Re: site-to-site VPN with credential prompts?

[tomas]

On Wed, Mar 26, 2025 at 09:41:55AM +0100, Nicolas George wrote:
> to...@tuxteam.de (HE12025-03-26):
> > I was once sitting at a $(DAYJOB) where they blocked everything but
> > 443 (and 80). I tunneled ssh over socat (with TLS, so that the handshake
> > didn't look suspect, in case their firewall sniffed that). Bonus: I
> > got to see whether they did MITM, since I made my own server and
> > client certs.
> 
> If behind a BOFH firewall, ssh is usually a lot easier to tunnel to
> sneak through than a VPN.

My bet was that 443 is always open because otherwise mid- and hi-
level mgmt would be on top of the poor admins because they couldn't
go to their share trading casinos: I won :)

> > Bigcorps are like that. It was not that the firewall department didn't
> > want to talk to me. It was that they bought a "product" without really
> > understanding how it works.
> 
> Must not comment. Must not comment.

My goto quote for this is Bruce Schneier's "Security is a process,
not a product" [1]. If, at a company, this earns me empty stares,
I try to not get involved in their security, but rather watch the
fireworks from afar.

Cheers

[1] https://www.schneier.com/essays/archives/2000/04/the_process_of_secur.html
-- 
t


signature.asc
Description: PGP signature

Re: reprepro dies with "duplicate sort specified but not supported in database"

[Alexandre Rossi]

Hi,

> > Where does this come from? reprepro is version 5.4.6+really5.3.2-1. Did
> > somebody tried to turn back the clock? How can I recover from this?
> > 
> PS: Seems this came up with the reprepro "upgrade" 2 days ago:
>
>   reprepro:amd64 (5.4.6-3, 5.4.6+really5.3.2-1)

Please see https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1100640

Summary is: use reprepro from experimental.

Cheers,

Alex

Re: Does secure old-stable kernel exist with required intel GPU support

[David Wright]

On Sun 23 Mar 2025 at 01:12:29 (-0400), Felix Miata wrote:
> songbird composed on 2025-03-22 08:16 (UTC-0400):
> > Felix Miata wrote:
> >> I'm not sure the subject is asking the right question, but determining 
> >> available
> >> versions of any Debian package has always vexed me. In opensuse, it's 
> >> pretty simple
> >> from shell prompt to get a list of packages available in currently 
> >> configured repos,
> >> one line each, including package name with version. I have yet to find an 
> >> equivalent
> > ...
>  
> >   i'm not able to dig into this further (way behind on this list)
> > but rmadison works ok for something like that:
>  
> > ...
> > root(5)~# rmadison linux-image-amd64
> > linux-image-amd64 | 4.19+105+deb10u16  | oldoldstable   | amd64
> > linux-image-amd64 | 5.10.127-2~bpo10+1 | buster-backports   | amd64
> > linux-image-amd64 | 5.10.223-1 | oldstable  | amd64
> > linux-image-amd64 | 6.1.90-1~bpo11+1   | bullseye-backports | amd64
> > linux-image-amd64 | 6.1.124-1  | stable-updates | amd64
> > linux-image-amd64 | 6.1.129-1  | stable | amd64
> > linux-image-amd64 | 6.12.12-1~bpo12+1  | stable-backports   | amd64
> > linux-image-amd64 | 6.12.17-1  | testing| amd64
> > linux-image-amd64 | 6.12.19-1  | unstable   | amd64
> > linux-image-amd64 | 6.13.7-1~exp1  | experimental   | amd64
>  
> Looks perfect, but for 43 things:
> # rmadison linux-image
> Command 'rmadison' not found, but can be installed with:
> apt install devscripts
> # apt-get install devscripts
> Reading package lists... Done
> Building dependency tree... Done
> Reading state information... Done
> The following additional packages will be installed:
[ … ]
> 0 upgraded, 44 newly installed, 0 to remove and 14 not upgraded.
> Need to get 3,993 kB of archives.
> After this operation, 10.3 MB of additional disk space will be used.
> Do you want to continue? [Y/n] n
> #

I would recommend that you add --no-install-recommends to your command
line. The difference here is dramatic:

  apt-get install devscripts
  0 upgraded, 149 newly installed, 0 to remove and 0 not upgraded.

  apt-get --no-install-recommends install devscripts
  0 upgraded, 11 newly installed, 0 to remove and 0 not upgraded.

On Sun 23 Mar 2025 at 08:48:32 (-0400), songbird wrote:
>   oh, well, ok, i have perl installed for other programs already.
> when i did install it there wasn't that much it brought in.

