Re: Xorg fails, no gui: systemd issue?

[tomas]

On Sun, Dec 10, 2023 at 12:57:59AM +0100, Anders Andersson wrote:
> On Sat, Dec 9, 2023 at 3:01 AM Andrew M.A. Cater  wrote:
> >
> > On Fri, Dec 08, 2023 at 06:13:59PM +, John - wrote:
> > > Since I last (3 December) upgraded the software (sid)  on my old 
> > > Thinkpad, my gui fails to come up. The last line of /var/log/Xorg.0.log 
> > > reads:
> > > (EE) systemd-login: failed to take device /dev/dri/card0: Message 
> > > recipient disconnected from message bus without replying
> > > I've  been trying for weeks to fix it, including tracking down all 
> > > suggests from googling the error message, without success. Can anyone 
> > > help me figure out what the problem is?
> > > Thanks.
> > >
> >
> > Hi John,
> >
> > Today is the 8th of December - strictly, that's barely a business week.
> >
> > Debian expressly comes with no guarantees. You are running sid a.k.a
> > unstable - that comes with still fewer guarantees other than breakage
> > from time to time.
> 
> I find this quite rude. Nothing in OPs post suggests that he demands
> or expects any help, yet you're quick to point out there are no
> guarantees. So? Who asked about guarantees?

It wasn't rude. Perhaps a bit unfortunate (because Debian, after all,
does come with some guarantees. The Social Contract, among others).

Andrew was talking specifically about *sid*. Sid does break from time
to time. This is to be expected, by the way it works.

> Aren't you even allowed to ask for help if you run sid? Where else to
> ask other debian users for pointers than the debian user list? Is
> there a specific list for sid users only?

I think Andrew's answer was more towards "in sid, things tend to fix
themselves, but this takes a bit of time".

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: mpv "some-gif-file.gif" fullscreen and quit, auto exits/kills wayland/ "logs out"

[Zenaan Harkness]

Some years ago, I ran sid for a few years. Only recall one Xorg issue
where I needed low level deb os surgery during that time.

Is sid still similarly stable, or less so in the wayland era (for
anyone using wayland/mutter)?

A benefit of using sid is that I am using software versions (e.g mpv
and mutter, gnome-desktop etc) that are much closer to current dev
versions, therefore my bug pursuit is more useful...

Feeling a bit like a raft in the ocean right now.

Re: mpv "some-gif-file.gif" fullscreen and quit, auto exits/kills wayland/ "logs out"

[Zenaan Harkness]

> Can reproduce on intel free driver too. Reproduced with .gif, .webm and
> .mp4.
>
> For reference, bug reported just now:
>
> https://github.com/intel/media-driver/issues/1743


Anyone know how I can test further, or should I report this against mutter?

mpv bug closed:
https://github.com/mpv-player/mpv/issues/13062

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057119
should probably be closed too I guess

but my desktop keeps crashing, so it's a real bug.

Any hints from anyone at the best approach at this point would be
really appreciated (testing?, report against mutter?)

Thanks in advance

Re: Xorg fails, no gui: systemd issue?

[Andy Smith]

Hello,

On Sun, Dec 10, 2023 at 12:57:59AM +0100, Anders Andersson wrote:
> On Sat, Dec 9, 2023 at 3:01 AM Andrew M.A. Cater  wrote:
> > Today is the 8th of December - strictly, that's barely a business week.
> >
> > Debian expressly comes with no guarantees. You are running sid a.k.a
> > unstable - that comes with still fewer guarantees other than breakage
> > from time to time.
> 
> I find this quite rude. Nothing in OPs post suggests that he demands
> or expects any help, yet you're quick to point out there are no
> guarantees. So? Who asked about guarantees?

It is difficult to tactfully suggest that someone may not be expert
enough to do something, and I would class trying to do your daily
work on Debian sid as an expert endeavour.

I took Andrew's mention of the date not as an admonishment that OP
was expecting too much of Debian developers, but rather as a reality
check that sid does sometimes enter a time of severe brokenness that
can last weeks, as a normal state of affairs that no one feels any
particular urgency to rectify, because that is the nature of the
thing. It is explicitly for developers.

> > You are expected to be able to fix breakage in sid yourself or
> > you get to keep both pieces :)
> >
> > You may find that the issue has been fixed if you update today: you may not.
> > There's not much there in logs to help any of the rest of us who don't
> > habitually run sid.
> 
> Aren't you even allowed to ask for help if you run sid?

"Allowed" is too strong a word, but sid is for developers and the
level of questions being asked here together with the level of
information supplied strongly suggest that OP is not a developer.
Questions about sid's status and issues within sid as discussed
between Debian developers tend to look very different. OP's post
looks like a typical end user support query, and that's not really
going to work for users of sid.

> Where else to ask other debian users for pointers than the debian
> user list? Is there a specific list for sid users only?

Probably debian-devel, or the team list for the specific package that
is thought to be having issues, or the bugs database for same. That
is, it really needs to be a discussion between developers, not a
user support conversation, so it's not really a "Debian user"
matter, except in the sense that any developer of Debian is also a
user of Debian.

If it's something that is broken in sid but not in a released
version of Debian then it can be difficult for debian-user to
assist.

