleetingly, gripping the human watching, urging to quickly
follow before too late.

As for why 'second hand', has something to do with originally clocks
just had an hour and minute hand, so when added it was the second minute
hand.

> > KM> Generally if I'm using linux I'm using linux, and I expect

hin

> > KM> to behave like linux.
> > The problem I'm finding is some utilties only use Windows, and some
> KM> And there's my problem...
> I have a bit of a suspicion there's some sort of under-the-table
> agreements for the 'Windows exclusively', otherwise why would companies
> manufacture for only part of the market?
KM> Linux desktops are something like 4% of the desktop market, and
KM> most of that is adjunct to the used-hardware market, and the
KM> stats probably includes Android tablets (Android is essentially
KM> linux).
I would guess part of the raeson for the only 4% is Linux wasn't really marketed until recently, and if one can save money by using the old equipment then let's use the old equipment.

No, it's because the vast majority of the desktop market is
enterprise business (you buy PCs by the each and keep 'em til
they die; they buy 'em by the pallet or the truckload and replace
them every three years, when they fall out of warranty support),
and enterprise business needs *legally* supported software. Only
RedHat does that, and only for servers, and even then linux does
not run the Adobe and Microsoft and AutoDesk products that are
most of what business uses.

Makes more sense. I was thinking from the consumer viewpoint. And
yes, if something (hardware/software) doesn't work with what I have/need
then it doesn't get purchased/used. Having a pretty look and just
sitting there doesn't get the job done.

Linux is really only *legally* supported in the narrow server
market, and ONLY for the base server OS, and ONLY for RedHat and
possibly Ubuntu Server and a couple others, none of any interest
to the consumer desktop. And not for ANY desktop hardware.

<Cue _Cops_ theme!>

And there's legal liability. My sister's architecture firm (she's
risen to 2nd in command after the founder and knows of what she
speaks) is big enough to have about a hundred offices worldwide,
and its own legal counsel. Who says you will NOT use unsupported
or not-industry-standard ANYTHING (not even company vehicles can
be out of warranty and support) because if you do and something
goes wrong with that building you designed, YOU ARE LIABLE JUST
BECAUSE OF THAT. Doesn't matter if you screwed up or not, that is
how the COURTS and JURIES will see it. So you will use CURRENT
VERSION of MAINSTREAM COMMERCIAL SOFTWARE and LIKE IT.

Makes sense from the legal viewpoint: maximum protection. ...And I have
no idea how it came up be remember decades ago when I worked at the
college somehow the topic came up of something to do with the stairs.
The person could not sue the college as long as properly maintained but
would sue the original designer or maybe it was the builder.

You can design a building perfectly well using FreeCAD. You also
open yourself up to millions (potentially billions with today's construction costs) in legal liability that you would NOT be
exposed to if you had instead used industry-standard software,
which is the current supported version of AutoCAD, period.
There's the relevant phrase: INDUSTRY STANDARD.

And that's probably the reason for those small books of Terms and
Conditions.

And people use at home what they use at work. It's just easier,
because you don't have to learn a second OS and wholly different
way of doing things. Especially not when the linux way of doing
things is often contrary and difficult.

Well, the college's system was good, the store's system was horrible....

So there's no incentive to use linux on the personal desktop,
outside of hobbyists and the perverse, who are a tiny market
segment. Frankly I'm surprised it's as high as 4%; at peak Apple
with all its fanboys only had 20%, and now it's a lot less, even
tho Apple puts a lot of money and effort into marketing and
support. (Back then Apple had the monopoly on graphical apps. As
of Win95, Windows ate Apple's lunch.)

Any idea what the European (we can limit to the EU) market is? I'm
thinking the use of Linux is significantly higher, and part of that
might be the legal system is more relaxed at times. ...Though looking
at one specific instance the relaxed atttude then is causing a trudge
through their legal system now.

Anyway, seems Linux is more used on that side of the pond, I'm thinking possibly because of a freer thought process. (Though they don't mess
with semi-monopolies like Google, etc.!)

KM> No one in their right financial mind would spend
KM> resources on a linux desktop version that requires more than
KM> setting a target flag in the compiler. This is much easier with
KM> modern compilers targeting modern OSs, hence a lot of major apps
KM> (eg. LibreOffice, Softmaker Office, the KDE stable of desktop
KM> apps) are often available across platforms.
Makes sense.

KDE even runs a "binary factory" that automates cross-platform
builds (Windows, Mac, Android, others), tho that lately migrated
to Gitlab and is no longer the nice handy interface it used to
be. *($# if I can figure it out now. It's just bloody awful.

Yes, seems a couple years ago Git did something which caused a lot of
people to move away; probably what you are talking about. From my
highly limited usage I seem to have more problems finding details now
than before.

KM> However, hardware-level utilities are generally coded in Assembly
KM> Language, which does not port easily to another OS, precisely
KM> because it does its own hardware access. Same reason drivers
KM> aren't portable as such.
So might have several 'barriers'. First is the compiled for the
original OS. The human creating the code is trained and familiar with
that OS, so to create code for a second OS needs to learn that also.
..A lot more details in there, of course.

And all that for a market segment that doesn't really exist.

Or 'barely exists' might be more accurate. Watch out for those timy up-starts!

> KM> And given it's a flash drive... unless it's purely a filesystem
> KM> error, AFAIK it's not recoverable.
> Highly limited personal experience here so probably yes. Most of the
KM> It becomes like recovering data off a regular memory chip: it
KM> just ain't there.
Solid state drives (of whatever format) have numerous good points but
also some major bad points. Backups are even better now!

Yeah, that exactly. And some lose data if they sit unpowered for
very long.

<Looking at small storage container> Oh-oh! ...As for the sitting
unpowered issue, makes sense: effectively a battery in there to keep the
1 or 0 state. Probably 1 as 0 is off but might be some leakage.
Eventually all those 1's use the stored energy.

Which is probably why they also suck a laptop dry just from
sitting there turned off.

It's technically not off but a form of standby.

> problem cases are off-brands, though I had a quirk with all of the black
> and yellow ADATA thumbdrives eventually failed yet all of the black and
> blues ones are fine. Same seller, same capacity, purchased about the
> same time. <shrug>
KM> I remember that. Likely different supplier for the chips and/or
KM> logic boards. Plus ADATA had a longstanding repute as crap, so I
KM> was not surprised. <g>
The good news was I didn't loose anything other than time.

*whew* !!

Really. LIS/implied up there, I tend to have multiple backups of
important stuff.

KM> Otherwise, I wouldn't pay money for a flash drive of any
KM> description.
I've mainly used them for SneakerNet and Live Boot.

Yeah, that sort of thing where it's disposable, okay. NOT for
everyday or storage.

I've had too much problem with them to trust fully. Some of the
problems directly to faulty thumbdrives, some appears to have been to
not knowing the rules. (Apparently USB 3 can do some weird things when
the cable is over ten feet!)

> KM> http://doomgold.com/images/linux/fragmented.jpg
> To my thinking the only way to keep a file from fragmenting is to have a
> rule it needs to be laid down in one contiguous section -- and that
> might a bit of added difficulty with those bad segments.
KM> That is in fact how DOS does it, if there's room. That's
KM> supposedly how linux does it. Except the evidence is that linux
KM> does anything but. One suspects if there is ONE fragment, it just
KM> blows the rule away entirely and writes segments any damn place.
So may as well say it writes in segments; just appears like it doesn't
at the beginning because there is so much available room and 'easier'
not to figure out where to fragment.

IOW, it fibs about how it does things, and what you won't admit
happens you can't fix.

Initially I'll say 'yup' but not so sure some of the problem isn't due
to repeating of incorrect information or taken out of context. I was
looking up the USB 3 cable length to verify was the 3m I remembered,
first hit said something about 90m or some way-off number. I knew it
was wrong so automaticalaly forgot it, but the way it was written if
someone was looking up cable lengths and didn't have an idea they would
have taken that 90m and not looked further. (I think the 90m was using
a powered repeater.)

Not saying the fragmentation issue isn't occurring, just sometimes a
deep dig is needed. For me fragmentation doesn't seem to be causing
problems. Sure, I might have to wait a few seconds for it to sort
through 506 GB the first time (before gets journalled or whatever the
proper term is). A huge business system -- that's bound to cause
problems not seem in my relatively teeny-tiny system. Semi-tangent: I
wait about that long for the hospital's on-line system to give me a
report. No idea how big my personal patient records are, but multiply
that by all the other patients and possibly payroll, general operations, security footage.....

KM> And my observation is that the linux filesystem is much more
KM> fragile in the event of an insult such as an ungraceful shutdown
KM> (and in the event, fsck will often just delete affected files,
KM> sucks to be you.)
I don't recall having that problem -- ungraceful shutdowns, yes; loss of files, no. Have lost some work because I wasn't able to save/update a
file before the system freeze, but that's not what we're talking about.

Nope. This is "instead of fixing the file table entry, we'll nuke
the file". There is no predicting when it will do that, either,
but if it decides it needs a long session with fsck, you can
count on it. You think the file is still there, and it's not.... fortunately all it's eaten are Youtube downloads, but some are no
longer available.

<wagging finger> How many times have we told you to do multiple
backups?!

And by multiple backups I'm thinking for really important stuff use
different types of hardware: years ago I had an audio CD not play
correctly. Thought just dirty (but I know to handle by edges, not use
as coaster, etc.); flipped over -- euwww! Looked like the metal layer
was peeling under the protective layer!

KM> I have also had linux delete whole swaths of the sacrificial
KM> drive because ONE file failed to copy TO that drive.
Linux senses your dislike and retaliates!!

This is the linux I love! But it cheats on me anyway!!

It knows you don't like its friends!

That 'taking a swath' sort of reminds me when I came home from work and
the computer was 'wrong' -- I don't recall the visual but ended up the
hard drive went from 250 GB to something like 20. Failure of the HDD,
even though running some version of Windows at the time. (I probably
didn't know Linux existed; TV recordings were done on multiple VCRs.)

> The fragmentation discussion almost points to a very good reason to
> start with a fresh system rather than upgrade. I don't recall reading
> about it specifically as they always seem to say 'files' which sort of
> gets thought of as the data -- stuff added by humans. Seems to me the
> Operating System files could just as easily get fragmented and
> eventually have problems.
KM> Really, it means only use SSDs and assume that periodically the
KM> system will still need a copy-and-back defragging. And never,
KM> ever write user data (especially cache or swap) to the same
KM> partition as the OS files.
Here the "big systems" have a SSD for the OS and hard drive for the
data. Off-hand not sure where the swap goes (which storage device).

SSDs and NVMes here for workspace and OS, HDD for storage.

Pretty much the same here. The smaller/'less important' systems are
running from a SSD (had to rephrase from what I originally typed because
of the Raspberry Pi's) and those have a backup routine to HDD.

> Network the two, so 60 MB available. I'LL NEVER RUN OUT OF SPACE!!!
KM> Along come 12TB drives, and.....
Right now it appears I won't need anything that large, though a couple
of systems here are using 4 TB drives.

Famous Last Words. <g>

Yup! ...Right now I don't need more, though I know that's what I've
said numerous times in the past.


> > drive manufacturer and version/family would also need to be

address as

> well -- why would Western Digital create Black series for regular use
> and Red for a NAS? Data is data, but one group has faster writes while
> another faster reads. ...Just seems to be too many variables.
KM> NAS is expected to be more archival, doesn't need the speed, but
KM> needs max data reliability (which may mean more platters per TB).
KM> Black is a much faster drive, intended for desktop and gaming
KM> use. I have long thought Blue and Green were QC grades rather
KM> than different drives.
The lower grades make sense: why discard a good unit just because it
doens't meet the upper standard? Lots of people out there who would be happy to pay a little less for a slightly slower drive.

Same reason we have Celeron and lower-speed CPUs to this day. The
ones in the center of the wafer are better quality. Around the
edges, to varying degrees less so. But still perfectly good
working chips, just can't count on 'em being as good. But that's
also why so many cheap CPUs can be drastically overclocked --
they're actually as good, but can't count on it for sales
purposes.

Right -- remmeber you saying that sometime back and from other sources. Personally I'm a bit shy on overclocking: running something constantly
in 'overdrive' mode seems like asking for problems. ...I have over-
clocked a few of the Pi's here without problems. More of a 'gentle overclock': couple of notches but nothing like 'flooring'. With three
Pi's which are essentilly duplicates of each other two could use a
higher value, the third wasn't happy and had to be turned down a notch.
That's more tweaking hardware, so the position on the wafer seems to be
making sense as for a 'why'.

KM> And some of it is just marketing.
Ooo! Pretty lights and flashy colours!! ...Some time back I bought some (to me) off colour RAM. The specs wre identical, just (say) $100 for
black and $90 for red. The computer didn't care, I'd only see the RAM
when I go inside. ...Eventually spent the $10 on something else.

LOL. I don't want all that bling, it's needless heat and power consumption.
So buy server RAM and no bling. <g>

I don't see the inside of my computers except when working on them so
could care less about flashing lights -- in fact I'd find 'em
distracting.

> And I'll bring up another tangent to file placement: I've noticed after
> a reboot if I ask for a GUI list of files in a subdirectory the first
> time it takes a while -- the more files the longer I'm waiting. But
> generally any time after the first inquiry it is an instantaneous
> response. Seems there's something working to counteract any slow
> response due to fragmentation.
KM> Has to rebuild the file cache, and apparently in Ubuntu that
KM> cache is not persistent.
Appears so. ...Maybe written to RAM for speedy access?

Evidently.

MythTV says it does write stuff to RAM for a speedier response. A while
back someone posted a question as to why a certain function in MythTV
was showing 100%/near 100% usage of their RAM when the OS was only
showing (say) 30%. The post was also asking if more RAM was needed.
The answers were MythTV out into RAM what it could, so essentially that parameter would always/eventually show a high RAM usage. ...Seems no
matter what OS is used the general rule is to install as much RAM as
possible was the system will make use of it.

KM> Also if you have spinning rust drives set to sleep after X-long,
KM> it takes about 30 seconds for the drive to wake back up.
Right. I think this one is one all the time; would seem if sleeping
then I'd have the wait issue daily.

Likely so. One of mine silently runs an MP3 all the time because
otherwise it won't stay awake. The downside is it records a
spin-up each time the file plays. It's recorded something like
17,000 spin-ups.

I've seen small TV networks like that: "Good Grief! they're showing this episode again?!". <g> Actually your little keep awake trick is clever!

> You need a larger hard drive to conveniently store the trivia! <g>
KM> Why is the trivia the size of the universe??!
That's a piece of trivia unto itself!

It never ends!

To infinity and beyond!!

KM> And the current Debian install (which hadn't been up in a year or
KM> so) is going "servers? what servers??" meaning it's gonna be a
KM> full reinstall instead of an upgrade. Good thing for it that it's
KM> just experimental and not an everyday-use setup.
May be a sneaky way to get around the upgrade problem!

LOL. Not so sneaky, and of course it wants to do an update during
install. NO NO NO I don't have six hours to babysit the damn
thing....

Start some time in the morning, let it do it's thing while you
periodically check in. ...Lunch ...Dinner ...Overnight ...Check when
wake up...... Of course you could sit there for an hour and as soon as
you move way it needs a user input!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Some people have gingerbread houses - I have a Pizza Hut.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

KM> My experience is that LiveCD, Virtual Machine, and Real Hardware
KM> are not equivalent, in behavior or performance. Many a Linux
KM> LiveCD can see the other PCs on the network, but the same one
KM> INSTALLED cannot.

My experiementation and recollections aren't sufficient to verify or
dispute. Here usually LiveCD is to do an installation and at that point
I don't care if that computer sees the other computers or not.

Whereas one of my criteria for eventual hardware install is "Can you see
the bloody network? At least sometimes??"

Virtual Machine.... know it sometimes has problems even transferring
data to the host machine even with that scratchpad function set. (Ever
do SneakerNet from one machine to the same physical machine?!)

Yeah, have seen that. You can set it bidirectional all you want, all you
get is "Huh??"

Or ... "Host? What host??" tho that tends to go along with "Network?
What network??" (can still see internet, usually...) Host is actually a network drive for the VM, and if it can't see the network...