I have perl installed but not as a language I program in. Yet
installing it didn't bring in all this stuff:

  libalgorithm-diff-perl libalgorithm-diff-xs-perl
  libalgorithm-merge-perl libaliased-perl libarchive-cpio-perl
  libarchive-zip-perl libarray-intspan-perl libcapture-tiny-perl
  libclass-inspector-perl libconfig-tiny-perl libconst-fast-perl
  libcontextual-return-perl libconvert-binhex-perl libdata-dpath-perl
  libdata-messagepack-perl libdata-validate-domain-perl
  libdebhelper-perl libdevel-size-perl libdistro-info-perl
  libemail-address-xs-perl libfile-chdir-perl libfile-find-rule-perl
  libfile-stripnondeterminism-perl libfont-ttf-perl
  libgetopt-long-descriptive-perl libgit-wrapper-perl
  libgitlab-api-v4-perl libhash-fieldhash-perl
  libhtml-html5-entities-perl libhttp-tiny-multipart-perl
  libio-prompter-perl libio-sessiondata-perl libio-string-perl
  libipc-run3-perl libiterator-perl libiterator-util-perl
  libjson-maybexs-perl libjson-perl libjson-xs-perl
  liblist-compare-perl liblist-someutils-perl
  liblist-someutils-xs-perl liblist-utilsby-perl
  liblog-any-adapter-screen-perl liblog-any-perl libmime-tools-perl
  libmoox-aliases-perl libmoox-struct-perl libmouse-perl
  libnet-domain-tld-perl libnumber-compare-perl libnumber-range-perl
  libobject-id-perl libossp-uuid-perl libparams-validate-perl
  libpath-iterator-rule-perl libpath-tiny-perl libperlio-gzip-perl
  libpod-constants-perl libre-engine-re2-perl
  libregexp-pattern-license-perl libregexp-pattern-perl
  libsereal-decoder-perl libsereal-encoder-perl libsoap-lite-perl
  libsort-key-perl libsort-versions-perl libstring-copyright-perl
  libstring-escape-perl libstring-shellquote-perl libsub-override-perl
  libsys-cpuaffinity-perl libtask-weaken-perl libtext-glob-perl
  libtext-levenshteinxs-perl libtext-markdown-discount-perl
  libtext-xslate-perl libtime-moment-perl libtype-tiny-perl
  libtype-tiny-xs-perl libtypes-serialiser-perl libunicode-utf8-perl
  libwant-perl libxmlrpc-lite-perl libyaml-libyaml-perl

  ~# rmadison linux-image-amd64
  linux-image-amd64 | 4.19+105+deb10u16  | oldoldstable| amd64
  linux-image-amd64 | 5.10.127-2~bpo10+1 | buster-backports| amd64
  linux-image-amd64 | 5.10.223-1 | oldstable   | amd64
  linux-image-amd64 | 6.1.90-1~bpo11+1   | bullseye-backports  | amd64
  linux-image-amd64 | 6.1.124-1  | stable-updates  | amd64
  linux-image-amd64 | 6.1.129-1  | stable  | amd64
  linux-image-amd64 | 6.1

Re: Installation Process

[Nicolas George]

Ife Wright (HE12025-03-28):
> I want to install debian but I don't understand why I have to erase
> everything on my hard disk to do it,I just want to install without erasing
> my hard disk

You do not need to erase anything, the installation process will
overwrite whatever it needs. That includes the root data structures of
most anything that could already been there, so the rest, the data that
will not have been overwritten, will be mostly unusable without a lot of
effort.

Regards,

-- 
  Nicolas George

Re: array suggestions

[Anssi Saari]

Dan Ritter  writes:

> Prices via Newegg 24 hours ago. Here's a 4x for $28:
> https://www.newegg.com/p/3C6-00SN-00045

Of course, now I understand. My problem is I don't have a spare x16 slot
to put the card into, just an x4. Hence the bridge chip needed. Next
build, definitely going for a motherboard that has two x16 slots.

> Note that if you need a bridge chip for PCIe bifurcation, those cards are more
> expensive, though you probably still overpaid a bit. Prices
> fluctuate a lot.

Shrug. It was the cheapest I could find at the time so I'm fine with it.

Re: Memory RAM SO_DIMM

[Felix Miata]

William Torrez Corea composed on 2025-04-04 23:01 (UTC-0600):

> I have two memory RAM SO-DIMM in my two slot

> The two memory have the following characteristics:

>- 12800MHz
>- 4GB
>- DDR3L

> But the BIOS only reflects 1600MHz, why? What happened to the rest?

12800 is MB/s bandwidth
1600 is MT/s transfer rate
two different measurements
https://en.wikipedia.org/wiki/DDR3_SDRAM
-- 
Evolution as taught in public schools is, like religion,
based on faith, not based on science.