> > This is explicitly *not* a sarcastic suggestion: if you can't run sid,
> > then I would suggest you reformat your disks and install Debian stable.
> > Most of the people either active on this list or lurking and reading on
> > the sidelines run Debian stable for a reason.
> 
> It does not come of as sarcastic, just condescending, as if OP does
> not already know what sid is.

I suspect from the toner of OP's message that they really don't
know what sid is. Again, it's tricky to politely suggest that
someone might be using the wrong tool for the job.

> > unless you are actively interested in testing and fixing
> > breakage as it occurs, there is little justification for running
> > sid as a daily operating system.
> >
> > Sid is explicitly *not* a chance to run the latest, greatest bleeding edge
> > software reliably on a sustained basis without the occasional crash or
> > significant problems.
> 
> Again, OP never claimed anything different. Something broke, as
> expected. OP asks for help trying to fix it.

I think it's a useful reminder of the purpose of sid and what to
expect.

> Can't you simply let people who want to help answer? If no one
> knows, then fine, OP is on his own.

I did not see Andrew telling people not to help OP, so if anyone
can, they are still able to. But user support for sid is more
difficult.

There are definitely people out there who mistakenly see sid as
"Debian, but with newer packages" and it seems possible that OP is
one of those people. If Andrew's response educates any such person
then I think it was worth it.

Much like with X/Y problems, where we must suggest that the user
needs to step back and explain what they are actually trying to
achieve, communicating that you think someone is using the wrong
tool—and why—is fraught with social pitfalls.

Thanks,
Andy

-- 
https://bitfolk.com/ -- No-nonsense VPS hosting

Re: Xorg fails, no gui: systemd issue?

[Anders Andersson]

On Sat, Dec 9, 2023 at 3:01 AM Andrew M.A. Cater  wrote:
>
> On Fri, Dec 08, 2023 at 06:13:59PM +, John - wrote:
> > Since I last (3 December) upgraded the software (sid)  on my old Thinkpad, 
> > my gui fails to come up. The last line of /var/log/Xorg.0.log reads:
> > (EE) systemd-login: failed to take device /dev/dri/card0: Message recipient 
> > disconnected from message bus without replying
> > I've  been trying for weeks to fix it, including tracking down all suggests 
> > from googling the error message, without success. Can anyone help me figure 
> > out what the problem is?
> > Thanks.
> >
>
> Hi John,
>
> Today is the 8th of December - strictly, that's barely a business week.
>
> Debian expressly comes with no guarantees. You are running sid a.k.a
> unstable - that comes with still fewer guarantees other than breakage
> from time to time.

I find this quite rude. Nothing in OPs post suggests that he demands
or expects any help, yet you're quick to point out there are no
guarantees. So? Who asked about guarantees?


> You are expected to be able to fix breakage in sid
> yourself or you get to keep both pieces :)
>
> You may find that the issue has been fixed if you update today: you may not.
> There's not much there in logs to help any of the rest of us who don't
> habitually run sid.

Aren't you even allowed to ask for help if you run sid? Where else to
ask other debian users for pointers than the debian user list? Is
there a specific list for sid users only?


> This is explicitly *not* a sarcastic suggestion: if you can't run sid,
> then I would suggest you reformat your disks and install Debian stable.
> Most of the people either active on this list or lurking and reading on
> the sidelines run Debian stable for a reason.

It does not come of as sarcastic, just condescending, as if OP does
not already know what sid is.


> If you are a Debian maintainer, you are expected to build new software in
> unstable for it to propagate to testing and (eventually) to the next Debian
> major release. Outside that, unless you are actively interested in testing
> and fixing breakage as it occurs, there is little justification for running
> sid as a daily operating system.
>
> Sid is explicitly *not* a chance to run the latest, greatest bleeding edge
> software reliably on a sustained basis without the occasional crash or
> significant problems.

Again, OP never claimed anything different. Something broke, as
expected. OP asks for help trying to fix it. Can't you simply let
people who want to help answer? If no one knows, then fine, OP is on
his own.

CUPS classes wrecking printing

[Gary Dale]

I've running Debian/Bookworm (stable) on an AMD64 system - a laptop. 
It's a fresh install of Debian from about 6 months back that has been 
kept up to date.


Each December I am involved in an event that requires me to use 3 
photo-printers to print a lot of 4x6 photos. It takes 2 or 3 printers to 
get the throughput so people aren't waiting for their photos.


I've been doing this for a decade using various photo printers. I've 
always just set up a CUPS "photo" class and added the printers to it. 
Then I'd use lpr -P photo  to send the 
output to whichever printer was free. I even did it last year using the 
same laptop and printers and things worked.


This year, because it was a new OS install, I had to connect the 
printers and install them again. This required the gutenprint drivers 
for two of the printers while the newest seems to work "driverless".


All the printers were tested individually and printed the CUPS test page 
perfectly.


However when I sent something to the "photo" class, whichever printer 
received the job just printed a page of bands of colour. I could send a 
picture to an individual printer OK but not send it to the "photo" class.


I got through the event by skipping the lpr -P photo... command and 
manually selecting a printer from Gwenview when I was viewing the 
picture earlier in the workflow (to verify it was worth printing). This 
was not ideal and I only got through it because this year's event was 
less than half its usual size.