I have resorted to turning data into an ISO image, then loading that in
the VM's optical drive and reading it with WinRAR. I guess that's
sneakernet on the same machine!

> KM> But when I need to run something that the host OS doesn't like...
> KM> that's why I do VMs.
> You do a heck of a lot more experimenting than I do!
KM> A whole lot. <g>

But still easier and safer in the long run to experiment on a disposable machine, even if doesn't always work the same.

Quicker, yeah, especially when I was pawing through a hundred distros
trying to find one I could love, never mind tolerate. Let's hear it for LiveISOs!!

KM> Still preferably on real hardware, but am hunting for a VM that
KM> will work on Roadkill... my regular XP VM wouldn't even finish
KM> loading.

First thing that comes to mind is insufficent space. ...Checking
mine..... 32-bit XP, Motherboard tab has 'Enable I/O APIC' tic'd.
Everything else seems relatively normal.

Tried all manner of settings. XP can run in less than 100mb RAM (have regularly seen it use only 80mb with no external drivers installed,
dunno why it does that only when dual booting with ReactOS, but it
does), but apparently 512mb was too much for the host... I did get
Win2K to make a nice VM and that runs fine, with no issues, tho the
video component of the Guest Additions took a good 15 minutes to trawl
through every video driver known to man....

> KM> Did you see the crazy thing (I think it was) MJD did, with VMs
> KM> inside of VMs until it went all the way from Newest Windows to
> KM> Oldest Windows??
> No but sounds like an interesting project! Wonder if considered going
> back to MS-DOS?! ..Wonder how much storage it takes? Presume on a
> NVMe just for a reasonable speed to load the most current version, then
> the next from that, and the next from that one....
KM> LOL, you can do that, if you have enough RAM!
Can that be increased with Virtual RAM?!

LOL, we may have to try that. <g>

I do have Win8.1 in a VM on XP64 (Win10 is better, but threw up all over
the older VirtualBox) and that works fine... useful when I need to
access my hosting, which no longer speaks to any XP FTP client, with the
weird exception of commandline FTP, which works just fine. If you don't
mind OMG tedium to do anything.

KM> LOL, the old Costco here (they've moved) had that issue. The
KM> parking lot was atop what used to be a dump. Flat when first
KM> paved, but a few decades later it was up hill and down dale in
KM> every direction, tho with the largest dent toward the middle. And
KM> I mean a serious slope, not just a little dip!

Mine might have yours beat: I recall there were some sinkholes in the
stores (probably also the parking lot but I was too young to drive so
didn't pay attention) which were cordoned off. We didn't do it but I remember Dad commented on running with the xcart would be like a rollercoaster.

Egads. I don't think ours had yet progressed to sinkholes.... but it had
sunk about 10 feet in the middle. But tilted enough that it doesn't
collect much water.

KM> Came across some idiot on Youtube "demonstrating how unsafe XP is
KM> online" .... first thing he does is DISABLE THE FIREWALL, and
KM> naturally it immediately collected every circulating network worm
KM> or virus. Uh, stupid, do that with ANY OS and it'll have the same
KM> thing happen!!

Right! Take any current system, disable firewalls and security stuff,
and see how long it lasts!

Linux is actually worse for this than Windows, because most of what come slithering past no firewall are network worms, and linux is MORE
vulnerable to worms, and those don't require the OS to do anything.
Whereas most Windows malware needs an application to infect, but
generally doesn't have a good attack surface for network worms, having
not been designed from the gitgo to be an internet server.

I see most every linux distro has finally decided that average people do
NOT need to run the Apache webserver full time (yes, they all did that,
it was part of why in the early days performance was dismal) which
greatly reduces the attack surface. Apache was the main ingress route
for linux malware... and why does ANY desktop system need to run it??

...I'd be willing to bet my old DEC Rainbow
100 running DOS 2.11 would be trashed quickly. ...Well, might take a
while: as I recall 4Kbps modem.

LOL. There really wasn't much that could infect DOS over a modem,
because DOS didn't execute anything by default. I can just barely see
some sort of BIOS firmware worm managing it, but... why??

KM> And then he says, "I don't think the firewall is much good" ...
KM> <headdesk>

Once disabled it is no good!

This is true. In fact, I can think of few programs that run better when they're not running...

> I've noticed Mozilla appears to do a full re-install instead of an update
> and I think LibreOffice does too.
KM> A lot of these monolithic programs do that. However, IIRC Ubuntu
KM> is now all containerized (or at least I heard it was going to
KM> be), which means you always replace the whole thing.

Could be. I had thought they would stick with the compartmentalization
so if one thing breaks it doens't take down the whole thing.

<blink> Oh, confuzzlement. I mean it replaces the whole contents of the container.

The idea is good, but from what I hear it's not entirely ready for prime
time.

However, it does preclude Dependency Hell, which makes DLL Hell look
like a raw beginner.

KM> Generally, but there's another advantage of Rolling...
KM> If I have to reinstall with every version upgrade, I won't use
KM> it. That simple.

LIS in some other message I tend to do a full upgrade just because it
makes more sense: I usually have a new (updated) machine and so have
changed enough it makes more sense to discover all the hardware new than
to have to new OS look at the old list and make revisions from that. (I
know I'm using human-thinking method.) Plus over the decades with
MS-DOS, Windows and a few flavours of Linux I haven't had the greatest
luck in upgrading and having everything work properly the first day.

In the old days, Windows did not upgrade gracefully, in part because it
tried to preserve all your programs and settings, and those had DLL dependencies all over the place. They seem to have fixed that with the
8/10/11 chain; now you can't tell what's been upgraded and what was a
scratch install. However, it no longer goes to special lengths to
preserve anything, other than what's in your /User profile.

Linux version upgrades used to be a mess, rarely worked right, and a
clean install was indeed the only sane alternative. Some have figured
out that regular users do not like doing a resinstall every six months,
and have finally got it right.

Fedora does a full version every six months. My Fedora install was
originally v32. It has since been 34, 35, 36 (it did this one all by
itself, shortly after I did the manual ugrade to 35... I didn't even do
any updates, I just left it running and one morning there it was), 37,
38, 39, 40. You can skip two major versions, but if you need to upgrade
by more than two, you have to do it stepwise, not all at once.

Debian is still cranky about it, not sure how good Ubuntu is, being
really Debian that's eaten too many donuts.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> My experience is that LiveCD, Virtual Machine, and Real Hardware
KM> are not equivalent, in behavior or performance. Many a Linux
KM> LiveCD can see the other PCs on the network, but the same one
KM> INSTALLED cannot.
My experiementation and recollections aren't sufficient to verify or dispute. Here usually LiveCD is to do an installation and at that point
I don't care if that computer sees the other computers or not.

Whereas one of my criteria for eventual hardware install is "Can
you see the bloody network? At least sometimes??"

I wonder if the problem is with whichever device is assigning the IP?
If it was just one computer or one brand of interface card I could sort
of see the problem among multiple devices, but as you are having
problems across numerous computers maybe 'the other end' of the
connection. ..."Live works, installed does not" - guess
Troubleshooting Point #1 is are the two the same IP? And maybe that's
causing an issue with your firewall or some sort of protective point: it originally allowed the fingerprint for when on Live CD, now something
has been changed with the full installation.

Virtual Machine.... know it sometimes has problems even transferring
data to the host machine even with that scratchpad function set. (Ever
do SneakerNet from one machine to the same physical machine?!)

Yeah, have seen that. You can set it bidirectional all you want,
all you get is "Huh??"

Yup. I've also tried the untic trick: click so it looks 'off' but as
long as the computer thinks it's 'on'.

Or ... "Host? What host??" tho that tends to go along with
"Network? What network??" (can still see internet, usually...)
Host is actually a network drive for the VM, and if it can't see
the network...
I have resorted to turning data into an ISO image, then loading
that in the VM's optical drive and reading it with WinRAR. I
guess that's sneakernet on the same machine!

Much more data than I'm moving around!

> KM> But when I need to run something that the host OS doesn't like...
> KM> that's why I do VMs.
> You do a heck of a lot more experimenting than I do!
KM> A whole lot. <g>
But still easier and safer in the long run to experiment on a disposable machine, even if doesn't always work the same.

Quicker, yeah, especially when I was pawing through a hundred
distros trying to find one I could love, never mind tolerate.
Let's hear it for LiveISOs!!

<crowd roars!>

KM> Still preferably on real hardware, but am hunting for a VM that
KM> will work on Roadkill... my regular XP VM wouldn't even finish
KM> loading.
First thing that comes to mind is insufficent space. ...Checking
mine..... 32-bit XP, Motherboard tab has 'Enable I/O APIC' tic'd. Everything else seems relatively normal.

Tried all manner of settings. XP can run in less than 100mb RAM
(have regularly seen it use only 80mb with no external drivers
installed, dunno why it does that only when dual booting with
ReactOS, but it does), but apparently 512mb was too much for the
host... I did get Win2K to make a nice VM and that runs fine,
with no issues, tho the video component of the Guest Additions
took a good 15 minutes to trawl through every video driver known
to man....

At least it's trying! (Very trying!!) ...Your trials and
tribulations sort of remind me of the problems I had with RoseReader --
the OLMR. The BBS was a beta site. I could never get RoseReader to
quite work right: close but not quite. My configuration was right,
confirmed by by others. We finally copied my RR files, config, etc., to floppy/CD (I don't recall the details) and someone else tested -- of
course he never had any problems!

I do have Win8.1 in a VM on XP64 (Win10 is better, but threw up
all over the older VirtualBox) and that works fine... useful when
I need to access my hosting, which no longer speaks to any XP FTP
client, with the weird exception of commandline FTP, which works
just fine. If you don't mind OMG tedium to do anything.

I have to admire your system security! <g>

KM> LOL, the old Costco here (they've moved) had that issue. The
KM> parking lot was atop what used to be a dump. Flat when first
KM> paved, but a few decades later it was up hill and down dale in
KM> every direction, tho with the largest dent toward the middle. And
KM> I mean a serious slope, not just a little dip!
Mine might have yours beat: I recall there were some sinkholes in the
stores (probably also the parking lot but I was too young to drive so
didn't pay attention) which were cordoned off. We didn't do it but I remember Dad commented on running with the cart would be like a rollercoaster.

Egads. I don't think ours had yet progressed to sinkholes.... but
it had sunk about 10 feet in the middle. But tilted enough that
it doesn't collect much water.

I had moved out here by then but from what I recall Dad said at the time
the owners of the mall finally had to expand somewhat to move the store tenants into new sites and then fix the old sites, move, repeat.

KM> Came across some idiot on Youtube "demonstrating how unsafe XP is
KM> online" .... first thing he does is DISABLE THE FIREWALL, and
KM> naturally it immediately collected every circulating network worm
KM> or virus. Uh, stupid, do that with ANY OS and it'll have the same
KM> thing happen!!
Right! Take any current system, disable firewalls and security stuff,
and see how long it lasts!

Linux is actually worse for this than Windows, because most of
what come slithering past no firewall are network worms, and
linux is MORE vulnerable to worms, and those don't require the OS
to do anything. Whereas most Windows malware needs an application
to infect, but generally doesn't have a good attack surface for
network worms, having not been designed from the gitgo to be an
internet server.
I see most every linux distro has finally decided that average
people do NOT need to run the Apache webserver full time (yes,
they all did that, it was part of why in the early days
performance was dismal) which greatly reduces the attack surface.
Apache was the main ingress route for linux malware... and why
does ANY desktop system need to run it??

No wonder your stuff doesn't work: you don't have your Indian guide! <g>

...Just checked: apache and apache2 not installed here, or at least on
this system. And yes, the more stuff added and running the slower the
system is going to be: can only do so much at a time.

...I'd be willing to bet my old DEC Rainbow
100 running DOS 2.11 would be trashed quickly. ...Well, might take a
while: as I recall 4Kbps modem.

LOL. There really wasn't much that could infect DOS over a modem,
because DOS didn't execute anything by default. I can just barely
see some sort of BIOS firmware worm managing it, but... why??

Because it's challenge.

KM> And then he says, "I don't think the firewall is much good" ...
KM> <headdesk>
Once disabled it is no good!

This is true. In fact, I can think of few programs that run
better when they're not running...

If the 'guard' at the firewall doesn't have to check each packet's
identity things would be faster. Not only clear sailing in and out of
the computer the CPU doesn't have to do the thinking of the goard.

KM> Generally, but there's another advantage of Rolling...
KM> If I have to reinstall with every version upgrade, I won't use
KM> it. That simple.
LIS in some other message I tend to do a full upgrade just because it
makes more sense: I usually have a new (updated) machine and so have
changed enough it makes more sense to discover all the hardware new than
to have to new OS look at the old list and make revisions from that. (I know I'm using human-thinking method.) Plus over the decades with
MS-DOS, Windows and a few flavours of Linux I haven't had the greatest
luck in upgrading and having everything work properly the first day.

In the old days, Windows did not upgrade gracefully, in part
because it tried to preserve all your programs and settings, and
those had DLL dependencies all over the place. They seem to have
fixed that with the 8/10/11 chain; now you can't tell what's been
upgraded and what was a scratch install. However, it no longer
goes to special lengths to preserve anything, other than what's
in your /User profile.

The probably eliminated a ton of variables and so testing time to just
stick with that handful of options.

Linux version upgrades used to be a mess, rarely worked right,
and a clean install was indeed the only sane alternative. Some
have figured out that regular users do not like doing a
resinstall every six months, and have finally got it right.

I'd like the latest and greatest but from past experience have found
"poop occurs" and stuff doesn't work, therefore the 'rolling hardware'
as a safety net. And doing upgrades takes time: while I'm babysitting
the machine I can't do too much else.

Fedora does a full version every six months. My Fedora install
was originally v32. It has since been 34, 35, 36 (it did this one
all by itself, shortly after I did the manual ugrade to 35... I
didn't even do any updates, I just left it running and one
morning there it was), 37, 38, 39, 40. You can skip two major
versions, but if you need to upgrade by more than two, you have
to do it stepwise, not all at once.
Debian is still cranky about it, not sure how good Ubuntu is,
being really Debian that's eaten too many donuts.

I've been able to skip one LTS upgrade most of the time. I'm recalling
one machine here I did have to do an 'intermediate install' but may have
been a three LTS skip.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... We have biscuits and Triscuits, where are monscuit and quadriscuits?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

> My experiementation and recollections aren't sufficient to verify or
> dispute. Here usually LiveCD is to do an installation and at that point
> I don't care if that computer sees the other computers or not.
KM> Whereas one of my criteria for eventual hardware install is "Can
KM> you see the bloody network? At least sometimes??"

I wonder if the problem is with whichever device is assigning the IP?

That would be the router, and it does just fine.

See this?

C:\WINDOWS>ping 192.168.0.5

Pinging 192.168.0.5 with 32 bytes of data:

Reply from 192.168.0.5: bytes=32 time<1ms TTL=64
Reply from 192.168.0.5: bytes=32 time<1ms TTL=64
Reply from 192.168.0.5: bytes=32 time<1ms TTL=64
Reply from 192.168.0.5: bytes=32 time<1ms TTL=64

Ping statistics for 192.168.0.5:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 0ms, Average = 0ms
===

This is Windows pinging the PCLinuxOS box.

Win10 and 11 can connect to said box. XP cannot. But it can ping it fine!
It's something in the network protocols that's different.
Also, something goofy in the Samba server.

It also affects FTP to an outside site, as of a few months ago XP no
longer works, but a lot longer for linux desktops. (It used to work
better, or more often.)

Meanwhile, the PCLOS box can only see one drive of the XP64 box. Sometimes.

It cannot see the Fedora box. Which doesn't see any Windows box.

If it was just one computer or one brand of interface card I could sort
of see the problem among multiple devices, but as you are having
problems across numerous computers maybe 'the other end' of the
connection. ..."Live works, installed does not" - guss

Across multiple different linux systems, too. They don't like each
other, they don't like XP, but they do let Win10 snoop.

Linux demands a login from XP, then won't accept any credentials.
Doesn't even ask who-are-you to Win10.