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata

how to remove debian in bios menu

[hlyg]

i press F12 during Dell boot, a list of options show

under section "UEFI BOOT:" there are 2 items: FreeBSD and debian

but all 3 disks installed use mbr, no wonder they don't work

most probably they are resulted from previous installation on gpt disks

how to remove them?

Re: laptop options

[Lists]

On 2025-03-23 08:37, lina wrote:

Dear all,

Which laptop option is friendly with Debian,
The purpose is related to work, not game.

Mainly for computation, R and some bioinformatic analysis,

Ideally at least > 16 cores, decent memory.

Thanks,


About half a year ago I bought a Lenovo Thinkpad P16 Gen 2 with some 
extras and I have been very satisfied with it. No trouble a all getting 
Debian running on it. As with you, this system is purely for work.


HTH

Grx HdV

Documentation for NetworkManager 1.30.0 -- Where?

[Richard Owlett]

I am running Debian 12.8 with MATE 1.26.0 .
When right-clicking on the Network Manager icon's "About" button it 
identifies itself as version 1.30.0 .
There is a link to http://www.gnome.org/projects/NetworkManager/ which 
results in a 404 error.


I have questions about enabling/disabling WiFi.
To ask in intelligently I must be familiar with the documentation.
Where do I find it?

TIA

Re: recursively share NFS

[gene heskett]

On 3/19/25 16:19, to...@tuxteam.de wrote:

On Wed, Mar 19, 2025 at 02:53:51PM -0400, Eben King wrote:

I have this machine "alexandria".  It mounts a directory from the nas
via NFS.  When I export a parent directory on alexandria, the mount
point appears empty, even though you can ssh to it and see everything
there that should be.  How do I get it to share the contents of that mount?

It is possible, see the file /etc/exports and its man page exports(7),
specifically the options nohide and crossmnt there.

There are caveats, though (e.g. you might get duplicate inode numbers
from the two file systems), described in the above man page.

Sorry to be so handwavy, but it's a while ago I had hands-on experience
with that.

Cheers


I've had nothing but intermittent trouble trying us use NFS like that, 
so a decade or more ago, I made it a habit to create


an /sshnet directory, and then a subdir for the share by machine $name 
in that, then used ssh to mount each machines /home/me dir via sshfs. 
Then I wrote a bash script that when executed, mounts each machine to 
its subdir so each machine is a direct path thru that subdir. I 
specifically deny root thru that connection which occasionally creates 
problems, but I can always put a root needing file in that machine user 
dir, and ssh into that machine to fix the perms and move it as needed. 
sshfs Just Works   NFS found more excuses to deny that than Berny Maddof 
at his best.


Everybody has his fav bag of tricks,  that's mine.

Cheers, Gene Heskett, CET.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis

Re: Pls help fixing /boot/efi and GRUB

[J]

>
> "sudo -i" is meant to approximate the behavior of "su -".  Before buster,
> nobody would have used that on a Debian system.  It's horrible.  The
> fact that people are now embracing it as a norm is even worse.
>

Why horrible?

Re: OT: Connect two computers with linux with wlan, but without any router

[John Hasler]

Hans writes:
> This looks strange for me, as I would think, the AP on the computer
> would also need some processing time for recognition, correction and
> routing to the host.

Every packet is routed by the kernel.  There is no seperate "AP".

How much delay matters?  Ping should be under a millisecond.
-- 
John Hasler 
j...@sugarbit.com
Elmwood, WI USA

Re: Pls help fixing /boot/efi and GRUB

[Nicolas George]

Greg (HE12025-03-27):
> I'm certain sudo has its use cases, but all I do personally is su to
> root and update and upgrade my stable Bookworm using apt, so I feel no
> need to complexify the issue with sudo.

The fallacy in here being assuming, without stating it and without
justifying it that sudo makes things more complex than su.

Please, when replying to this, double check you do not commit the second
fallacy to assume that one thing is simpler than another when the
difference is you already know the first and not the second.

Regards,

-- 
  Nicolas George

Re: Package kde-config-mobile-networking - what does it do?

[Greg Wooledge]

On Tue, Apr 01, 2025 at 19:54:19 -0400, Timothy M Butterworth wrote:
> Try running `sudo apt info kde-config-mobile-networking` and see what the
> info says.

You don't even need sudo for that.

Re: libsmbios/smbios-utils removed from Debian?