This was not an lpr problem because I also couldn't print to the class 
from Gwenview. CUPS classes seem to be broken.

Re: ToG Linux (first draft of a RFC) ...

[Albretch Mueller]

On 12/9/23, Albretch Mueller  wrote:
> the CIA was giving money to Ukrainian people but in
> order to get it they had to use their cell phones ;-)

 which (cell phones) they would also get "for free", mind you.

 And well ..., yes, even if you remove the networking hard and
software, all RF devices need are electrons moving around in ways you
can control and encode. Some time ago, I heard the expected news that
they had managed to sandwich a RF circuitry on the layers of a chip!
So, expect for everybody to start making their own chips at least for
their most critical infrastructure!

 There will however always be physical ways to hack a hack. Imagine a
bugged cell phone, you could always physically extend the
functionality you need and keep the cell phone itself encased in a
Faraday cage at all times.

 lbrtchx

Re: Apt 2 not fully installed or removed.

[Timothy M Butterworth]

On Sat, Dec 9, 2023 at 9:07 AM Andy Smith  wrote:

> Hello,
>
> On Sat, Dec 09, 2023 at 02:26:41AM -0500, Timothy M Butterworth wrote:
> > I figured out the problem and apt is working fine now.
>
> It would be nice for you to elaborate, both to satisfy my curiosity
> 😀 and to provide a search engine hit for anyone encountering this
> in future.
>
> Thanks,
> Andy
>
> --
> https://bitfolk.com/ -- No-nonsense VPS hosting
>
> Andy,

I deleted the files /var/lib/dpkg/info/php-horde.postrm and
/var/lib/dpkg/info/libapache2-mod-fcgid.postinst. Then I did a sudo apt
remove php-horde and sudo apt install libapache2-mod-fcgid.

I modified /etc/apache2/envvars setting /etc/default/locale and commented
out export LANG=C. This was to make NextCloud work, with the required
en_us-UTF8 character set. I am not sure why php-horde was getting stuck on
it.

Tim



-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
⠈⠳⣄⠀⠀

Re: IMPORTANT: do NOT upgrade to new stable point release

[Joe]

On Sat, 9 Dec 2023 19:18:20 +
Michael Kjörling <2695bd53d...@ewoof.net> wrote:

> On 9 Dec 2023 20:54 +0200, from ale...@nanoid.net (Alexis Grigoriou):
> > I just upgraded to Bookworm this morning. I did reboot a couple of
> > times but there seems to be no problem (yet). Is there anything I
> > should look for or do other than rebooting?  
> 
> If you upgraded this morning, then I would expect that you are okay
> for now.
> 
> Per #5 in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
> the bug is present in kernel Debian package version 6.1.64-1. If you
> are on 6.1.55-1 (current Bookworm stable per last night) you _likely_
> aren't affected. If you are on 6.1.55-1 (or earlier), just hold off on
> upgrades for now; and if you need to upgrade something else, take
> great care for now to ensure that no Linux kernel packages get
> upgraded to any version < 6.1.66, and preferably not < 6.1.66-1.
> 
> For versions, check:
> 
> * uname -v
> * dpkg -l linux-image-\*
> 
> In that bug report thread, #21 lists 6.1.66 as fixed upstream, and #28
> indicates that 6.1.66-1 includes the fix from upstream, and that it is
> being published.
> 

It appears from the link in the bug report that 6.5.x kernels
(sid/trixie) are not affected. Does anyone know otherwise?

https://lore.kernel.org/stable/20231205122122.dfhhoaswsfscuhc3@quack3/

Is the bug likely to affect all architectures? I have a Pi bookworm
(armhf) on 6.1.63-1, with 6.1.58-1 also installed. I can probably roll
back to 6.1.54-1 if necessary.

-- 
Joe

Re: IMPORTANT: do NOT upgrade to new stable point release

[Michael Kjörling]

On 9 Dec 2023 14:26 -0500, from g...@wooledge.org (Greg Wooledge):
>> If you upgraded this morning, then I would expect that you are okay
>> for now.
> 
> That doesn't appear to be true.
> 
>> Per #5 in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
>> the bug is present in kernel Debian package version 6.1.64-1. If you
>> are on 6.1.55-1 (current Bookworm stable per last night) you _likely_
>> aren't affected.
> 
> This is the kernel I got this morning:
> 
> ii  linux-image-6.1.0-14-amd64 6.1.64-1 amd64Linux 6.1 for 64-bit 
> PCs (signed)

Alexis Grigoriou is in UTC+0200 per the email headers; you are in
UTC-0500. Thus "this morning" is 7 hour later for you than for Alexis,
only because of that.

-- 
Michael Kjörling 🔗 https://michael.kjorling.se
“Remember when, on the Internet, nobody cared that you were a dog?”