Troubleshooting Point #1 is are the two the same IP? And maybe that's causing an issue with your firewall or some sort of protective point: it originally allowed the fingerprint for when on Live CD, now something
has been changed with the full installation.

Nope, wouldn't be anything like that. It's somewhere in the network
protocols that do not like each other much. I'd guess a live CD loads a generic version of Samba (or at least generic settings), which proceeds
to handle things properly, but installed version is not so generic and therefore does not work. And may not even install Samba. (I'm lookin' at
you, PCLOS...)

> Virtual Machine.... know it sometimes has problems even transferring
> data to the host machine even with that scratchpad function set. (Ever
> do SneakerNet from one machine to the same physical machine?!)
KM> Yeah, have seen that. You can set it bidirectional all you want,
KM> all you get is "Huh??"

Yup. I've also tried the untic trick: click so it looks 'off' but as
long as the computer thinks it's 'on'.

Yeah, sometimes there's a flipped flag somewhere.

KM> Or ... "Host? What host??" tho that tends to go along with
KM> "Network? What network??" (can still see internet, usually...)
KM> Host is actually a network drive for the VM, and if it can't see
KM> the network...
KM> I have resorted to turning data into an ISO image, then loading
KM> that in the VM's optical drive and reading it with WinRAR. I
KM> guess that's sneakernet on the same machine!

Much more data than I'm moving around!

Oh, an ISO can be very small, just make an image file with some ISO creator.

KM> Let's hear it for LiveISOs!!

<crowd roars!>

I can't hear you!!

KM> host... I did get Win2K to make a nice VM and that runs fine,
KM> with no issues, tho the video component of the Guest Additions
KM> took a good 15 minutes to trawl through every video driver known
KM> to man....

At least it's trying! (Very trying!!) ...Your trials and
tribulations sort of remind me of the problems I had with RoseReader --
the OLMR. The BBS was a beta site. I could never get RoseReader to
quite work right: close but not quite. My configuration was right,
confirmed by by others. We finally copied my RR files, config, etc., to floppy/CD (I don't recall the details) and someone else tested -- of
course he never had any problems!

I remember that Rose Reader existed. <g>

But yeah, sometimes it's something very obscure and not even related to
the software in question. Some TSR with a wrong address byte, that sort
of thing.

KM> I do have Win8.1 in a VM on XP64 (Win10 is better, but threw up
KM> all over the older VirtualBox) and that works fine... useful when
KM> I need to access my hosting, which no longer speaks to any XP FTP
KM> client, with the weird exception of commandline FTP, which works
KM> just fine. If you don't mind OMG tedium to do anything.

I have to admire your system security! <g>

Perfectly secure. <g>

KM> I see most every linux distro has finally decided that average
KM> people do NOT need to run the Apache webserver full time (yes,
KM> they all did that, it was part of why in the early days
KM> performance was dismal) which greatly reduces the attack surface.
KM> Apache was the main ingress route for linux malware... and why
KM> does ANY desktop system need to run it??

No wonder your stuff doesn't work: you don't have your Indian guide! <g>

I think mine is the one with his ear pressed to the road.

Cowboy: Who just went by?
Chief Ear-to-Road: Two wagons, six cowboys, and a dozen steers.
Cowboy: Wow, you got all that from listening to the ground?
Chief Ear-to-Road: No, from when they ran me over!!

..Just checked: apache and apache2 not installed here, or at least on
this system. And yes, the more stuff added and running the slower the
system is going to be: can only do so much at a time.

Yeah, as I noted, it stopped a few years back. But it was a point of
utter stupidity for about 20 years, at least with the major distros.

> ...I'd be willing to bet my old DEC Rainbow
> 100 running DOS 2.11 would be trashed quickly. ...Well, might take a
> while: as I recall 4Kbps modem.
KM> LOL. There really wasn't much that could infect DOS over a modem,
KM> because DOS didn't execute anything by default. I can just barely
KM> see some sort of BIOS firmware worm managing it, but... why??

Because it's challenge.

Because it's probably impossible. <g>

> KM> And then he says, "I don't think the firewall is much good" ...
> KM> <headdesk>
> Once disabled it is no good!
KM> This is true. In fact, I can think of few programs that run
KM> better when they're not running...

If the 'guard' at the firewall doesn't have to check each packet's
identity things would be faster. Not only clear sailing in and out of
the computer the CPU doesn't have to do the thinking of the goard.

On a very slow sPC you could see ZoneAlarm lagging the system. But by
the middle-Pentium era, we have enough horsepower that it's not a problem.

KM> In the old days, Windows did not upgrade gracefully, in part
KM> because it tried to preserve all your programs and settings, and
KM> those had DLL dependencies all over the place. They seem to have
KM> fixed that with the 8/10/11 chain; now you can't tell what's been
KM> upgraded and what was a scratch install. However, it no longer
KM> goes to special lengths to preserve anything, other than what's
KM> in your /User profile.

The probably eliminated a ton of variables and so testing time to just
stick with that handful of options.

I think the real issue is that Windows has become so complex (how on
earth does Win11 need 25GB on disk and 4GB of RAM just to admire its
navel??) that there is no supporting any variables anymore.

KM> Linux version upgrades used to be a mess, rarely worked right,
KM> and a clean install was indeed the only sane alternative. Some
KM> have figured out that regular users do not like doing a
KM> resinstall every six months, and have finally got it right.

I'd like the latest and greatest but from past experience have found
"poop occurs" and stuff doesn't work, therefore the 'rolling hardware'
as a safety net. And doing upgrades takes time: while I'm babysitting
the machine I can't do too much else.

Yeah, good policy.

.. We have biscuits and Triscuits, where are monscuit and quadriscuits?

This is a good question!
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Win10 and 11 can connect to said box. XP cannot. But it can ping it fine! It's something in the network protocols that's different.
Also, something goofy in the Samba server.

That may be it right there. Linux will apply updates to samba to
(allegedly) quash vulnerabilities. I can rememeber when they applied one and my Warp 4 box could no longer access the samba shares on the linux server. Apparently OS/2 relied on the vulnerability being there in order to work.

Since XP's samba has probably not been updated in a while it may have the
same issue.

It also affects FTP to an outside site, as of a few months ago XP no
longer works, but a lot longer for linux desktops. (It used to work
better, or more often.)

To get around this (with an old Win98 machine I have that does not
otherwise have access to the Internet), I installed an ftp server on the
linux machine for the occassional times I need to transfer something
between boxes.

Across multiple different linux systems, too. They don't like each
other, they don't like XP, but they do let Win10 snoop.

I have never had an issue with linux systems, that are all running NFS,
seeing each other. Then again, I have not tried it with the variation of distros that you seem to be.

Mike


* SLMR 2.1a * Southern DOS: Y'all reckon? (Yep/Nope)
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

Hi Ky!

> My experiementation and recollections aren't sufficient to verify or
> dispute. Here usually LiveCD is to do an installation and at that point
> I don't care if that computer sees the other computers or not.
KM> Whereas one of my criteria for eventual hardware install is "Can
KM> you see the bloody network? At least sometimes??"
I wonder if the problem is with whichever device is assigning the IP?

That would be the router, and it does just fine.
C:\WINDOWS>ping 192.168.0.5
Reply from 192.168.0.5: bytes=32 time<1ms TTL=64
This is Windows pinging the PCLinuxOS box.

Obvious a nice fast time.

...I'm trying to recall my issue a couple months ago; sort of similar as
an established box stopped connecting. Ping worked fine but Remmina,
ssh, etc., did not. I think I had to re-establish the ssh fingerprint
thing but not sure.

If it was just one computer or one brand of interface card I could sort
of see the problem among multiple devices, but as you are having
problems across numerous computers maybe 'the other end' of the
connection. ..."Live works, installed does not" -

Across multiple different linux systems, too. They don't like
each other, they don't like XP, but they do let Win10 snoop.
Linux demands a login from XP, then won't accept any credentials.
Doesn't even ask who-are-you to Win10.

Here's a quirk I ran in to: I think up to a couple of Ubuntu versions
ago I have to go into Settings to reset/re-allow Sharing. Turn off then
click all the options back on. In "Remote Desktop" be sure to tic
'Enable Legacy VNC Protocol' and then tic the three vertical dots at the
right, and 'follow instructions'. The other 'trick' is the legacy
password can only be eight characters long and here Remmina has to be
set for use of a legacy password and not encrypted.

Be sure to also turn on Wired (or Wireless) connection at the bottom of
the menu pop-up!!

Well, that sort of mixed two problems together but might give you a idea
of what to try. I've got a document or two I could e-mail with my
set-up notes; if you want let me know.

Troubleshooting Point #1 is are the two the same IP? And maybe that's causing an issue with your firewall or some sort of protective point: it originally allowed the fingerprint for when on Live CD, now something
has been changed with the full installation.

Nope, wouldn't be anything like that. It's somewhere in the
network protocols that do not like each other much. I'd guess a
live CD loads a generic version of Samba (or at least generic
settings), which proceeds to handle things properly, but
installed version is not so generic and therefore does not work.
And may not even install Samba. (I'm lookin' at you, PCLOS...)

Installed version probably get picky as figure that is going to be a
system with information on it whereas the trial version is temporary and
more for testing.

> Virtual Machine.... know it sometimes has problems even transferring
> data to the host machine even with that scratchpad function set. (Ever
> do SneakerNet from one machine to the same physical machine?!)
KM> Yeah, have seen that. You can set it bidirectional all you want,
KM> all you get is "Huh??"
Yup. I've also tried the untic trick: click so it looks 'off' but as
long as the computer thinks it's 'on'.

Yeah, sometimes there's a flipped flag somewhere.

Figured you tried, just go back to basics troubleshooting. (Reboot
fixes a lot of problems -- except sometimes for Sharing causes
problems!)

KM> Or ... "Host? What host??" tho that tends to go along with
KM> "Network? What network??" (can still see internet, usually...)
KM> Host is actually a network drive for the VM, and if it can't see
KM> the network...
KM> I have resorted to turning data into an ISO image, then loading
KM> that in the VM's optical drive and reading it with WinRAR. I
KM> guess that's sneakernet on the same machine!
Much more data than I'm moving around!

Oh, an ISO can be very small, just make an image file with some
ISO creator.

True, just sounds impressive. Or maybe just to me as I've used ISOs
only to install OSs.

KM> Let's hear it for LiveISOs!!
<crowd roars!>

I can't hear you!!

Handing over an antique ear trumpet. ...Running away because realized
can also be used as a weapon!

At least it's trying! (Very trying!!) ...Your trials and
tribulations sort of remind me of the problems I had with RoseReader --
the OLMR. The BBS was a beta site. I could never get RoseReader to
quite work right: close but not quite. My configuration was right, confirmed by by others. We finally copied my RR files, config, etc., to floppy/CD (I don't recall the details) and someone else tested -- of
course he never had any problems!

I remember that Rose Reader existed. <g>
But yeah, sometimes it's something very obscure and not even
related to the software in question. Some TSR with a wrong
address byte, that sort of thing.

Yes, it bacame obvious it was something in the way my system interacted
with RoseReader. The 'good news' was I wasn't the only one with
problems: some others' never worked, a few like me 'worked except for',
and others always worked. We never did figure a commonality (like Intel worked, AMD did not).

KM> I do have Win8.1 in a VM on XP64 (Win10 is better, but threw up
KM> all over the older VirtualBox) and that works fine... useful when
KM> I need to access my hosting, which no longer speaks to any XP FTP
KM> client, with the weird exception of commandline FTP, which works
KM> just fine. If you don't mind OMG tedium to do anything.
I have to admire your system security! <g>

Perfectly secure. <g>

It's so secure you can't even access!

KM> I see most every linux distro has finally decided that average
KM> people do NOT need to run the Apache webserver full time (yes,
KM> they all did that, it was part of why in the early days
KM> performance was dismal) which greatly reduces the attack surface.
KM> Apache was the main ingress route for linux malware... and why
KM> does ANY desktop system need to run it??
No wonder your stuff doesn't work: you don't have your Indian guide! <g>

I think mine is the one with his ear pressed to the road.

Well that's an alternate way to describe fallen-down drunk! <g>

Cowboy: Who just went by?
Chief Ear-to-Road: Two wagons, six cowboys, and a dozen steers.
Cowboy: Wow, you got all that from listening to the ground?
Chief Ear-to-Road: No, from when they ran me over!!

..Just checked: apache and apache2 not installed here, or at least on
this system. And yes, the more stuff added and running the slower the system is going to be: can only do so much at a time.

Yeah, as I noted, it stopped a few years back. But it was a point
of utter stupidity for about 20 years, at least with the major
distros.

They (the powers that be) probably figured it was necessary, or the
developers needed it for their testing then would forget to
remove/de-activate it in the distribution. ...Yeah, twenty years seems
like a really long time to keep forgetting.

> ...I'd be willing to bet my old DEC Rainbow
> 100 running DOS 2.11 would be trashed quickly. ...Well, might take a
> while: as I recall 4Kbps modem.
KM> LOL. There really wasn't much that could infect DOS over a modem,
KM> because DOS didn't execute anything by default. I can just barely
KM> see some sort of BIOS firmware worm managing it, but... why??
Because it's challenge.

Because it's probably impossible. <g>

Now THAT'S a challenge!! <g>

> KM> And then he says, "I don't think the firewall is much good" ...
> KM> <headdesk>
> Once disabled it is no good!
KM> This is true. In fact, I can think of few programs that run
KM> better when they're not running...
If the 'guard' at the firewall doesn't have to check each packet's
identity things would be faster. Not only clear sailing in and out of
the computer the CPU doesn't have to do the thinking of the goard.

On a very slow sPC you could see ZoneAlarm lagging the system.
But by the middle-Pentium era, we have enough horsepower that
it's not a problem.

Yeah, I sort of miss the days when one could read any error messages
while the computer was booting.

KM> In the old days, Windows did not upgrade gracefully, in part
KM> because it tried to preserve all your programs and settings, and
KM> those had DLL dependencies all over the place. They seem to have
KM> fixed that with the 8/10/11 chain; now you can't tell what's been
KM> upgraded and what was a scratch install. However, it no longer
KM> goes to special lengths to preserve anything, other than what's
KM> in your /User profile.
The probably eliminated a ton of variables and so testing time to just
stick with that handful of options.

I think the real issue is that Windows has become so complex (how
on earth does Win11 need 25GB on disk and 4GB of RAM just to
admire its navel??) that there is no supporting any variables
anymore.

That's probably closer to the real reason. ...Things get more and more complicated when trying to create something to run on every combination.

KM> Linux version upgrades used to be a mess, rarely worked right,
KM> and a clean install was indeed the only sane alternative. Some
KM> have figured out that regular users do not like doing a
KM> resinstall every six months, and have finally got it right.
I'd like the latest and greatest but from past experience have found
"poop occurs" and stuff doesn't work, therefore the 'rolling hardware'
as a safety net. And doing upgrades takes time: while I'm babysitting
the machine I can't do too much else.

Yeah, good policy.

Plus takes a lot of stress (and panic) out of things: *it* doesn't work,
but everything else does so no big deal. Use the working computer to
find out how to fix the non-working computer.

.. We have biscuits and Triscuits, where are monscuit and quadriscuits?

This is a good question!

And then I wonder how they'd be: if a biscuit was as crunchy as a
Triscuit it would be bad eating. Would a monscuit be mushy? A
quadriscuit hard that hardtack?


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Went shopping for cherries and mic stands: bought a bing bought a boom.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

MIKE POWELL wrote:

Win10 and 11 can connect to said box. XP cannot. But it can ping it fine!
It's something in the network protocols that's different.
Also, something goofy in the Samba server.

That may be it right there. Linux will apply updates to samba to
(allegedly) quash vulnerabilities. I can rememeber when they applied one and my Warp 4 box could no longer access the samba shares on the linux server. Apparently OS/2 relied on the vulnerability being there in order to work.

That is likely so. It used to be better.

Since XP's samba has probably not been updated in a while it may have the same issue.

Probably. :/

It also affects FTP to an outside site, as of a few months ago XP no
longer works, but a lot longer for linux desktops. (It used to work
better, or more often.)