Re: IMPORTANT: do NOT upgrade to new stable point release

[Richmond]

Greg Wooledge  writes:

> On Sat, Dec 09, 2023 at 07:18:20PM +, Michael Kjörling wrote:
>> If you upgraded this morning, then I would expect that you are okay
>> for now.
>
> That doesn't appear to be true.
>
>> Per #5 in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
>> the bug is present in kernel Debian package version 6.1.64-1. If you
>> are on 6.1.55-1 (current Bookworm stable per last night) you _likely_
>> aren't affected.
>
> This is the kernel I got this morning:
>
> ii linux-image-6.1.0-14-amd64 6.1.64-1 amd64 Linux 6.1 for 64-bit PCs
> (signed)
>
> This is the current result of looking for a newer one:
>
> 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
>
> Based on the warnings given here, I rebooted to the prior kernel:
>
> unicorn:~$ uname -a Linux unicorn 6.1.0-13-amd64 #1 SMP
> PREEMPT_DYNAMIC Debian 6.1.55-1 (2023-09-29) x86_64 GNU/Linux
>
> I guess I'll wait and see what happens next.

I upgraded some time today to a December kernel. I have now gone back to
September 29 kernel. But is there a way to tell what if anything got
corrupted? I am using a 32 bit system and ext4.

I booted this:

 6.1.0-13-686-pae #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1 (2023-09-29)
 i686 GNU/Linux

then:

aptitude remove linux-image-6.1.0-14-686-pae

File systems mounted under `/media/root/` ?

[Stefan Monnier]

Recently I noticed some unused ext4 filesystems (i.e. filesystems that
aren't in /etc/fstab, that I normally don't mount, typically because
they're snapshots or backups) "magically" mounted as
`/media/root/`.

This is on a headless ARM board running Debian stable.

Not sure when this happen, but I noticed t least once happening in
response to `vgchange -ay`.

Any idea who/what does that, and how/where I can control it?


Stefan

Re: IMPORTANT: do NOT upgrade to new stable point release

[Greg Wooledge]

On Sat, Dec 09, 2023 at 07:18:20PM +, Michael Kjörling wrote:
> If you upgraded this morning, then I would expect that you are okay
> for now.

That doesn't appear to be true.

> Per #5 in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
> the bug is present in kernel Debian package version 6.1.64-1. If you
> are on 6.1.55-1 (current Bookworm stable per last night) you _likely_
> aren't affected.

This is the kernel I got this morning:

ii  linux-image-6.1.0-14-amd64 6.1.64-1 amd64Linux 6.1 for 64-bit 
PCs (signed)

This is the current result of looking for a newer one:

0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

Based on the warnings given here, I rebooted to the prior kernel:

unicorn:~$ uname -a
Linux unicorn 6.1.0-13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1 
(2023-09-29) x86_64 GNU/Linux

I guess I'll wait and see what happens next.

Re: IMPORTANT: do NOT upgrade to new stable point release

[Michael Kjörling]

On 9 Dec 2023 20:54 +0200, from ale...@nanoid.net (Alexis Grigoriou):
> I just upgraded to Bookworm this morning. I did reboot a couple of
> times but there seems to be no problem (yet). Is there anything I
> should look for or do other than rebooting?

If you upgraded this morning, then I would expect that you are okay
for now.

Per #5 in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
the bug is present in kernel Debian package version 6.1.64-1. If you
are on 6.1.55-1 (current Bookworm stable per last night) you _likely_
aren't affected. If you are on 6.1.55-1 (or earlier), just hold off on
upgrades for now; and if you need to upgrade something else, take
great care for now to ensure that no Linux kernel packages get
upgraded to any version < 6.1.66, and preferably not < 6.1.66-1.

For versions, check:

* uname -v
* dpkg -l linux-image-\*

In that bug report thread, #21 lists 6.1.66 as fixed upstream, and #28
indicates that 6.1.66-1 includes the fix from upstream, and that it is
being published.

-- 
Michael Kjörling 🔗 https://michael.kjorling.se
“Remember when, on the Internet, nobody cared that you were a dog?”

Re: ToG Linux (first draft of a RFC) ...

[Albretch Mueller]

On 12/9/23, Greg Wooledge  wrote:
>> On 09.12.23 at 10:13, Albretch Mueller wrote:
>> >   As anyone could see you could even run a network of detached
>> > computers without networking interfaces in a "touch of God" kind of
>> > way ...

 Thank you. I should have more clearly stated that those computing
devices would go each about their particular business situationally in
an air-gapped mode to then each go about their "collective
intelligence" on a separate computer, a "server" of sorts.

> What I don't understand is what these computers would be *doing*.  Why
> does he need them at all?  If he needs them, why does he need them to
> be detached from each other and from the rest of the world?

 These days, everything from microwave ovens to pacemakers are
computing devices, but why should they be wifi-enabled? People don't
seem to even realize that since the 1990's they have been driving
computers on wheels. Then you hear that Vladimir Putin assassinated
Michael Hastings for saying the same Joe Biden said only two decades
before; you hear targeted individuals talking about of smoke rings
(something that can't happen in nature by itself) to then hear about
COVID-19 and how it was all started in China by some dissident
"freedom-loving" bats, ... I even heard once as part of those marginal
comments you hear which make you go like, say what?, that courtesy of
U.S. tax payers the CIA was giving money to Ukrainian people but in
order to get it they had to use their cell phones ;-)