To get around this (with an old Win98 machine I have that does not
otherwise have access to the Internet), I installed an ftp server on the linux machine for the occassional times I need to transfer something
between boxes.

Clever trick. What FTP server did you use?

Across multiple different linux systems, too. They don't like each
other, they don't like XP, but they do let Win10 snoop.

I have never had an issue with linux systems, that are all running NFS, seeing each other. Then again, I have not tried it with the variation of distros that you seem to be.

Yeah, there's the one I like best, then the several that I can make do
with if I have to, because PCLinuxOS is a one man band, Tex's health
isn't good, and dunno if there's anyone going to continue the distro
when he hangs it up. So have been trying to come up with another I like
as well... so far the answer is "not really."

But I keep Fedora up to date because it's been useful (tho the
performance is really suck by comparison... 5 seconds to the desktop vs
2 minutes), tho I'd not like to be using it for everyday. And everything
else I like less.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

To get around this (with an old Win98 machine I have that does not otherwise have access to the Internet), I installed an ftp server on the linux machine for the occassional times I need to transfer something between boxes.

Clever trick. What FTP server did you use?

I installed vsftpd. It was pretty simple. The Win98 ftp client works fine with it (knock-on-wood!). I also made sure that the ftp port on that
machien was not open to the router (and the outside world).

Yeah, there's the one I like best, then the several that I can make do
with if I have to, because PCLinuxOS is a one man band, Tex's health
isn't good, and dunno if there's anyone going to continue the distro
when he hangs it up. So have been trying to come up with another I like
as well... so far the answer is "not really."

That is a good reason to be trying out other distros, though. I think you
told me before, but which one is PCLinuxOS based on?

Mike

* SLMR 2.1a * Do not look in laser with remaining eye.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

MIKE POWELL wrote:

To get around this (with an old Win98 machine I have that does not
otherwise have access to the Internet), I installed an ftp server on the >>> linux machine for the occassional times I need to transfer something
between boxes.

Clever trick. What FTP server did you use?

I installed vsftpd. It was pretty simple. The Win98 ftp client works fine with it (knock-on-wood!). I also made sure that the ftp port on that
machien was not open to the router (and the outside world).

https://security.appspot.com/vsftpd.html#download

I'll have to try it, thanks.

Yeah, there's the one I like best, then the several that I can make do
with if I have to, because PCLinuxOS is a one man band, Tex's health
isn't good, and dunno if there's anyone going to continue the distro
when he hangs it up. So have been trying to come up with another I like
as well... so far the answer is "not really."

That is a good reason to be trying out other distros, though. I think you

Yeah. Next one I become addicted to, I'd like to outlive me.

Fedora is tolerable, but I don't LIKE it the way I do PCLOS. And the performance is nowhere near as good. Boot to desktop:
PCLOS: 5-10 seconds on a modern machine, 30 seconds on a 16 year old laptop. Fedora: 2 minutes on a modern machine, I wouldn't even TRY it on the laptop.

told me before, but which one is PCLinuxOS based on?

Mandrake, a LONG time ago, before Mandrake itself died. It's become its
own thing, tho the nearest current relation is OpenMandriva (but OM is
not nearly as clean and bug-free).
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Yeah. Next one I become addicted to, I'd like to outlive me.

Yes, that is my hope also. ;)

Fedora is tolerable, but I don't LIKE it the way I do PCLOS. And the performance is nowhere near as good. Boot to desktop:
PCLOS: 5-10 seconds on a modern machine, 30 seconds on a 16 year old laptop. Fedora: 2 minutes on a modern machine, I wouldn't even TRY it on the laptop.

I would not bother trying that on a old laptop, either!

told me before, but which one is PCLinuxOS based on?

Mandrake, a LONG time ago, before Mandrake itself died. It's become its
own thing, tho the nearest current relation is OpenMandriva (but OM is
not nearly as clean and bug-free).

Have you ever looked at Mageia? I have never tried it but it is supposed
to be a fork of Mandriva, forked about the same time as OM when Mandriva stopped supporting an open source version.

I was sorry to see Mandrake go. It was one of the first distros I tried.
The installer was slick and appeared to work perfectly with my video card. However, the installer failed to set the actual OS up like it did during
the installer video tests and left the system unusable. :( I had hoped
one day maybe to try it again on more modern hardware but it morphed into Mandriva shortly thereafter.

Mike


* SLMR 2.1a * Oxymoron: Hillary Rodham Clinton and open-door meetings.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

KY MOFFET wrote to MIKE POWELL <=-

Hi Ky!

But I keep Fedora up to date because it's been useful (tho the
performance is really suck by comparison... 5 seconds to the
desktop vs 2 minutes), tho I'd not like to be using it for
everyday. And everything else I like less.

Just out of curiosity is the boot hardware the same? Here I shocked
myself a while back by replacing the hard drive with a SSD in a couple
of old computers and the boot time went from a couple of minutes to
maybe 15 seconds.

...There is also an option to wait for network connections which could
delay booting. And of course the disk check process.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... How do dragons blow out candles?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

MIKE POWELL wrote:

Yeah. Next one I become addicted to, I'd like to outlive me.

Yes, that is my hope also. ;)

LOL. Old Farts Farting About Linux. <g>

Acronymy :D

Fedora is tolerable, but I don't LIKE it the way I do PCLOS. And the
performance is nowhere near as good. Boot to desktop:
PCLOS: 5-10 seconds on a modern machine, 30 seconds on a 16 year old laptop. >> Fedora: 2 minutes on a modern machine, I wouldn't even TRY it on the laptop.

I would not bother trying that on a old laptop, either!

I'm not sure it would even boot. It has perhaps the oldest laptop CPU
that was x64, and only 2GB RAM. Even PCLOS isn't snappy, but it's not sluggish. Mint might run, Puppy would, and Q4OS (which is quite slick on
old hardware). I wouldn't bet on anything else.

told me before, but which one is PCLinuxOS based on?

Mandrake, a LONG time ago, before Mandrake itself died. It's become its
own thing, tho the nearest current relation is OpenMandriva (but OM is
not nearly as clean and bug-free).

Have you ever looked at Mageia? I have never tried it but it is supposed
to be a fork of Mandriva, forked about the same time as OM when Mandriva stopped supporting an open source version.

Yeah, I always check out Mageia when a new version comes out, but
it's... very very slow, and not stable, and it gets more sluggish with
every version. It should be nice, they try hard, but apparently there is
zero clue about optimization.

OpenMandriva isn't sluggish like that, but it's not very "finished".

Kubuntu always reminds me that Ubuntu only really cares about the Gnome desktop. It just feels like a poor relation, plus has Ubuntu's... heft.

I was sorry to see Mandrake go. It was one of the first distros I tried.

Yeah, I was very sad about that.

I'd had RedHat 6 long before (it stank in every way, was unstable, and
was a horrible slug, and when it forgot its own password, I nuked it),
but Mandrake 7.2 with some early KDE desktop was the first one I got to
run well and showed promise of being everyday-usable. Wasn't there yet,
but could see it coming, and I actually *liked* it, whereas I'd hated
RedHat and whatever else I'd looked at. And funny thing, when I went
looking for a linux I could love, all the ones I liked best were
Mandrake descendants.

The installer was slick and appeared to work perfectly with my video card. However, the installer failed to set the actual OS up like it did during
the installer video tests and left the system unusable. :( I had hoped
one day maybe to try it again on more modern hardware but it morphed into Mandriva shortly thereafter.

Mandriva, or now OpenMandriva, is the nearest direct descendant, but
PCLOS seems to have better preserved the general goodness. Mandrake
really was geared toward being usable by ordinary people with no linux knowledge or experience.

However, Tex just switched to some other installer, and it's not near as
easy as Drak -- with Drak it was two clicks to approve the defaults,
five minutes, done. Could hand it to any idiot and they'd have a
successful install. The new installer, not so, you have to pay attention
and there's more stuff to select up front, and it no longer
automagically makes the sensible 3 partitions (OS, swap, user).
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

KY MOFFET wrote to MIKE POWELL <=-

Hi Ky!

KM> But I keep Fedora up to date because it's been useful (tho the
KM> performance is really suck by comparison... 5 seconds to the
KM> desktop vs 2 minutes), tho I'd not like to be using it for
KM> everyday. And everything else I like less.

Just out of curiosity is the boot hardware the same? Here I shocked
myself a while back by replacing the hard drive with a SSD in a couple
of old computers and the boot time went from a couple of minutes to
maybe 15 seconds.

Yeah, especially on an older system that makes a huge difference. Less
so on a faster one ... I've had both spinning rust and SSD in the Dells
and have found when you get up into the i7 range, the system itself is
fast enough that it shrinks the data transfer difference, having
processed it faster. Disk difference shouldn't be more than about 2x-4x slower, not 10x-20x slower.

Anyway, Fedora is on a slightly faster system with twice the RAM, but on spinning rust (however a very fast laptop drive). But it's done all the
disk read by the time it shows a desktop ... the desktop just isn't
usable for a while, because it's still thinking.

..There is also an option to wait for network connections which could
delay booting. And of course the disk check process.

Nope, this is just being slow. Plus lagged down by Discover insisting on
going and checking for updates before you have a usable desktop. Need to
find where to turn that off. Discovered you CAN turn that off in Win11,
and did so, take that Microsoft.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> But I keep Fedora up to date because it's been useful (tho the
KM> performance is really suck by comparison... 5 seconds to the
KM> desktop vs 2 minutes), tho I'd not like to be using it for
KM> everyday. And everything else I like less.
Just out of curiosity is the boot hardware the same? Here I shocked
myself a while back by replacing the hard drive with a SSD in a couple
of old computers and the boot time went from a couple of minutes to
maybe 15 seconds.

Yeah, especially on an older system that makes a huge difference.
Less so on a faster one ... I've had both spinning rust and SSD
in the Dells and have found when you get up into the i7 range,
the system itself is fast enough that it shrinks the data
transfer difference, having processed it faster. Disk difference
shouldn't be more than about 2x-4x slower, not 10x-20x slower.

OK. Countering the SSD being much faster than a hard drive the older processors are slower: less speed and less cores. Both of those factors
would slow booting. I haven't done any experimentation on current fast systems as build 'em with the SSD for the OS and the hard drive for the
data. Quite sure the hard drive is providing a bit of a bottleneck but
I'll accept that as a trade off for safety.

Anyway, Fedora is on a slightly faster system with twice the RAM,
but on spinning rust (however a very fast laptop drive). But it's
done all the disk read by the time it shows a desktop ... the
desktop just isn't usable for a while, because it's still
thinking.

Possibly listing all the files for later quick access. With my main
Ubuntu system I've noticed the Desktop comes up quickly and but when I
ask for something there is a bit of first time hesistation - how long
depends on what it is. Firefox and Thunderbird a few seconds; Nautilus
can pause for 10-15 seconds while it reminds itself where everything is.
As Nautilus again and instantaneous response.

..There is also an option to wait for network connections which could
delay booting. And of course the disk check process.

Nope, this is just being slow. Plus lagged down by Discover
insisting on going and checking for updates before you have a
usable desktop. Need to find where to turn that off. Discovered
you CAN turn that off in Win11, and did so, take that Microsoft.

Yes, that's sort of an unnecessary delay: updates are good (generally)
but do when convenient for the user. At least Ubuntu asks to install
updates -- I've got one pending this morning because I have to reboot to complete. ...When done updating will ask to reboot now or later, just
seems to me if there's that much change it requires a reboot the system
is a little unstable: running the old but some new is waiting to rush in
to take it's place.



¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Be with people who want more FOR you, not more FROM you.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> But I keep Fedora up to date because it's been useful (tho the
KM> performance is really suck by comparison... 5 seconds to the
KM> desktop vs 2 minutes), tho I'd not like to be using it for
KM> everyday. And everything else I like less.
Just out of curiosity is the boot hardware the same? Here I shocked
myself a while back by replacing the hard drive with a SSD in a couple
of old computers and the boot time went from a couple of minutes to
maybe 15 seconds.

Yeah, especially on an older system that makes a huge difference.
Less so on a faster one ... I've had both spinning rust and SSD
in the Dells and have found when you get up into the i7 range,
the system itself is fast enough that it shrinks the data
transfer difference, having processed it faster. Disk difference
shouldn't be more than about 2x-4x slower, not 10x-20x slower.

Well 'just for fun' I counted the number of seconds from this morning's reboot-required because of a major OS update to when I trupe in my
password: 36 seconds.

Longer than you're indicating but I'm probably counting stuff you're
not: I started counting when the system started it's reboot process, so
at shutdown: has to do the shutdown processes. It then also wanted to
do a disk check, so several seconds there. Then whatever changes and
checks it makes for the upgrade.

And I don't think I did a follow-up on an old problem here, mainly
because wanted to test what I finally found as the probable solution.
For some time (months, maybe years) this system was 'unstable' with
regards to USB stuff. USB2, USB3, where (port, hub's ports), didn't
matter: possibility of a lock up. Error message in system logs
indicated something like 'device 2, port 5' which I couldn't figure out
what it was. Slow forward <g> and through process of elimination think figured out what was being gestured to: a USB3 hub on the left of my
desk. The hub and connected devices seem to be fine as disconnected and
test and reconnect and test, and I got the same error when using a
different hub (different brand, fewer ports, but possibly the same
chip).

What I'm thinking (and haven't tested yet) is the USB3 cable is too
long. Found waaay after the fact 3m/10' is the limit. No idea what I
am using but guessing to USB3 extensions, and those would bring it over
the limit.

USB3 Powered Extenders are around $80 -- not exactly making my credit
card happy. Half-thinking to try using a spare USB3 hub (powered) at
the halfway point. Right now the physical placement is a problem. Also
the Round-ToIt. <g>


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Scientists have grown vocal cords in Petri dish. Results speak for self. --- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

I have two debian bookworm systems. One system is running the RDP daemon to accept remote desktop requests. The other is using Remmina as a client.

If I log into the host system locally, it correctly uses the IceWM windows manager. However, when I log on via Remmina from the other system, the connection uses the lxqt manager.

How do I set the host up so that it uses the proper windows manager for the user instead of defaulting to lxqt? I tried googling the issue but the words and phrases I used seemed to bring up links mostly for Windows issues and/or connection issues. With both machines being debian and the connections being fine, they were not of much help. ;)

Thanks!
#
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

Hi Mike!

I have two debian bookworm systems. One system is running the
RDP daemon to accept remote desktop requests. The other is using
Remmina as a client.

If I log into the host system locally, it correctly uses the
IceWM windows manager. However, when I log on via Remmina from
the other system, the connection uses the lxqt manager.

How do I set the host up so that it uses the proper windows
manager for the user instead of defaulting to lxqt? I tried
googling the issue but the words and phrases I used seemed to
bring up links mostly for Windows issues and/or connection
issues. With both machines being debian and the connections
being fine, they were not of much help. ;)

I'm not too sure how much this will help you to figure out the problem
but may give a few hints.

It seems Remmina and Wayland don't get along all that well. Remmina
plays nice with X11, but Bookworm uses Wayland. There is a way to
switch to Wayland back to X11 -- should be easy enough to find else with
a bit of digging I could find it in my notes. I elected not to switch
back as thinking if I revert back then ll run into problems because
eventually X11 will be depreciated and I'm back to the current problem.

IceWM seems to use (or plays nice with) Wayland. LXQT seems to be based
on X11: I was looking around for hints and one of the hits was "LXQt
Desktop Now '100%' Ready For Wayland", which to me implies it currently
uses X11, which might be why Remmina uses it.


So back to your issue, based on my playing-arounds between my desktop
using Ubuntu 22.04 and using Remmina to connect to the various other
computers (desktops using Ubuntua 22.04 and earlier) plus Raspberry Pi's
using Bookwork, Bullseye (I don't think prior), Remmina generally
connects nicely to anyone except if running Bookworm. Wayland is the
issue per the Raspberry Pi forums, probably have also seen in Ask Ubuntu
and Spiceworks, just skimming over as for me "known issue, not an easy
fix, others smarter than me are having the same problem...."

(This all sounds so much better in my head than when reading!)

You might want to try an alternative remote desktop utility; here I've
been using TigerVNC. ...Using Remmina because of its nice chart/listing feature: I can scan the chart for "Pi Clock" instead of 192.168.4.64).
For the Wayland-running systems I use TigerVNC and right now just type
in the IP.

(When experimenting with an alternative remote utilty, or anything, I'd recommed trying on a VM - virtual machine. That way if you don 't like
it it's not leaving traces behind in your real system, regardless of
pirging, etc.)




¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Running into stationary objects can be painful according to recent pole. --- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

It seems Remmina and Wayland don't get along all that well. Remmina
plays nice with X11, but Bookworm uses Wayland. There is a way to
switch to Wayland back to X11 -- should be easy enough to find else with
a bit of digging I could find it in my notes. I elected not to switch
back as thinking if I revert back then ll run into problems because eventually X11 will be depreciated and I'm back to the current problem.

That was sort of my thinking when I upgraded to Bookworm... that I was
better off going ahead with Wayland because X11 is closer to "going
away." ;)

IceWM seems to use (or plays nice with) Wayland. LXQT seems to be based
on X11: I was looking around for hints and one of the hits was "LXQt
Desktop Now '100%' Ready For Wayland", which to me implies it currently
uses X11, which might be why Remmina uses it.

That could very well be it. I assumed it was something happening on the
host system (i.e. the RDP daemon was choosing LXQt) but maybe it is
directly related to Remmina as the client.

You might want to try an alternative remote desktop utility; here I've
been using TigerVNC. ...Using Remmina because of its nice chart/listing feature: I can scan the chart for "Pi Clock" instead of 192.168.4.64).
For the Wayland-running systems I use TigerVNC and right now just type
in the IP.

It is not a super big problem as the system I am remoting into has plenty
of horsepower. A couple of other machines I have are not really capable of running something as lean as even LXQt so I only ever use IceWM on them
when logged on locally. I don't currently remote into them too often.

That said, this is a suggestion I need to keep in mind. I may give Tiger a
try if I have some spare time. Is it a client, or do you need to install something special on the host end also?

Mike

* SLMR 2.1a * Ultimate office automation: networked coffee machines.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

MIKE POWELL wrote:

It is not a super big problem as the system I am remoting into has plenty
of horsepower. A couple of other machines I have are not really capable of running something as lean as even LXQt so I only ever use IceWM on them
when logged on locally. I don't currently remote into them too often.

If you need even smaller, DamnSmallLinux with JWM might do it.

https://en.wikipedia.org/wiki/Damn_Small_Linux https://en.wikipedia.org/wiki/JWM

I like JWM on a small system; I prefer if it I'm using Puppy.


X11 is not going away any time soon, as Wayland still has its issues,
but X11 ceased development in 2012. People say "because it's a finished product" but I think the truth is that it's such spaghetti code that new people really cannot work on it, and old timers don't see the point.
It's a crutch on top of a shim held up by a brace.

Here is a good presentation on the problems that are baked into X11, and
why Wayland became necessary:

https://www.youtube.com/watch?v=cQoQE_HDG8g

On modern systems we don't realize how =SLOW= X11 is, but on a very old
box you could watch the screen redraw in slow motion.... it is probably
the biggest performance bottleneck in all of linux.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

It seems Remmina and Wayland don't get along all that well. Remmina
plays nice with X11, but Bookworm uses Wayland. There is a way to
switch to Wayland back to X11 -- should be easy enough to find else with
a bit of digging I could find it in my notes. I elected not to switch
back as thinking if I revert back then ll run into problems because eventually X11 will be depreciated and I'm back to the current problem.

That was sort of my thinking when I upgraded to Bookworm... that
I was better off going ahead with Wayland because X11 is closer
to "going away." ;)

Right. I have no idea why X11 can't do what Wayland does, or even what
the difference is: video display looks the same to me. "Obviously"
something is, determined by people who know more about it that I do.

IceWM seems to use (or plays nice with) Wayland. LXQT seems to be based
on X11: I was looking around for hints and one of the hits was "LXQt
Desktop Now '100%' Ready For Wayland", which to me implies it currently
uses X11, which might be why Remmina uses it.

That could very well be it. I assumed it was something happening
on the host system (i.e. the RDP daemon was choosing LXQt) but
maybe it is directly related to Remmina as the client.

Based on my very limited experience I would say Remmina is the problem
in that it doesn't have a Wayland-compatible element. (For those
reading along and not familiar with Remmina one needs to select a
protocol to view the rmeote screen.)

You might want to try an alternative remote desktop utility; here I've
been using TigerVNC. ...Using Remmina because of its nice chart/listing feature: I can scan the chart for "Pi Clock" instead of 192.168.4.64).
For the Wayland-running systems I use TigerVNC and right now just type
in the IP.

It is not a super big problem as the system I am remoting into
has plenty of horsepower. A couple of other machines I have are
not really capable of running something as lean as even LXQt so I
only ever use IceWM on them when logged on locally. I don't
currently remote into them too often.
That said, this is a suggestion I need to keep in mind. I may
give Tiger a try if I have some spare time. Is it a client, or
do you need to install something special on the host end also?

TigerVNC is installed only on the one 'main' system, the one you are one
using to look at the remote systems. The remote systems do not have to
have anything special other than VNC enabled (I'm thinking the Raspberry
Pi bring-up routine; to desktops similar). Might also have to have
Remmina installed on the remote systems, and I think that's only to have
needed libraries installed. Seemed like one new Pi I was playing with
didn't need Remmina and the other only connected after Remmina was
installed. (The currently-in-use Pi don't count as they already had
Remmina installed.)

So TigerVNC is only needed on the 'main' computer. The remote system
needs VNC activated and may or may not need Remmina to install some
mystery stuff. (I probably need to look at the how to properly install TigerVNC!)

...OK, since I'm a bit doi-it-right-the-first-time compulsive did a
quick look and found this article: https://picockpit.com/raspberry-pi/tigervnc-and-realvnc-on-raspberry-pi- bookworm-os/

Some of the steps don't seem necessary but I'm only following along in
my mind. One "not necessary" was in his 'VNC Viewer Connect' picture:
he is using a port number whereas I don't. (Maybe he revised his port
numbers for security reasons? -- Just told the world!)

* SLMR 2.1a * Ultimate office automation: networked coffee
machines. --- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)
þ RNET 2.10U: ILink: Christian Fellowship þ cfbbs.no-ip.com þ
856 933-7096

¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... She was like a magnet: attractive from back, repulsive from front.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

Forgot to add this to my previous message re: Wayland and Remmina and
TigerVNC (oh my! <g>)....

Wayland seems to not work headless -- without a monitor attached. Well,
IICC works, just defaults to a lower-resolution 4:3 remote display. The current work-around is to add a Dummy Monitor -- little critter
emulating a monitor.

Last batch I ordered from Amazon:
Dummy Monitor B0B5TMXLY5
Adapter cable B00JDRHQ58 RPi 4/5 is micro HDMI, dummy
monitor is HDMI

I sort of like having an indicator LED on the dummy monitor: I sort of
like things to help tell me if something is working or not: Dummy
Monitor's blue LED on: we've got power! Also I'd go for a high-ish
resolution for future compatibility: here the computer monitors are "1K"
1080 but eventually will fail and so possibly replaced by something with
a higher resolution; I try not to limit myself when buying new stuff.
For my current needs 4K seems more than enough.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Her eyes twinkled, like the moustache of a man with a cold.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Right. I have no idea why X11 can't do what Wayland does, or even what
the difference is: video display looks the same to me. "Obviously"
something is, determined by people who know more about it that I do.

Ky came up with a few reasons that sounded pretty good, especially if you
have a slower system. ;)

mike


* SLMR 2.1a * Perhaps this situation requires a more Klingon response.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

Wayland seems to not work headless -- without a monitor attached. Well,
IICC works, just defaults to a lower-resolution 4:3 remote display. The current work-around is to add a Dummy Monitor -- little critter
emulating a monitor.

Last batch I ordered from Amazon:
Dummy Monitor B0B5TMXLY5
Adapter cable B00JDRHQ58 RPi 4/5 is micro HDMI, dummy
monitor is HDMI

Luckily the system I am remoting into does have a monitor. If I intend for
a system to be headless, I usually just ssh into it and use the cli.
However, now I know that if I need to remote into a desktop, there is a
handy fix for it!

Mike


* SLMR 2.1a * I forget the dream, but I'm missing a pajama button...
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

MIKE POWELL wrote:

Right. I have no idea why X11 can't do what Wayland does, or even what
the difference is: video display looks the same to me. "Obviously"
something is, determined by people who know more about it that I do.

Ky came up with a few reasons that sounded pretty good, especially if you have a slower system. ;)

Puppy might be good too. It runs entirely from RAM, rather like a VM, so
it is very slick even on old hardware, and 340mb RAM is enough (that's
how much my very old laptop has, and it runs fine there, and doesn't
need much CPU either). It's based on Ubuntu tho much pared down, with
JWM as the default desktop.

Also there are a lot of flavors of Puppy (over 500 last I checked). So
there's one for everything. Some are maintained.

https://puppylinux-woof-ce.github.io/

Current flavors:
https://puppylinux-woof-ce.github.io/screenshots.html

Install notes:
https://puppylinux-woof-ce.github.io/install.html

[I would only do a full install, it's less trouble later on]
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

Right. I have no idea why X11 can't do what Wayland does, or even what
the difference is: video display looks the same to me. "Obviously" something is, determined by people who know more about it that I do.

Ky came up with a few reasons that sounded pretty good,
especially if you have a slower system. ;)

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Scientists have grown vocal cords in Petri dish. Results speak for self. --- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

Wayland seems to not work headless -- without a monitor attached. Well, IICC works, just defaults to a lower-resolution 4:3 remote display. The current work-around is to add a Dummy Monitor -- little critter
emulating a monitor.

Luckily the system I am remoting into does have a monitor. If I
intend for a system to be headless, I usually just ssh into it
and use the cli. However, now I know that if I need to remote
into a desktop, there is a handy fix for it!

Just remember to use the Mirror option or set the Dummy Monitor as the
second monitor else you'll be wondering where the menus and Desktop
items went!!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... She was like a magnet: attractive from back, repulsive from front.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Mike!

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

<looks at large stack of HDs each with a different OS, decides to hide
them so Barry won't have a brain seizure>
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

I used to experiment a lot with different OSes but have pretty much settled
now on one -- debian linux -- and a few of its close derivatives. For
example, I have two raspberry pis and they are running Raspbian, which is a debian derivative made specifically for them that pretty much works just
the same.

I do have a couple of machines that are old enough that they either lack
memory or more than two cores, and I run devuan on them. That is another debian derivative that doesn't include systemd. It seems to be a little
less memory intensive than debian proper.

In all cases, the user experience is very similar. ;)

Mike


* SLMR 2.1a * The only good MAC is a Big Mac.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

MIKE POWELL wrote:

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

I used to experiment a lot with different OSes but have pretty much settled now on one -- debian linux -- and a few of its close derivatives. For example, I have two raspberry pis and they are running Raspbian, which is a debian derivative made specifically for them that pretty much works just
the same.

I do have a couple of machines that are old enough that they either lack memory or more than two cores, and I run devuan on them. That is another debian derivative that doesn't include systemd. It seems to be a little
less memory intensive than debian proper.

Should be, yes.

Devuan's desktop and general way of doing things now default to the
PCLinuxOS incarnation of KDE, which I was pleased to see (tho last
update somewhat messed up... oh well, it's not my everyday). PCLOS is
much lighter on the hardware than most -- startup is 5 seconds flat on a midrange i7, and under 30 seconds on a 15 year old laptop that wasn't
much to start with.

PCLOS is a one-man-band, Tex is getting on in years and health not good,
so I've been hoping to find another distro I like as well, in case no
one takes up the torch. So I was glad to see Devuan at least adding the desktop. (Never liked Debian, tho..)
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

<looks at large stack of HDs each with a different OS, decides to
hide them so Barry won't have a brain seizure>

Too late! My brain has seized control! ...Alt Delete - gaaa!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Has anyone seen my jacket? It's white w/sleeves to make you hug yourself. --- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

> Saw those. I'll admit to be resistng using a variety of operating
> systems as just gets too confusing for me. I tend to use the computer
> as templates of each other.
KM> <looks at large stack of HDs each with a different OS, decides to
KM> hide them so Barry won't have a brain seizure>

Too late! My brain has seized control! ...Alt Delete - gaaa!

.. Has anyone seen my jacket? It's white w/sleeves to make you hug yourself.

Mine is more stylish... it has very long sleeves that tie in the back....
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

Saw those. I'll admit to be resistng using a variety of operating
systems as just gets too confusing for me. I tend to use the computer
as templates of each other.

I used to experiment a lot with different OSes but have pretty
much settled now on one -- debian linux -- and a few of its close derivatives. For example, I have two raspberry pis and they are
running Raspbian, which is a debian derivative made specifically
for them that pretty much works just the same.

I'm quite sure you and I could switch between various OSs if we needed
to -- the key being 'needed to'. For me just easier -- actually was
sort of the reason I got started with Ubuntu: at the time was using
Mythbuntu for recording TV shows and Windows XP for my other computing
needs. Windows kept 'demanding' money for upgrades -- ended up hardware
as well as software -- and to understand the Ubuntu part of Mythbuntu I
sort of needed a 'play' machine to experiment with without screwing up
my primary machine.

I do have a couple of machines that are old enough that they
either lack memory or more than two cores, and I run devuan on
them. That is another debian derivative that doesn't include
systemd. It seems to be a little less memory intensive than
debian proper.

I'm using a few around 25 year old machines as Frontends for playback of
the recorded shows. Did swap the hard drive to a SSD and max out the
RAM if I hadn't already. I know a couple of the machines that max is 8
GB. Zip right along! ...One machine for certain has reached it's
maximum OS: will run Ubuntu 20.04 but not 22.04.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Shout out to the people who don't know what the opposite of in is!
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

PCLOS is a one-man-band, Tex is getting on in years and health not good,
so I've been hoping to find another distro I like as well, in case no
one takes up the torch. So I was glad to see Devuan at least adding the desktop. (Never liked Debian, tho..)

That is good to see someone else pick it up. When I first wanted to try
linux, I got started with a CLI-only version of Slackware. I got a basic understanding of the command line. When I got a faster system I wanted to
try a GUI. I tried several different distros, all based on either Slackware
or Red Hat. The installers on a few were actually graphical, and those installer would do a *great* job identifying and correctly working with my video card.

However, who knows why, despite all of the "does this look good?" tests,
etc., the installers did, when it actually came to installing a *desktop*
that worked correctly, they all failed miserably.

A friend recommended a debian-based distro. I forget what it was called
but it was the FOSS version of Correl Linux. It didn't even have a fancy graphical installer, but it worked! They stopped maintaining it, so I
migrated to another debian based distro. I am surprised but I have finally forgotten the name! The fellow who maintained it passed away after a couple
of releases, so I migrated to debian proper.

I have run into a couple of issues from time to time, but they are usually
easy to fix.

Now Ubuntu is a different animal. Despite being debian based, I have had
no luck with it. I have an SBC that supposedly only works with it. I got
it installed fine the first time but, when it finally came time to upgrade
to a new version, that resulted in a non-working machine.

I have tried it other times in the past and that is always the same
story... I can get it installed and working until it is time for a version upgrade. I have never had a version upgrade that resulted in a working
machine afterwards.

That SBC now runs debian proper just fine.

Mike


* SLMR 2.1a * All true wisdom is found on T-shirts. --And in taglines.
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

BARRY MARTIN wrote:

sort of needed a 'play' machine to experiment with without screwing up
my primary machine.

That's always good. And you know one stays when you keep using it. :)

I'm using a few around 25 year old machines as Frontends for playback of
the recorded shows. Did swap the hard drive to a SSD and max out the
RAM if I hadn't already. I know a couple of the machines that max is 8
GB. Zip right along! ...One machine for certain has reached it's
maximum OS: will run Ubuntu 20.04 but not 22.04.

Nah, only old enough to vote... 8GB capacity arrived with the Core2Duo
in 2006.