 Think monitoring devices in hospitals, schools, power plants, ... I
once heard that some "intelligence department" knew the grounding
truth about some matter which happened in some remote place in Russia,
because they had been monitoring the cell phones of not only "we the
people", but even the police and, of course, when you hear such thing,
since neither Physics nor "God" have "blue-eyed sons" (contrary to
what some Israelis/Jewish people may think) or as they say "what is
good for the goose is good for the gander", that also means that "the
Chinese", "Russians", ... are able to do the same thing (of course, in
their case they do it "because they hate freedom"). Now, imagine that
at least the police would use a "ToG network" (to call it something)
without any networking capabilities on a hard- and software level
(just the necessary functionality, for example, passively getting GPS
coordinates for which you don't need the whole networking stack) for
their computing devices with a One-Time pad lease for the session they
will be using it (and they would be physically powered for the amount
of time they need to be used) which they must relinquish after each
day of work to their base and which data would be encrypted (reusing
the initial OTP) in ways that only the server which leased the OTP
would be able to decrypt. Think of how they hacked the enigma machine
and how Nazis suspected such a thing to have happened:

 https://en.wikipedia.org/wiki/Erhard_Maertens

 What could those "freedom hating" Chinese, "what the heck is freedom"
Russians, all those "intelligence departments", ... do about any of
it?

> The only things I can imagine are ...

> Whatever he's doing, I'm confident he won't tell us, or at least not in
> a way I'll be able to understand.

 I explained the basic idea to a friend over the phone and clarified
to him a number of doubts he had about it. He got it. He also realized
that it wasn't much of a hassle, really. And he told me that I come up
with such ideas because I have been forced to look at reality from a
different point of view. In a Hegelian, karmic way it is a good thing
that people looking at reality from different vantage points can talk
to one another, even though, as they say, "misunderstanding is as
mutual as love should be".  No XY problem whatsoever, and I am not
trying to hide anything from anyone (whatever "hiding", "privacy", ...
could possibly mean these days). As Mike Rogers (of Pink Floyd fame)
said when he was being accused of being a racist, anti-zionist or a
zionist, (or whatever he was accused of), ...: "if I am, at the very
least, I should be conscious of it".

 Keeping an external drive you never connect to the Internet could be
understood as a sneakernet aspect of it.

 lbrtchx

Re: IMPORTANT: do NOT upgrade to new stable point release

[Alexis Grigoriou]

On Sat, 2023-12-09 at 13:09 -0500, Dan Ritter wrote:
> 
> 
> The new kernel release is reported to contain an ext4 data
> corruption bug. It's prudent not to upgrade, or if you have
> started to upgrade, not to reboot, until a new kernel release
> is prepared.
> 

I just upgraded to Bookworm this morning. I did reboot a couple of
times but there seems to be no problem (yet). Is there anything I
should look for or do other than rebooting?

Re: Xorg fails, no gui: systemd issue?

[John -]

 Thanks for answering and thanks for the advice, most of which I agree with, 
since I've been running sd since 2006.Since I am now 84, I'm not as good at 
figuring things out as I used to be, so if anyone else can offer help, I'd be 
grateful.
On Friday, December 8, 2023 at 03:36:19 PM EST, Andrew M.A. Cater 
 wrote:  
 
 On Fri, Dec 08, 2023 at 06:13:59PM +, John - wrote:
> Since I last (3 December) upgraded the software (sid)  on my old Thinkpad, my 
> gui fails to come up. The last line of /var/log/Xorg.0.log reads:
> (EE) systemd-login: failed to take device /dev/dri/card0: Message recipient 
> disconnected from message bus without replying
> I've  been trying for weeks to fix it, including tracking down all suggests 
> from googling the error message, without success. Can anyone help me figure 
> out what the problem is?
> Thanks.
>

Hi John,

Today is the 8th of December - strictly, that's barely a business week.

Debian expressly comes with no guarantees. You are running sid a.k.a
unstable - that comes with still fewer guarantees other than breakage
from time to time.

You are expected to be able to fix breakage in sid
yourself or you get to keep both pieces :) 

You may find that the issue has been fixed if you update today: you may not.
There's not much there in logs to help any of the rest of us who don't 
habitually run sid.

This is explicitly *not* a sarcastic suggestion: if you can't run sid,
then I would suggest you reformat your disks and install Debian stable.
Most of the people either active on this list or lurking and reading on
the sidelines run Debian stable for a reason.

If you are a Debian maintainer, you are expected to build new software in
unstable for it to propagate to testing and (eventually) to the next Debian
major release. Outside that, unless you are actively interested in testing
and fixing breakage as it occurs, there is little justification for running
sid as a daily operating system.

Sid is explicitly *not* a chance to run the latest, greatest bleeding edge
software reliably on a sustained basis without the occasional crash or 
significant problems.

With every good wish, as ever,

Andy

(amaca...@debian.org)

  

Re: IMPORTANT: do NOT upgrade to new stable point release

[Markus Schönhaber]

09.12.23, 19:09 +0100, Dan Ritter:

> https://fulda.social/@Ganneff/111551628003050712
> 
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843
> 
> The new kernel release is reported to contain an ext4 data
> corruption bug. It's prudent not to upgrade, or if you have
> started to upgrade, not to reboot, until a new kernel release
> is prepared.

Thank you very much for the hint, Dan!

-- 
Regards
  mks

Re: IMPORTANT: do NOT upgrade to new stable point release

[Nate Bargmann]

Thanks for the tip.  I updated this morning well before any
announcements and having seen this I rebooted into the 6.1.0-12 (6.1.52)
package.  Good thing old kernels are kept around.