And yeah, the Core2 and 8GB is now the practical divide between "still generally useful" and "just too durn slow" but they're starting to fall
out of "useful" for all but these limited specialty tasks that don't
need a lot of horsepower (for modern values of "a lot").

https://hyperboleandahalf.blogspot.com/2010/04/alot-is-better-than-you-at-everything.html
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

> Saw those. I'll admit to be resistng using a variety of operating
> systems as just gets too confusing for me. I tend to use the computer
> as templates of each other.
KM> <looks at large stack of HDs each with a different OS, decides to
KM> hide them so Barry won't have a brain seizure>
Too late! My brain has seized control! ...Alt Delete - gaaa!
.. Has anyone seen my jacket? It's white w/sleeves to make you hug yourself.

Mine is more stylish... it has very long sleeves that tie in the
back....

Ah! The Pacer Style! Very popular among upper echelon executives who
like to strut!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Lady Godiva - What do mean this isn't business casual?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

MIKE POWELL wrote:

PCLOS is a one-man-band, Tex is getting on in years and health not good,
so I've been hoping to find another distro I like as well, in case no
one takes up the torch. So I was glad to see Devuan at least adding the
desktop. (Never liked Debian, tho..)

That is good to see someone else pick it up. When I first wanted to try

Yeah, Tex has it optimized and the installer set up so it's WAY slicker
than anything else. Full install of the everything-and-the-kitchen-sink edition takes five minutes and two clicks. Reboot on an i7-3xxx is 30
seconds from power-cycle to desktop. And it's very reliable.

linux, I got started with a CLI-only version of Slackware. I got a basic understanding of the command line. When I got a faster system I wanted to try a GUI. I tried several different distros, all based on either Slackware or Red Hat. The installers on a few were actually graphical, and those installer would do a *great* job identifying and correctly working with my video card.

The first linux I installed was RedHat6. It was dreadful. I had Win95 on
the same hardware and it ran rings around RH. RH was downright sluggish,
and there was no getting it to play nice with the fairly-ordinary
vidcard (stuck at 640x480). One day it forgot its password, and that was
when I made it go away.

However, who knows why, despite all of the "does this look good?" tests, etc., the installers did, when it actually came to installing a *desktop* that worked correctly, they all failed miserably.

Yeah, it used to be pretty bad. Amateur hour plus too many cooks and
only doing the part they liked. The first one I had any success with was Mandrake 7.2, but it wasn't sufficiently complete for an everyday
desktop. But I did like KDE, far as I got. I still prefer it.

A friend recommended a debian-based distro. I forget what it was called
but it was the FOSS version of Correl Linux. It didn't even have a fancy graphical installer, but it worked! They stopped maintaining it, so I migrated to another debian based distro. I am surprised but I have finally forgotten the name! The fellow who maintained it passed away after a couple of releases, so I migrated to debian proper.

If you want to run WordPerfect for linux, you have to run Corel Linux as
the OS. Tho it's reportedly not very stable.

Now Ubuntu is a different animal. Despite being debian based, I have had
no luck with it. I have an SBC that supposedly only works with it. I got
it installed fine the first time but, when it finally came time to upgrade
to a new version, that resulted in a non-working machine.

I loathe Ubuntu, think Gnome makes Win10 look usable, and no longer even
look at any flavor of Ubuntu (Kubuntu is the poor relation there, and
just not a very good incarnation of KDE).

I have tried it other times in the past and that is always the same
story... I can get it installed and working until it is time for a version upgrade. I have never had a version upgrade that resulted in a working machine afterwards.

With the SBC or linux in general?

Here... no SBCs but assorted random PCs.

Used to be until about 7-8 years ago you could COUNT on doing a full reinstall, because version upgrade almost never worked, and sometimes
updates didn't work either. Since I regard reinstalling as a mortal sin,
this was a total dealbreaker for me. But it's been a lot better since:

Devuan wasn't quite right after the most recent update (not upgrade).
Don't recall quite what, like it messed up some of my visual settings or something. Not a regular use setup so didn't much care, I'll probably
just redo it from scratch and hope they've fixed whatever.

My Fedora setup has been ugraded from v32 to v39 and I see v40 is out.
Upgrade is an overnight job but I have not so far had it screw up. It
has sometimes told me "if you want to clean up this mess, input this
command" and I did and it fixed whatever hadn't upgraded. This is its
way of preventing dependency hell, I guess. I only use the CLI for
updates because I don't like Discover. I've been doing this long enough
that I have all the update/upgrade command in the buffer and it's just up-arrow enter.

PCLOS is rolling so there is no such thing as upgrade, and updates are continuous. Occasional minor glitches but overall I find I prefer
rolling... problems get fixed a LOT faster. Synaptic is ugly but it
works really nice for updates.

There's a Debian install in the stack, but I dislike Debian (even with
KDE desktop) and since I never use it, haven't been arsed to update it.
I suppose it's a full version up by now.

That SBC now runs debian proper just fine.

That should be true for any hardware that "requires" some downstream
distro. Frex, if it wants Ubuntu, there's no reason it can't run
Debian... so long as it's compiled for that CPU.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

KY MOFFET wrote:

If you want to run WordPerfect for linux, you have to run Corel Linux as
the OS. Tho it's reportedly not very stable.

By weird coincidence, MichaelMJD on Youtube just uploaded a video on
this very topic.

I am following me around in advance. <g>
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

I have tried it other times in the past and that is always the same story... I can get it installed and working until it is time for a version upgrade. I have never had a version upgrade that resulted in a working machine afterwards.

With the SBC or linux in general?

Ubuntu on any machine, SBC or PC. The upgrade always leaves the system
in question unusable.

Used to be until about 7-8 years ago you could COUNT on doing a full reinstall, because version upgrade almost never worked, and sometimes
updates didn't work either. Since I regard reinstalling as a mortal sin,
this was a total dealbreaker for me. But it's been a lot better since:

Same here. I can remember needing to do the full reinstall, but have not
had to do that with debian (or devuan) in quite a while... well, except for when I accidentally ran a machine out of space. ;)

Devuan wasn't quite right after the most recent update (not upgrade).
Don't recall quite what, like it messed up some of my visual settings or something. Not a regular use setup so didn't much care, I'll probably
just redo it from scratch and hope they've fixed whatever.

I use Devuan on two machines. One is a PC that I do use the GUI on. That machine has always had a quirk where, every so often, the screen will just disappear and the machine requires a reboot. Do you remember ever trying
to load a GIF or other graphics on an XT and having it not work? That is
what the screen will suddenly look like... a big mess. ;)

It does that no matter what distro is running on it so I figure it is
something wrong with the box itself.

On the other machine, I do not have the GUI installed and use the cli only.
I find it works well in that situation.

I tried installing it on a third machine that I wanted to use the GUI on
and wound up installing debian instead after some goofy things happened
(forget what now).

PCLOS is rolling so there is no such thing as upgrade, and updates are continuous. Occasional minor glitches but overall I find I prefer
rolling... problems get fixed a LOT faster. Synaptic is ugly but it
works really nice for updates.

Does PCLOS have a cli install program, like apt or ???

That SBC now runs debian proper just fine.

That should be true for any hardware that "requires" some downstream
distro. Frex, if it wants Ubuntu, there's no reason it can't run
Debian... so long as it's compiled for that CPU.

I *think* the part that may not work, but that I don't use on this one, is
the 40-pin whatchamacallit. I think the maintainers only make that code available to ubuntu. Since I don't use that bit, debian works just fine
and I suspect if I wanted to use that bit I could find the code and compile
it.

Mike


* SLMR 2.1a * Overdrawn? No way! I still have checks left!
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

Hi Ky!

sort of needed a 'play' machine to experiment with without screwing up
my primary machine.

That's always good. And you know one stays when you keep using
it. :)

It won't just be taking up space?!

I'm using a few around 25 year old machines as Frontends for playback of
the recorded shows. Did swap the hard drive to a SSD and max out the
RAM if I hadn't already. I know a couple of the machines that max is 8
GB. Zip right along! ...One machine for certain has reached it's
maximum OS: will run Ubuntu 20.04 but not 22.04.

Nah, only old enough to vote... 8GB capacity arrived with the
Core2Duo in 2006.

So maybe closer to 20 -- almost like presenting a false ID! ...One of
the old-enough-to-vote is the one where it says it can use 8 GB (4x 2GB)
but it is picky and only wants 6 (2x2GB and 2x1GB). A while back did a
bit of experimenting: moved the 2GB pairs around, memory testing, etc.
Nope: only wanted 6 GB and played possum at 8.

And yeah, the Core2 and 8GB is now the practical divide between
"still generally useful" and "just too durn slow" but they're
starting to fall out of "useful" for all but these limited
specialty tasks that don't need a lot of horsepower (for modern
values of "a lot").

https://hyperboleandahalf.blogspot.com/2010/04/alot-is-better-than -you-at-every
hing.html

Well, modified that to what looked right, even 'hing' to 'thing' - page
not found but did give me some interesting-looking options so clicked.
...Did eventually find the article: cut off the alot- stuff and read a
couple of quirky articles and the third or fourth one down is what was
looking for.

Copy and paste: https://hyperboleandahalf.blogspot.com/2010/04/alot-is-better-than-you-a t-everything.html

Looks the same. <shrug> May have been those eagle talons.

But since this was about old computers and their horsepower (so that's
the engine revving -- thought it was just a cooling fan gone hyper) the
older computers are gradually being replaced by the new, sleek,
Raspberry Pi's. The RPi4's seem to be quite fine for my MythTV
Frontend usage (playback of recorded OTA shows). I'm expecting the RPi5
to be even better but right now can't check: explanation is a little complicated but boils down to RPi5 wants Bookworm and MythTV v31 wants
Buster. (Anyone wanting details post.)



¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... antistalking (v.) - Learning a person's routine in order to avoid them.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

MIKE POWELL wrote to KY MOFFET <=-

Now Ubuntu is a different animal. Despite being debian based, I
have had no luck with it. I have an SBC that supposedly only
works with it. I got it installed fine the first time but, when
it finally came time to upgrade to a new version, that resulted
in a non-working machine.

I have tried it other times in the past and that is always the
same story... I can get it installed and working until it is time
for a version upgrade. I have never had a version upgrade that
resulted in a working machine afterwards.

I started with Ubuntu because it was the 'mother' of Mythbuntu which I
was using to record my TV shows. At the time Mythbutu was the OS and
utility rolled in one; several years back they gave up trying to
maintain everything so now MythTV installs on a Ubuntu system. ...Back
to the original days, I was casually looking for a Windows replacement
and made sense to try something on which my TV view computers were based
on -- easier to troubleshoot if I'm semi-familiar with how the thing
works.

As far as upgrading to the new version, I've also had problems. Seems
to be a lot smoother with the current versions but the older
versions..... Maybe around 20.04 they figured things out. I also could
have complicated things by trying to upgrade a computer near the bottome
of the minimum hardware requirements (like my computer which says will
take 8 GB but it won't boot if more thna 6 GB).

Anyway, the problems seem to be resolved by installing using the nomodeset option (the editing it out when things work properly). There's also an
option is some BIOSs/UEFI's which has to be turned off to install a
non-Windows OS.

I've been using the Trickle Down Upgrade: I get the new system. It gets
built and tested while the current system is running. The new system
then gets swapped in, tested, and then the old system gets upgraded to
the new OS, replacement, not updating. This system then replaces
another system, etc.

One little problem is MythTV requires the Frontend (viewing) and Backend (recording) versions to match, so I have to have at least two computers upgraded. The good news is now I should be able to use Raspberry Pi 4
or 5 for the new Frontend, making the physical swap a lot easier. (The
old computers are big and have to fit the available space: a tower
format won't fit a desktop cubbyhole.)


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Don't give up on your dreams: go back to bed!
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

MIKE POWELL wrote:

I have tried it other times in the past and that is always the same
story... I can get it installed and working until it is time for a version >>> upgrade. I have never had a version upgrade that resulted in a working
machine afterwards.

With the SBC or linux in general?

Ubuntu on any machine, SBC or PC. The upgrade always leaves the system
in question unusable.

Ah. Well, one more reason to dislike Ubuntu.

Used to be until about 7-8 years ago you could COUNT on doing a full
reinstall, because version upgrade almost never worked, and sometimes
updates didn't work either. Since I regard reinstalling as a mortal sin,
this was a total dealbreaker for me. But it's been a lot better since:

Same here. I can remember needing to do the full reinstall, but have not
had to do that with debian (or devuan) in quite a while... well, except for when I accidentally ran a machine out of space. ;)

LOL. I would hear the word "upgrade reinstall" and head for the hills,
so I never had the displeasure.

I use Devuan on two machines. One is a PC that I do use the GUI on. That machine has always had a quirk where, every so often, the screen will just disappear and the machine requires a reboot. Do you remember ever trying
to load a GIF or other graphics on an XT and having it not work? That is what the screen will suddenly look like... a big mess. ;)

It does that no matter what distro is running on it so I figure it is something wrong with the box itself.

Bad video RAM, probably at the tail end of the addressing space and
marginal rather than failed, so it only occasionally messes up.

PCLOS is rolling so there is no such thing as upgrade, and updates are
continuous. Occasional minor glitches but overall I find I prefer
rolling... problems get fixed a LOT faster. Synaptic is ugly but it
works really nice for updates.

Does PCLOS have a cli install program, like apt or ???

Yes, but I've never used it. It's some unholy hybrid of APT and RPM.

That SBC now runs debian proper just fine.

I *think* the part that may not work, but that I don't use on this one, is the 40-pin whatchamacallit. I think the maintainers only make that code available to ubuntu. Since I don't use that bit, debian works just fine
and I suspect if I wanted to use that bit I could find the code and compile it.

Nah, there's no code for the GPIO (40 pin) header, and nothing to do
with Ubuntu It's all hardware and firmware. You have to write and run
Python scripts to control each pin individually. ExplainingComputers
youtube channel has done a few vids on simple scripts, to control a
weather station and a toy robot, stuff like that. Mostly useful to
makerspace, not to general-purpose computing.

https://www.raspberrypi-spy.co.uk/2012/06/simple-guide-to-the-rpi-gpio-header-and-pins/
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

As far as upgrading to the new version, I've also had problems. Seems
to be a lot smoother with the current versions but the older
versions..... Maybe around 20.04 they figured things out. I also could
have complicated things by trying to upgrade a computer near the bottome
of the minimum hardware requirements (like my computer which says will
take 8 GB but it won't boot if more thna 6 GB).

IIRC, the issue I last ran into -- I suspect -- had something to do with
how they have moved some "parts" of the OS and accessories from the
completely free version to one that maybe involves paid support, or at
least some level beyond what just the FOSS version I was using. apt threw
some messages I am not used to seeing with other debian variants that lead
me to believe as much.

Whatever packages I had that they had since "decoupled" left the system unstable. They were not familiar to me or I might have realized what was
going on sooner.

Previous times that was not the issue, and I would still wind up with a
system that was not at the same level of "useability" as it was before the upgrade.

IMHO, it probably has something to do with them mixing in things from
debian stable along with things from other branches (testing, unstable)
that screws it up. Or maybe it is just some other ubuntu-specific modifications that they make.

Anyway, the problems seem to be resolved by installing using the nomodeset option (the editing it out when things work properly). There's also an option is some BIOSs/UEFI's which has to be turned off to install a non-Windows OS.

I have run into some issues when installing debian/devuan while being near
the bottom of the minimum requirements. I usually go with the
non-graphical install option and that seems to get me around the issues.

One little problem is MythTV requires the Frontend (viewing) and Backend (recording) versions to match, so I have to have at least two computers upgraded. The good news is now I should be able to use Raspberry Pi 4
or 5 for the new Frontend, making the physical swap a lot easier. (The
old computers are big and have to fit the available space: a tower
format won't fit a desktop cubbyhole.)

Sounds complicated, but it also sounds like you have a good process in
place for making it work. ;) It is also good that you can use a Pi. Not
only do they require less space but they should also draw less power.

Mike


* SLMR 2.1a * Tinnn Rooooooooof! --Rusted!
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

It does that no matter what distro is running on it so I figure it is something wrong with the box itself.