- Nate

-- 
"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."
Web: https://www.n0nb.us
Projects: https://github.com/N0NB
GPG fingerprint: 82D6 4F6B 0E67 CD41 F689 BBA6 FB2C 5130 D55A 8819



signature.asc
Description: PGP signature

IMPORTANT: do NOT upgrade to new stable point release

[Dan Ritter]

https://fulda.social/@Ganneff/111551628003050712

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057843

The new kernel release is reported to contain an ext4 data
corruption bug. It's prudent not to upgrade, or if you have
started to upgrade, not to reboot, until a new kernel release
is prepared.


-dsr-

Re: Could not find interfaces configuration file /etc/network/interfaces in Debian Linux 11 (bullseye)

[Turritopsis Dohrnii Teo En Ming]

On Friday, December 8th, 2023 at 11:23 PM, John Hasler  
wrote:


> Mr. Turritopsis Dohrnii Teo En Ming writes:
> 
> > You managed to install OpenWRT on an Ubiquiti router?
> 
> 
> Yes. It was quite straightforward. Instructions on the OpenWRT site.
> --
> John Hasler
> j...@sugarbit.com
> Elmwood, WI USA

I have seen the web UI of OpenWRT. Quite impressive.

Regards,

Mr. Turritopsis Dohrnii Teo En Ming
Targeted Individual in Singapore

Re: ToG Linux (first draft of a RFC) ...

[John Hasler]

Greg writes:
> Is he simply talking about sneakernet?  A human administrator, whom I
> imagine to be the "god" in this scenario, walks around and room and
> types things on each computer as needed?

Carrying removable media around.
-- 
John Hasler 
j...@sugarbit.com
Elmwood, WI USA

Re: ToG Linux (first draft of a RFC) ...

[John Hasler]

Arno writes:
> At this point it becomes quite clear that we have a misunderstanding
> at a very low level. Sentences like "run a network of ... computers
> without networking interfaces" are something I can not really grasp
> with the facilities I have.

You could run a slow network by mailing removable media around.  In the
early days Australia was on Usenet by way of airmailed taps.  Then
there's https://www.rfc-editor.org/rfc/rfc2549.

Though consider: the earliest computer viruses were transmitted by
floppy disk...
-- 
John Hasler 
j...@sugarbit.com
Elmwood, WI USA

Re: ToG Linux (first draft of a RFC) ...

[Greg Wooledge]

On Sat, Dec 09, 2023 at 02:50:16PM +0100, Arno Lehmann wrote:
> On 09.12.23 at 10:13, Albretch Mueller wrote:
> >   As anyone could see you could even run a network of detached
> > computers without networking interfaces in a "touch of God" kind of
> > way,
> 
> At this point it becomes quite clear that we have a misunderstanding at a
> very low level. Sentences like "run a network of ... computers without
> networking interfaces" are something I can not really grasp with the
> facilities I have.

Is he simply talking about sneakernet?  A human administrator, whom I
imagine to be the "god" in this scenario, walks around and room and
types things on each computer as needed?

What I don't understand is what these computers would be *doing*.  Why
does he need them at all?  If he needs them, why does he need them to
be detached from each other and from the rest of the world?

The only things I can imagine are:

 * Calculating something that takes a long time to calculate.  Maybe the
   problem can be trivially parallelized, in such a way that he can
   type the necessary parameters for each piece of the calculation on
   each node.  Obviously it would be better if the nodes were networked
   to each other, instead of requiring manual collation of the results,
   but we've already established that the OP is insane.

 * Tools in the creation of some kind of work of art (visual, musical,
   etc.).  A computer runs whatever software is used in this creative
   endeavor.  At the end, a file is created, and this is copied onto
   removable media, which is then sent to a publisher.  I don't see
   why he would need multiple computers in this scenario, unless he's
   got many projects going on simultaneously, and he wants one computer
   dedicated to each project.

Whatever he's doing, I'm confident he won't tell us, or at least not in
a way I'll be able to understand.

Re: debian forgot usr pw

[Pocket]

On 12/9/23 01:29, Timothy M Butterworth wrote:



On Fri, Dec 8, 2023 at 7:56 AM Pocket  wrote:


On 12/8/23 00:05, John Hasler wrote:
> Gene writes:
>> AND (horrors) have written it down.
> That's the right thing to do.

Well you could always use the universal password of password

I use for example i use the following

for the root account the password is root

for my user account of pocket the password is pocket

No one will every guess those password so I am completely protected


Thanks now I have your passwords and your IP address from the SMTP 
header. 2001:41b8:202:deb:216:36ff:fe40:4002 Darn SSH is configured to 
use certificates. Your security is stronger than you let on.


LOL

whois 2001:41b8:202:deb:216:36ff:fe40:4002
% This is the RIPE Database query service.
% The objects are in RPSL format.
%
% The RIPE Database is subject to Terms and Conditions.
% See https://apps.db.ripe.net/docs/HTML-Terms-And-Conditions

% Note: this output has been filtered.
%   To receive output for a database update, use the "-B" flag.

% Information related to '2001:41b8:202::/48'

% Abuse contact for '2001:41b8:202::/48' is 'ab...@man-da.de'

inet6num:   2001:41b8:202::/48
netname:    DE-MANDA-DEBIAN-V6-01
descr:  Debian Darmstadt Network
country:    DE
admin-c:    DSAT1-RIPE
tech-c: MAND2-RIPE
status: ASSIGNED
mnt-by: MANDA-MNT
created:    2015-07-09T09:24:37Z
last-modified:  2015-07-09T09:24:37Z
source: RIPE # Filtered

role:   Debian System Administrators Team
address:    Software in the Public Interest, Inc
address:    Debian Project
address:    1732 1st Ave #20327
address:    New York, NY 10128-5177
address:    United States
org:    ORG-DA330-RIPE
remarks:    
remarks:    **  in case of emergency find us on IRC   **
remarks:    **  irc://irc.debian.org/#debian-admin    **
remarks:    
remarks:    **
remarks:    
remarks:    **   Direct peering requests to   **
remarks:    ** peer...@debian.org  **
remarks:    

That doesn't belong to me





-- 


It's not easy to be me



--
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
⠈⠳⣄⠀⠀


--
It's not easy to be me

Re: ToG Linux (first draft of a RFC) ...

[Arno Lehmann]

Hello,

On 09.12.23 at 10:13, Albretch Mueller wrote:

On 12/7/23, Arno Lehmann  wrote:

it's quite interesting that you use a platform such as wordpress,
running code you can not control, to discuss such matters.


  I was just brainstorming, dumping a stream of consciousness with a
relatively comprehensive outline of the main ideas.


Your paranoia needs an adjustment, because the above is what would make 
you targetable.


...

- You can not use the same hardware air gapped and non air gapped.


  I beg to differ and at the end of the day this is something that can
be physically/technically proved.


It has been proven.

...

  Well, not really! Booting a Debian Live DVD doesn't take more time
than booting Windows (from scratch) and the whole idea of using a
package extensions USB pen drive would automate updates. This
basically is all there is to maintaining it.


No.


You would be basically
making use of the BIOS and RAM of a computer


You can not trust those.

...

  As anyone could see you could even run a network of detached
computers without networking interfaces in a "touch of God" kind of
way,



At this point it becomes quite clear that we have a misunderstanding at 
a very low level. Sentences like "run a network of ... computers without 
networking interfaces" are something I can not really grasp with the 
facilities I have.


Cheers,

Arno

--
Arno Lehmann

IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück

Re: Apt 2 not fully installed or removed.

[Greg Wooledge]

On Sat, Dec 09, 2023 at 02:26:41AM -0500, Timothy M Butterworth wrote:
> > Removing php-horde (5.2.23+debian0-6) ...
> > /var/lib/dpkg/info/php-horde.postrm: 28: /etc/apache2/envvars:
> > /etc/default/locale: Permission denied

> I figured out the problem and apt is working fine now.

Was it the permissions on the /etc/default/locale file by any chance?
Or maybe the /etc/default directory?

Re: ToG Linux (first draft of a RFC) ...

[Albretch Mueller]

On 12/9/23, Albretch Mueller  wrote:
>  As anyone could see you could even run a network of detached
> computers without networking interfaces in a "touch of God" kind of
> way, some sort of "leased One-time pad touches of God" specifically
> for each, all coordinated through and which data/information would end
> up in a kind of "server", you could even use cell phones to do such
> thing ...

 I could even envision industries around such specifications. The only
reason why such things haven’t happened (would not ever happen?) is
because police, politicians and IT companies (which these days are all
the same) want to run society as if we were all rats in a maze they
control real time in a predictive and cross correlating way in which
everything is ephemeral to "we the people" while they keep a click by
click, keystroke by keystroke, breath by breath, ...  data
Doppelgänger of each of us.

 Something "my paranoia" noticed as part of the Snowden revelations
which IMO hasn’t been aired, questioned, discussed enough was that the
NSA, as part of their "all tangible things" doctrine, was most
interested in people’s medical records. Why?!? Why would "they" care
about people’s health? Isn’t the safety and betterment of society what
they should be interested in? To top it all "we the people" didn’t get
the extent to which they were making fun of us when they said that:
"what matters is how we use that information, not that we collect
it"!!!

 This is how the world we are living in looks like:

// __ 36C3 - The sustainability of safety, security and privacy:

 https://www.youtube.com/watch?v=2m5EMkVTydI

 (7:35) Internet of things or -Internet of Targets-
 (8:15) just as a car has got about 50 computers in it
 (11:20) hospital safety usability failures kill about 2000 people a
year in the UK, just about car accidents (and he was just talking
about impedance in the GUIs! not even about "errare humanum est")
 (12:30) some dude manage to gain access to 450,000 active pace makers over wifi
 (15:20) modern cars have about 10 radio frequency interfaces
 (23:20) initially light bulbs could be on for more than a century,
... these days companies make it from almost impossible to illegal to
fix things in order to make money
 ...
~
 At some point it all became so unbelievable, out of it all weird that
I had to take as some sarcastic theatrics. Like when he showed
hospital rooms and the number of network-enabled, RF devices in it.