Bad video RAM, probably at the tail end of the addressing space and
marginal rather than failed, so it only occasionally messes up.

I should see if the MB will allow me to drop another video card in there
and maybe try that. IIRC, I don't think I have tried that before. The
good thing is that I don't use that box much... usually only when I need
access to a CD/DVD player/writer does it get fired up.

Does PCLOS have a cli install program, like apt or ???

Yes, but I've never used it. It's some unholy hybrid of APT and RPM.

RPM. That is the name I could not remember. ;)

Mike


* SLMR 2.1a * AAAAA - American Association Against Acronym Abuse
--- SBBSecho 3.20-Linux
* Origin: ILink: CCO - capitolcityonline.net (454:3/105)

Hi Mike!

KY MOFFET wrote to MIKE POWELL <=-

<snip-rip-tear just leaving enough for what is the reference>

There's a Debian install in the stack, but I dislike Debian (even
with KDE desktop) and since I never use it, haven't been arsed to
update it. I suppose it's a full version up by now.

That SBC now runs debian proper just fine.

That should be true for any hardware that "requires" some
downstream distro. Frex, if it wants Ubuntu, there's no reason it
can't run Debian... so long as it's compiled for that CPU.

I won't disagree with Ky as he has his set of needs-wants-desires that
differ from others. Not right, not wrong. LIS in an earlier message, I switched to Ubuntu because that's what MythBuntu was using and being semi-familiar with the OS helped me figure out what wasn't quite right
with my Myth devices. For me also worked out as the Raspberry Pi's used
an operating system similar to Ubuntu.

Is Ubuntu THE one? According to Ky no, and I'll agree with him up to a
point. I've had my share of well-this-is-stupid-why-isn't-it-fixed.
Ubuntu seems to be fairly well supported by having a wide variety of
optional utilities available (and at a super-low price!!), though that statement is biased as I haven't really tried other OSs -- my main beef
is when only offered for Windows and then I sometimes whine (anyone
catch the homonym of 'whine' and 'WINE'?!).

I think some of the 'complications' of whichever version of Linux one
selects is because Windows does things as a benevolent bully. Microsoft
has the money and manpower to figure out how to have the installer
programmes make the discovery of which CPU its looking at, etc. I'm
thinking Linux can also, just maybe not worth the pages of code to do
it. I know one of the original concepts was to keep the installation
small enough to fit on a CD: 650 MB.

As for the bully part, it seems Microsoft has told compmuter component manufactures "do it my way and don't play with anyone else". Probably
not in so many words as that's inviting legal problems, but sure seems
to me like there's a bit of fear going on.



¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... "I just dialed a wrong 800 number. Will I be charged?"
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

MIKE POWELL wrote:

It does that no matter what distro is running on it so I figure it is
something wrong with the box itself.

Bad video RAM, probably at the tail end of the addressing space and
marginal rather than failed, so it only occasionally messes up.

I should see if the MB will allow me to drop another video card in there
and maybe try that. IIRC, I don't think I have tried that before. The
good thing is that I don't use that box much... usually only when I need access to a CD/DVD player/writer does it get fired up.

Linux can be real fussy about vidcards. Sometimes it'll throw up all
over a new one unless it's in the very same family. Sometimes it doesn't
care. I haven't bothered sorting it out, might be how well the distro
handles newly-added hardware.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Mike!

KM> That should be true for any hardware that "requires" some
KM> downstream distro. Frex, if it wants Ubuntu, there's no reason it
KM> can't run Debian... so long as it's compiled for that CPU.

I won't disagree with Ky as he has his set of needs-wants-desires that
differ from others. Not right, not wrong. LIS in an earlier message, I

That wasn't what I said... To clarify for the distro impaired: <g>

If it runs Ubuntu, it should have NO trouble with Debian (PROVIDED it
was compiled for that CPU, eg. Rasbian for ARM), because Ubuntu is
Debian under the hood, but Debian is lighter-weight. Devuan is also
basically Debian, but again not so heavy. So unless there's some very
specific added function (Ubuntu has been known to, uh, customize the
desktop beyond all recognition, or in your case Myth only knowing Ubuntu
for the other end) they should be *functionally* interchangeable, tho
Debian or Devuan may run better on an older system.

And more extreme, given Mint is based on Ubuntu... Mint is only about a quarter as system-intensive as Ubuntu, and somewhat less than Debian
too. So a system that runs Mint fine may be sloggy with Debian and
struggle with Ubuntu, even tho under the hood they are all Debian.

Again, Puppy is Ubuntu under the hood, so any system that runs Ubuntu
should also run Puppy, tho Puppy is much lighter weight.

There's a similar ecosystem surrounding Fedora and what used to be
Mandrake. PCLOS is derived from Mandrake, so it's over here, and what
runs Fedora or Mageia or Mandriva will generally run any of their
kinfolk (tho some are much lighter than others).

And there are other ancestral ecologies that are vertically mostly interchangeable in terms of "if it runs Big brother, it will run Parent
or Little brother".

The fact that I dislike Debian and loathe Ubuntu doesn't change any of that.

Is Ubuntu THE one? According to Ky no, and I'll agree with him up to a point. I've had my share of well-this-is-stupid-why-isn't-it-fixed.
Ubuntu seems to be fairly well supported by having a wide variety of
optional utilities available (and at a super-low price!!), though that

Oh, they've discovered the Store. Who could have predicted....

statement is biased as I haven't really tried other OSs -- my main beef
is when only offered for Windows and then I sometimes whine (anyone
catch the homonym of 'whine' and 'WINE'?!).

LOL. I've had zero luck with WINE. Gave up and use XP in a VM.

Whine whine whine!

I think some of the 'complications' of whichever version of Linux one
selects is because Windows does things as a benevolent bully. Microsoft

So does Red Hat in the linux world, and Ubuntu too because it's is the
base for so many downstream distros (and that happened because both
could PAY coders to work on it full time, something no one else did).
You will do things our way or the highway. So we have systemd and
Wayland proliferating throughout the distro world whether they're ready
for prime time or not, because the major upstream distros most minor
distros depend on switched, and if you don't like it, you're Devuan or
PCLOS or some other one-man-band distro of no market significance.

has the money and manpower to figure out how to have the installer
programmes make the discovery of which CPU its looking at, etc. I'm

It exists for most hardware, but if you're not a Mandrake descendant,
you may not have it. Mandrake did things Differently... it actually went
and did proper hardware detection long before anyone else even thought
of it (that's why PCLOS is better at this, it started life as Mandrake).
Well, at least we don't have to manually set up the X server anymore!!

thinking Linux can also, just maybe not worth the pages of code to do
it. I know one of the original concepts was to keep the installation
small enough to fit on a CD: 650 MB.

That went away a loooong time ago. Not long after giving up the idea
that it would be wholly a command-line OS.

I remember the first time I saw some early version (1996) of GIMP in
action... the image was in one window, and the parameters to be changed
were displayed in another window, and you typed in the parameters to be altered at the CLI prompt. Interesting system for the time, but ... jury-rigged is polite. This is why it took GIMP so long to decide that
maybe it was desirable to have the whole program in one window!

As for the bully part, it seems Microsoft has told compmuter component manufactures "do it my way and don't play with anyone else". Probably
not in so many words as that's inviting legal problems, but sure seems
to me like there's a bit of fear going on.

Uh, no. Microsoft's only function there is to install drivers, which
come from the hardware manufacturer. If the mfgr doesn't want to provide Microsoft with a driver to include in Windows, then it'll scrape by on
the default driver (if any, otherwise it just won't work) until you
break down and install the mfgr's driver. NVidia did this, because they
didn't want to provide the necessary source code.

What Microsoft CAN do is set minimum system requirements, like Win11
needing a TPM chip or it won't install (without mucking about in the Registry). Or refuse to include a known-problem driver.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KY MOFFET wrote to MIKE POWELL <=-

I use Devuan on two machines. One is a PC that I do use the GUI on. That machine has always had a quirk where, every so often, the screen will just disappear and the machine requires a reboot. Do you remember ever trying
to load a GIF or other graphics on an XT and having it not work? That is what the screen will suddenly look like... a big mess. ;)
It does that no matter what distro is running on it so I figure it is something wrong with the box itself.

Bad video RAM, probably at the tail end of the addressing space
and marginal rather than failed, so it only occasionally messes
up.

Well I'm going to disagree with the bad video RAM statement thought I'll probably be learning a lot by the 'rebuttal'!

Here have two essentially headless systems: video connected to a KVM
switch; rarely need to work directly on either unit. One has an antique
GT218 video card, the other uses built-in graphics from the CPU. Both
have eventually 'gone to black' in that they work fine except for no
video.

The other computers to the KVM switch do not have the loss of video
issue.

Doesn't eliminate video addressing statement as a possibility. I don't
recall how to look up; guessing "all f's" in the range end-point would
mean it's at the far end.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Do invisible cats drink evaporated milk?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

KY MOFFET wrote to MIKE POWELL <=-

> I use Devuan on two machines. One is a PC that I do use the GUI on. That
> machine has always had a quirk where, every so often, the screen will just
> disappear and the machine requires a reboot. Do you remember ever trying
> to load a GIF or other graphics on an XT and having it not work? That is
> what the screen will suddenly look like... a big mess. ;)
> It does that no matter what distro is running on it so I figure it is
> something wrong with the box itself.
KM> Bad video RAM, probably at the tail end of the addressing space
KM> and marginal rather than failed, so it only occasionally messes
KM> up.

Well I'm going to disagree with the bad video RAM statement thought I'll probably be learning a lot by the 'rebuttal'!

It's the description of the fault that tells me it's most likely bad
video RAM.
That's quite distinctive -- you get a solid mass of random ASCII
characters. We used to see that a lot in the Olden Tymes.

When the card is loose, or the GPU is failing, you get random lines.

When the cable is loose, you'll lose some color (and have frex a pink
screen).

This isn't absolute but it's a pretty good guideline.

Doesn't eliminate video addressing statement as a possibility. I don't recall how to look up; guessing "all f's" in the range end-point would
mean it's at the far end.

Addressing shouldn't cause this kind of visual screw up; it would cause
a lockup or no video at all due to the conflict. And if you're not a
modern gamer, or rendering video, you're not gonna fill up that video
RAM anyway.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Mike!

As far as upgrading to the new version, I've also had problems. Seems
to be a lot smoother with the current versions but the older
versions..... Maybe around 20.04 they figured things out. I also could have complicated things by trying to upgrade a computer near the bottome
of the minimum hardware requirements (like my computer which says will
take 8 GB but it won't boot if more thna 6 GB).

IIRC, the issue I last ran into -- I suspect -- had something to
do with how they have moved some "parts" of the OS and
accessories from the completely free version to one that maybe
involves paid support, or at least some level beyond what just
the FOSS version I was using. apt threw some messages I am not
used to seeing with other debian variants that lead me to believe
as much.
Whatever packages I had that they had since "decoupled" left the
system unstable. They were not familiar to me or I might have
realized what was going on sooner.

I'll admit to kind of starting at the "I'm not familiar" step because I generally don't deal with the hardware and Operating System levels all
that often -- no need to. I have noticed they do move things around:
Swap File came to mind and the why magically disappeared. %) ...Well,
in the Raspberry Pi OS the boot configuration file was move to the
firmware subdirectory; if a piece of software doesn't update that during
an upgrade I would image things can break -- sort of your 'decoupling'.

Previous times that was not the issue, and I would still wind up
with a system that was not at the same level of "useability" as
it was before the upgrade.

I've sort of had that issue also but I can't always point to the upgrade process: some of the failure/instability could be attributed to me swapping/adding/other changing of the hardware, being unfamilar with
UEFI, and the like. ...Had one brand new installation and the system
was unstable: ended up the problem was faulty RAM (freshly unpacked).

IMHO, it probably has something to do with them mixing in things
from debian stable along with things from other branches
(testing, unstable) that screws it up. Or maybe it is just some
other ubuntu-specific modifications that they make.

I'd guess a mix of all that! I'm sort of learned to test new stuff on a Virtual Machine because if don't like it or misconfigured easier to
blast the VM -- the uninstall, purge, etc., processes don't always
remove all traces.

Anyway, the problems seem to be resolved by installing using the nomodeset option (the editing it out when things work properly). There's also an option is some BIOSs/UEFI's which has to be turned off to install a non-Windows OS.

I have run into some issues when installing debian/devuan while
being near the bottom of the minimum requirements. I usually go
with the non-graphical install option and that seems to get me
around the issues.

I haven't learned to do that yet!

One little problem is MythTV requires the Frontend (viewing) and Backend (recording) versions to match, so I have to have at least two computers upgraded. The good news is now I should be able to use Raspberry Pi 4
or 5 for the new Frontend, making the physical swap a lot easier. (The
old computers are big and have to fit the available space: a tower
format won't fit a desktop cubbyhole.)

Sounds complicated, but it also sounds like you have a good
process in place for making it work. ;) It is also good that
you can use a Pi. Not only do they require less space but they
should also draw less power.

Yes, the 'rolling upgrade process' can be a litte detail-filled!


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Powered by Miracle Software; our motto: "If it works, it's a Miracle!"
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

KM> That should be true for any hardware that "requires" some
KM> downstream distro. Frex, if it wants Ubuntu, there's no reason it
KM> can't run Debian... so long as it's compiled for that CPU.
I won't disagree with Ky as he has his set of needs-wants-desires that differ from others. Not right, not wrong. LIS in an earlier message, I

That wasn't what I said... To clarify for the distro impaired: <g>

Oops! That's what I get for not being overly familiar with the stuff!

Is Ubuntu THE one? According to Ky no, and I'll agree with him up to a point. I've had my share of well-this-is-stupid-why-isn't-it-fixed.
Ubuntu seems to be fairly well supported by having a wide variety of optional utilities available (and at a super-low price!!), though that

Oh, they've discovered the Store. Who could have predicted....

<Trying to come up with an Economist-based reply and....> <g>

statement is biased as I haven't really tried other OSs -- my main beef
is when only offered for Windows and then I sometimes whine (anyone
catch the homonym of 'whine' and 'WINE'?!).

LOL. I've had zero luck with WINE. Gave up and use XP in a VM.
Whine whine whine!

It's 1700 somewhere! <bseg>

As for WINE, I rarely use it because I sort of forget about it. Think
the last time I used it was from DNS Benchmarks (GRC.com) and according
to my notes that was around a year ago. I mostly use 'random' Windows utilities for repairing thumbdrives -- which has a bit of a problem
because if the host system (Ubuntu in my case) doesn't detect or
properly detect it won't pass through to the VM.



¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Hmmmm... is it supposed to smoke like that?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

It does that no matter what distro is running on it so I figure it is something wrong with the box itself.

Bad video RAM, probably at the tail end of the addressing space and
marginal rather than failed, so it only occasionally messes up.

I should see if the MB will allow me to drop another video card
in there and maybe try that. IIRC, I don't think I have tried
that before. The good thing is that I don't use that box much...
usually only when I need access to a CD/DVD player/writer does it
get fired up.

Another consideration is the video cable and connections. I had a
computer downstairs with no problems with video but occasionally the
sound would not work. Ended up the HDMI cable needed to be shoved in al
lthe way: plastic cover around the cable end touched the chassis and so
it felt like it was fully inserted but was not.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... If England goes metric will they still call it Scotland Yard?
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Mike!
I'd guess a mix of all that! I'm sort of learned to test new stuff on a Virtual Machine because if don't like it or misconfigured easier to
blast the VM -- the uninstall, purge, etc., processes don't always
remove all traces.

Assuming you can get it to install and run in the VM. You should hear my
VM War Stories....

Yes, the 'rolling upgrade process' can be a litte detail-filled!

"Rolling" is supposed to refer to the software, not the server cabinet!


I've come to greatly prefer a rolling distro for an everyday desktop.
There are NO UPGRADES and therefore NO REINSTALLS.

Someone took the oldest extant PCLOS from 2010 to 2021 (I think it was)
using only the rolling update process via Synaptic, and only twice did
he have to stop and twiddle something at the command line. But normally
it's just click-the-usuals, go away for a while, and it's done.