 Since things happen for a reason, as part of explaining why Anderson
could have at the very least asked why this is all happening)

 lbrtchx

Re: ntpsec as server questions

[Max Nikulin]

On 08/12/2023 23:12, Bonno Bloksma wrote:

So, in pseudo code
   bool isleapyear (int year) {
   return false;


I've heard such a calendar was in use in ancient Egypt.
https://en.wikipedia.org/wiki/Sothic_cycle
Its disadvantage was that crop reaping and tax paying dates were slowly 
becoming out of sync. However every 1461 years it was returning to 
originally established order.

Re: Hardware TOTP on Linux

[Andre Rodier]

On Fri, 2023-12-08 at 19:11 -0500, Jeffrey Walton wrote:
> On Wed, Dec 6, 2023 at 7:36 AM Andre Rodier  wrote:
> > 
> > First, if this post is off-topic, feel free to give me other
> > mailing lists.
> 
> The eevblog may be another place to ask. But be warned, the folks on
> the eevblog can get really deep into the weeds on subjects like this.
> If you are not careful, they might provide you with circuit designs
> and have you build your own hardware.
> 
> Also see  and
> .
> 
> Jeff

Thanks, Jeff. If circuit design is the solution, why not ? It would be
fun.

Thanks again,
André

Re: ToG Linux (first draft of a RFC) ...

[Albretch Mueller]

On 12/7/23, Arno Lehmann  wrote:
> it's quite interesting that you use a platform such as wordpress,
> running code you can not control, to discuss such matters.

 I was just brainstorming, dumping a stream of consciousness with a
relatively comprehensive outline of the main ideas.

> Wouldn't it be more reasonable to self host

 That will definitely happen at some point. I will have to test first
the Linux initialization process (it’s runlevels) and how to make it
dance together with GRUB nicely (no mysteries whatsoever there).

> ... using a hoster providing
> decent privacy and aonymity or a technology such as Tor?

 I am not into protagonism and that is not my main line of research,
occupation. I would like to culture (invite more like-minded people to
own) that open source project. Anyone could take over hosting it (I
would pay for the first two years) and anything we do we would openly
(well, almost! ;-)). As they say: true security, privacy, ... can not
be hidden. All we do and say we would to the four winds. Once it is
vetted we could even ask nicely for it to be included as part of the
Debian or some other hosting.

> Also, what I know about secure, air-gapped systems, can be summarized
> quite easily:
>
> - You can not use the same hardware air gapped and non air gapped.

 I beg to differ and at the end of the day this is something that can
be physically/technically proved. Basically, how could you hack a
computer which you booted without a physical networking interface and
(part of the objectives) without loading the networking capabilities
from the kernel by exploiting Linux' runlevels? All you would need to
do is automating updates to that configuration.

> - Maintaining such systems is a pain.

 Well, not really! Booting a Debian Live DVD doesn't take more time
than booting Windows (from scratch) and the whole idea of using a
package extensions USB pen drive would automate updates. This
basically is all there is to maintaining it. You would be basically
making use of the BIOS and RAM of a computer (you don't even need to
own), you would keep the whole OS and all extras you need in your
shirt's front pocket. If they mess with the BIOS you will notice it
because the thing will not work and it would report the BIOS change
and exactly how, what the difference is and for basic physical reasons
you can't infect a computer's RAM.

> - There are no shortcuts.

 Well, no! ... and this is a good thing! We both, "hackers" and "we
the people", have to follow step by step procedures (what Ancient
Greek thinkers called "techne" and later we meant by "functions" up to
Descartes, before all that non-sensical "black box", I/O mindset took
over), what makes the difference is "the touch of God" and that no one
can take away from you that you could take care of your own security,
privacy (as existential philosophers would say: "absolutely no one,
nothing can take away your freedom").

 Notice that I am not just talking about computer soft and hardware. I
got my education as a theoretical Physicist (basically a double-major
in Physics and Math) an der TU Dresden, so I tend to see, understand
every through its physics.

 Experiment:
 1) use a hermetic metal (not plastic, looking like metal!) box (one
of those they use for candies)
 2) turn on your cell phone and carefully put it inside (making sure
it stays on)
 3) close the metal box
 4) right in front of that box place a call to your own phone using another one.
 * since EM waves can't reach your phone it would not only be
functionally off the grid, but off the confines of the universe! and
"Vladimir Putin" couldn't do sh!t about it!
 Isn't that cool!?!?! Now, doesn't it make it even cooler that you can
do such thing without spending one cent?

> Small anecdote: A colleague recently visited a US agencies secure site
> to help them with some software deployment. He could bring one DVD-R,
> not -RW, there. No electronic equipment.

 Well, yes! and how would those kinds of anecdotes speak against a
"touch of God"?

 As anyone could see you could even run a network of detached
computers without networking interfaces in a "touch of God" kind of
way, some sort of "leased One-time pad touches of God" specifically
for each, all coordinated through and which data/information would end
up in a kind of "server", you could even use cell phones to do such
thing ...

 lbrtchx

Re: Image handling in mutt

[Eric S Fraga]

On Friday,  8 Dec 2023 at 17:06, Pocket wrote:
> In Unix and Linux there isn't a file extension, that is a microsoft
> invention.

Predates MS by years.  Systems like RSTS/E on PDP-11s, just to name one.
-- 
Eric S Fraga via gnus (Emacs 30.0.50 2023-09-14) on Debian 12.2