The other advantage is that problems get fixed NOW, not when the next
major version rolls around.

The downside is you have to do your updates regularly, because it can
get out of sync if you let things go too long. (Tho Synaptic at least is pretty good at squaring things up.) But it's fast enough, given it's
usually small, to do 'em often.

So my experience is that overall, rolling is more stable, because it's
always up to date and there are never any major upheavals.

Speaking therewhich, I need to do the twice-yearly version upgrade on
the Fedora system, which I keep forgetting because it's an overnight job (sometimes two nights) and takes some babysitting before it gets going.
Grrr.
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

> KM> That should be true for any hardware that "requires" some
> KM> downstream distro. Frex, if it wants Ubuntu, there's no reason it
> KM> can't run Debian... so long as it's compiled for that CPU.
> I won't disagree with Ky as he has his set of needs-wants-desires that
> differ from others. Not right, not wrong. LIS in an earlier message, I
KM> That wasn't what I said... To clarify for the distro impaired: <g>

Oops! That's what I get for not being overly familiar with the stuff!

I have become painfully familiar. <g>

> Is Ubuntu THE one? According to Ky no, and I'll agree with him up to a
> point. I've had my share of well-this-is-stupid-why-isn't-it-fixed.
> Ubuntu seems to be fairly well supported by having a wide variety of
> optional utilities available (and at a super-low price!!), though that
KM> Oh, they've discovered the Store. Who could have predicted....

<Trying to come up with an Economist-based reply and....> <g>

....anything you're willing to pay for, you get more of!!

> statement is biased as I haven't really tried other OSs -- my main beef
> is when only offered for Windows and then I sometimes whine (anyone
> catch the homonym of 'whine' and 'WINE'?!).
KM> LOL. I've had zero luck with WINE. Gave up and use XP in a VM.
KM> Whine whine whine!

It's 1700 somewhere! <bseg>

<tips tricorn hat>

As for WINE, I rarely use it because I sort of forget about it. Think
the last time I used it was from DNS Benchmarks (GRC.com) and according
to my notes that was around a year ago. I mostly use 'random' Windows utilities for repairing thumbdrives -- which has a bit of a problem
because if the host system (Ubuntu in my case) doesn't detect or
properly detect it won't pass through to the VM.

Generally if I'm using linux I'm using linux, and I expect things to
behave like linux.

But yeah, there are some things linux still doesn't do well, and may
never do well, or even at all.

Frex, there is NO dedicated RTF editor for linux. I've looked. No,
LibreOffice is not satisfactory (insert rant about clean formatting code
vs printer-defined formatting and how the latter needs to be stripped
out for publication, and also to avoid random screwups). So... XP in a
VM, and my dedicated RTF editor to the rescue.

It's a nuisance, but... better than fighting with WINE. And XP will
always speak to the whole network, which linux never will. (Does not
like random other linux boxen, never mind random Windows. And cannot be trusted to write files without fragmenting them all over the place.)
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

> I use Devuan on two machines. One is a PC that I do use the GUI on.

ha

> machine has always had a quirk where, every so often, the screen will

us

> disappear and the machine requires a reboot. Do you remember ever

rying

> to load a GIF or other graphics on an XT and having it not work? That

s

> what the screen will suddenly look like... a big mess. ;)
> It does that no matter what distro is running on it so I figure it is
> something wrong with the box itself.
KM> Bad video RAM, probably at the tail end of the addressing space
KM> and marginal rather than failed, so it only occasionally messes
KM> up.
Well I'm going to disagree with the bad video RAM statement thought I'll probably be learning a lot by the 'rebuttal'!

It's the description of the fault that tells me it's most likely
bad video RAM.
That's quite distinctive -- you get a solid mass of random ASCII characters. We used to see that a lot in the Olden Tymes.
When the card is loose, or the GPU is failing, you get random
lines.

Ahhh! I don't recall the random characters part. The make-and-break of
the data stream would cause the card to try to make sense of it and output
what it thought it was told.

When the cable is loose, you'll lose some color (and have frex a
pink screen).

Right.

This isn't absolute but it's a pretty good guideline.

Good starting points which usually work.

Doesn't eliminate video addressing statement as a possibility. I don't recall how to look up; guessing "all f's" in the range end-point would
mean it's at the far end.

Addressing shouldn't cause this kind of visual screw up; it would
cause a lockup or no video at all due to the conflict. And if
you're not a modern gamer, or rendering video, you're not gonna
fill up that video RAM anyway.

Probably so: this is another of my Black Box areas and so stuff in,
something happens, stuff out. On the -- guess could call it
'clarification' area know there are maximum resolutions so I would guess
there is some sort of upper limit. Doesn't seem to be pertinent as the
video card and monitor seem to always adjust to each other.

So if I remember correctly Mike's problem was after a while his monitor
would go black and need a reboot to get things going again. I don't
recall if he stated a time but seemed he implied weeks or months. I
want to get in on the thread because I have a similar problem with one essentially headless system which tends to not talk to a monitor after
some time -- I'm thinking around 1.5 to 2 months.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... I have a 10 foot snake.
You can't fool me, snakes don't have feet.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

BARRY MARTIN wrote:

Hi Ky!

Boo!

KM> It's the description of the fault that tells me it's most likely
KM> bad video RAM.
KM> That's quite distinctive -- you get a solid mass of random ASCII
KM> characters. We used to see that a lot in the Olden Tymes.
KM> When the card is loose, or the GPU is failing, you get random
KM> lines.

Ahhh! I don't recall the random characters part. The make-and-break of
the data stream would cause the card to try to make sense of it and output what it thought it was told.

I think it happens when the amount of usable RAM is so constrained
(probably to just the first address bank) that it can't process more
than the first few pixels, and the result is Random ASCII Characters in 25-line textmode (usually the bottom 32 bytes of the ASCII table at
that, so it's hearts and spades and such), having displayed all it
managed to scrape up.

KM> When the cable is loose, you'll lose some color (and have frex a
KM> pink screen).

Right.

Why it's usually pink, I dunno, tho I have seen blue and green, just not
as often. Probably depends on the layout of the cable head, and which
side is pulling loose, and how it sits on the board, that sort of thing.
Ie. which pins are being pulled out, more likely on the side typically
away from gravity.

Gravity always wins!

KM> This isn't absolute but it's a pretty good guideline.

Good starting points which usually work.

Usually good enough, if it's hardware. If it's software, can be any damn thing.

> Doesn't eliminate video addressing statement as a possibility. I don't
> recall how to look up; guessing "all f's" in the range end-point would
> mean it's at the far end.

KM> Addressing shouldn't cause this kind of visual screw up; it would
KM> cause a lockup or no video at all due to the conflict. And if
KM> you're not a modern gamer, or rendering video, you're not gonna
KM> fill up that video RAM anyway.

Probably so: this is another of my Black Box areas and so stuff in,
something happens, stuff out. On the -- guess could call it
'clarification' area know there are maximum resolutions so I would guess there is some sort of upper limit. Doesn't seem to be pertinent as the
video card and monitor seem to always adjust to each other.

Modern OSs probe the vidcard and the monitor, which proceeds to tell the
OS what it is capable of, producing a range of options that if wrong
will do no worse than look goofy, but with neither damage the hardware
nor leave you with a blanked-out display that needs a reinstall or CLI
magic to fix.

Modern OSs now understand this stuff, and have done so pretty reliably
for about 25 years. Part of the longstanding problem with linux was that
until about 10 years ago, or a bit longer for some of the more advanced distros, you had to know your hardware parameters and sometimes input
them manually. It seems to have finally got this right. When you're
doing it for free, as has mostly been the case, hardware programming is
not near as sexy as cute apps and games, so it gets way less attention.

So if I remember correctly Mike's problem was after a while his monitor
would go black and need a reboot to get things going again. I don't
recall if he stated a time but seemed he implied weeks or months. I
want to get in on the thread because I have a similar problem with one essentially headless system which tends to not talk to a monitor after
some time -- I'm thinking around 1.5 to 2 months.

Fedora used to do that, tho the problem seems to have gone away as of
v39. Except it would decide to not speak to the outside world after
about a week, unless regularly rebooted. Meanwhile PCLOS had been
running for several months, and WinXP for about a year and a half, with
no such issues.

As we skeap ?? speak Fedora is sitting there upgrading to v40. It's been
at it since 11pm last night, tho a lot of the time was 5GB of downloads
on a 3Mbps connection. Had to do a full update first, then run the
upgrade (fortunately the commands are still handy in the console
buffer... it started life as v32). It is now running the final step and
will be done in about an hour. You can see why I find rolling a lot less trouble.

And I get my first look at KDE Plasma v6. Given the newness thereof
(just released) for the next while there will be a lot of updates,
possibly to the point of a Whole New Monkey. But they are usually pretty
good about getting the major bugs out (not least because unlike say
Xfce, which is one guy and change, it's a team of a hundred-plus folks.)

http://www.wholenewmonkey.com/
þ RNET 2.10U: ILink: Techware BBS þ Hollywood, Ca þ www.techware2k.com

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

(Originally to Mike)

I'd guess a mix of all that! I'm sort of learned to test new stuff on a Virtual Machine because if don't like it or misconfigured easier to
blast the VM -- the uninstall, purge, etc., processes don't always
remove all traces.

Assuming you can get it to install and run in the VM. You should
hear my VM War Stories....

I don't play all that much but have had a few instances where it just
doesn't play nice. I've also been using Raspberry Pi's for a disposable machine. No, not going to toss the Pi itself, but can overwrite the SD
card and so remove all traces of the failed experiment.

Yes, the 'rolling upgrade process' can be a litte detail-filled!

"Rolling" is supposed to refer to the software, not the server
cabinet!

Only when forget to set the brakes on the cabinet's wheels!

I've come to greatly prefer a rolling distro for an everyday
desktop. There are NO UPGRADES and therefore NO REINSTALLS.
Someone took the oldest extant PCLOS from 2010 to 2021 (I think
it was) using only the rolling update process via Synaptic, and
only twice did he have to stop and twiddle something at the
command line. But normally it's just click-the-usuals, go away
for a while, and it's done.
The other advantage is that problems get fixed NOW, not when the
next major version rolls around.

That option has advantages and disadvantages. I do like the 'fixed
now', assuming it doesn't break something else. (Of course the latter
is a possibility any time.) Ubuntu does have LivePatch, which is
probably your fix-it-now.

Around here I'd prefer manual version upgrades, meaning to go from
version 22 to version 23, not the more-minor updates. Have had old
computers no longer work properly with upgrades: IMO not a fault of the
OS, though one could say it didn't check for compatibility.

The downside is you have to do your updates regularly, because it
can get out of sync if you let things go too long. (Tho Synaptic
at least is pretty good at squaring things up.) But it's fast
enough, given it's usually small, to do 'em often.

Agree: though with the limited experience I have it seems Ubuntu does a reasonable job: it seems one can take a 'basic' version (say 22.04.03),
which does the installaion and original versions, then at the end of the installation ask for the updates and it will d/l umpteen files, od the
magic, and now 22.04.39. (Making up the numbers.)

So my experience is that overall, rolling is more stable, because
it's always up to date and there are never any major upheavals.

I can see thar.

Speaking therewhich, I need to do the twice-yearly version
upgrade on the Fedora system, which I keep forgetting because
it's an overnight job (sometimes two nights) and takes some
babysitting before it gets going. Grrr.

And if you're like me the babysitting is the major part of the delay!
Good reason to have two monitors so can have one on the upgrading system
and the other for while-I'm-sitting-here-do-e-mail, etc.


¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... Q: How prevent diseases caused by biting insects?
A: Don't bite any.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)

Hi Ky!

> KM> That should be true for any hardware that "requires" some
> KM> downstream distro. Frex, if it wants Ubuntu, there's no reason it
> KM> can't run Debian... so long as it's compiled for that CPU.
> I won't disagree with Ky as he has his set of needs-wants-desires that
> differ from others. Not right, not wrong. LIS in an earlier message, I
KM> That wasn't what I said... To clarify for the distro impaired: <g> Oops! That's what I get for not being overly familiar with the stuff!

I have become painfully familiar. <g>

That was also part of your job. Same as back in the olden days when I
was selling computers and they were still more of a novelty I resisted
using some utilities even though they were clearly better than what
Windows was providing: I was sort of providing customer support and
figured I'd better be intimately familiar with what the customer's
system had.

> Is Ubuntu THE one? According to Ky no, and I'll agree with him up to a
> point. I've had my share of well-this-is-stupid-why-isn't-it-fixed.
> Ubuntu seems to be fairly well supported by having a wide variety of
> optional utilities available (and at a super-low price!!), though that
KM> Oh, they've discovered the Store. Who could have predicted....
<Trying to come up with an Economist-based reply and....> <g>

....anything you're willing to pay for, you get more of!!

If it's more expensive it has to be better! (For the seller's bank
acocunt!)

> statement is biased as I haven't really tried other OSs -- my main beef
> is when only offered for Windows and then I sometimes whine (anyone
> catch the homonym of 'whine' and 'WINE'?!).
KM> LOL. I've had zero luck with WINE. Gave up and use XP in a VM.
KM> Whine whine whine!
It's 1700 somewhere! <bseg>

<tips tricorn hat>

(Hmm: did Ky slide over to the year or did he miss the 24-hour version
of 5 o'clock somewhere?)

As for WINE, I rarely use it because I sort of forget about it. Think
the last time I used it was from DNS Benchmarks (GRC.com) and according
to my notes that was around a year ago. I mostly use 'random' Windows utilities for repairing thumbdrives -- which has a bit of a problem
because if the host system (Ubuntu in my case) doesn't detect or
properly detect it won't pass through to the VM.

Generally if I'm using linux I'm using linux, and I expect things
to behave like linux.

The problem I'm finding is some utilties only use Windows, and some
years back even the Ubuntu/Linux upper echelon said need to use Windows utilities to do stuff like repair damaged thumbdrives as there was
nothing in the Ubuntu/Linux repertoire. (I don't recall who or what
group was quoted, just helped me to figure out why I'm not finding what
I wanted: at the time didn't exist.)

But yeah, there are some things linux still doesn't do well, and
may never do well, or even at all.

Agree, and while could be an inconvenience not necessarily a bad thing.
Using my thumbdrive repair example, Linux might "never" be able to repair/recover and now (time - 2024) it may be cheaper and easier to
toss a failed thumbdrive. OTOH if something really important on it
that needs to be recovered then worth the cost to send to a recovery
service.

Frex, there is NO dedicated RTF editor for linux. I've looked.
No, LibreOffice is not satisfactory (insert rant about clean
formatting code vs printer-defined formatting and how the latter
needs to be stripped out for publication, and also to avoid
random screwups). So... XP in a VM, and my dedicated RTF editor
to the rescue.

Haven't fiddle but thinking gedit and mousepad. ... I haven't used a dot-matrix printer in years so zero experience. Only thing I do with
plain text is fiddle with a little coding.

It's a nuisance, but... better than fighting with WINE. And XP
will always speak to the whole network, which linux never will.
(Does not like random other linux boxen, never mind random
Windows. And cannot be trusted to write files without fragmenting
them all over the place.)

Could be. (Not disagreeing with you, just insufficient mackground on my
end.) I do agree it does seem odd Linux doesn't have a defregment
option -- hey: good reason to do a fresh install as that puts the files
back together! <g>

I'll admit to having problems 'seeing' other computers around here but
the correction usually was to fix the ssh connection and then Remmina or TigerVNC would run correctly. Both of those do have some sort of a
reliance on the host system having a visible/usable screen: as was
mentioned somewhere in the earlier levels of this thread, I have (and I
think Mike also or similar) issue with the host system's monitor not connecting. Seems like when that occurs rare to get to reconnect the
video.

¯ ®
¯ BarryMartin3@MyMetronet.NET ®
¯ ®


... After things go from bad to worse, the cycle will repeat itself.
--- MultiMail/Win32 v0.47
þ wcECHO 4.2 ÷ ILink: The Safe BBS þ Bettendorf, IA

--- QScan/PCB v1.20a / 01-0462
* Origin: ILink: CFBBS | cfbbs.no-ip.com | 856-933-7096 (454:1/1)