[Thomas Schmitt]

Hi,

Václav Ovsík wrote:
> https://tracker.debian.org/news/1626825/removed-243-2-from-unstable/
> https://tracker.debian.org/pkg/libsmbios
> [...]
> I am surprised the smbios-utils is removed from unstable.

  https://tracker.debian.org/news/1626825/removed-243-2-from-unstable/
links to the discussion which led to the removal:
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1099864


Have a nice day :)

Thomas

Re: Trixie Wayland login issue

[Timothy M Butterworth]

On Thu, Mar 20, 2025 at 10:41 AM Timothy M Butterworth <
timothy.m.butterwo...@gmail.com> wrote:

> All,
>
> After installing updates this morning I am not able to login to a KDE
> Wayland session. The login just cycles back to the login screen. I am able
> to login to a X11 session. Is anyone else experiencing this issue?
>

I rebooted and now it works again.

Tim


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


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

Re: Frequent freezing around login screens

[Max Nikulin]

On 27/03/2025 14:34, W. Pepperdine wrote:

dmesg (http://paste.debian.net/1365575/)


I am confused. I have impression that you described your trouble as 
rather severe freeze with no reaction on keyboard an mouse. Am I wrong? 
How have you managed to get dmesg output in that state? 
/var/log/kern.log has wall time timestamp, not monotonic ones.



[   35.014008] systemd-journald[289]: Time jumped backwards, rotating.


Some applications may have bugs related to backward time jumps. How much 
that jump was? It might be something wrong with time configuration


timedatectl

There is a little chance that RTC battery might be discharged (causing 
issues with BIOS settings).



[   35.040339] systemd-journald[289]: Failed to read journal file 
/var/log/journal/c113f4b56932414d90d5b7ade9c142d9/user-1001.journal for 
rotation, trying to move it out of the way: Device or resource busy
[   35.040583] systemd-journald[289]: Failed to read journal file 
/var/log/journal/c113f4b56932414d90d5b7ade9c142d9/user-1000.journal for 
rotation, trying to move it out of the way: Device or resource busy


I have not seen this kind of messages so far. At least they give some 
keywords for search queries.


May it happen that no space left on some partition?

df -h


journalctl -b -p3 (http://paste.debian.net/1365578/)


Just "-b" without additional argument like "-1" means current boot. It 
is useless. Get logs for some boot when you experienced a hang. Do not 
use "-p" filtering. If it is not comfortable for you to post detailed logs


journalctl -e -b ID_OF_FAILED_BOOT

then try to figure out suspicious entries (not necessary errors) 
yourself. Repeat the same without "-e" to get all logs for that boot.

Re: array suggestions

[Will Mengarini]

* Eben King  [25-03/21=Fr 15:32 -0400]:
> I have a 2x1TB RAID-1 array on one of my computers.  It holds
> a backup.  It's starting to become too small, not because it's
> shrinking, but because I'm getting more stuff.  So, I need to
> do something that ends up with a larger array using 3, maybe 4
> disks.  It'd be nice if it supports disks of disparate sizes (and
> actually uses the extra space), so I can upgrade by attrition.
>
> I am by no means an expert at mdadm.  Heck, I'm barely competent at
> it.  So I have no particular attachment to it.  My friend uses btrfs
> to make a (for me) massive array, some 6-8 disks and probably 40 TiB
> of space.  But it seems he spends a lot of time on administrivia,
> balancing the array and whatnot.  Maybe that's because it's so
> large?  Dunno.  I've heard there are other filesystems that do
> similar things, but I'm not familiar with them.  Any recommendations?

LVM is used for this kind of thing.  You can install LVM over RAID
(which is better than installing RAID over LVM):

https://serverfault.com/questions/217666/what-is-better-lvm-on-raid-or-raid-on-lvm

This gives a reasonable procedure to follow:

https://tomlankhorst.nl/setup-lvm-raid-array-mdadm-linux

However, note that if your filesystems aren't already part of
LVM, you'll either need to clobber them (trusting backups for
restoration, which is slow as well as scary) or shrink your
existing partition(s) and use a new one for LVM, planning
to extend that new partition by later adding more drives.

Hmm, what consumer-level motherboard supports
more than 2 NVMe drives?  Or are you using SATA?

If you need vastly more information and have many hours
available for leisurely reading, you might want to look at
.

For dealing with humongous multiline URLs like that, I sometimes find
  smush() {
tr -d '[:space:]' <<< "$*";
echo
  }
to be helpful.

Re: Unable to install GRUB in dummy

[David Wright]

On Sun 30 Mar 2025 at 11:50:52 (+0800), hlyg wrote:
> i install with debian-12.10.0-amd64-netinst.iso
> 
> during final stage of installation, it fails to install grub
> 
> "Executing 'grub-install dummy' failed."
> "This is a fatal error."
> 
> how to solve it?

Presumably that error message was from the screen. Have you looked
at /var/log/installer/syslog for more expansive error messages?

Cheers,
David.

Re: Unable to install GRUB in dummy

[hlyg]

On 4/1/25 05:20, Andrew M.A. Cater wrote:

Make it simple for yourself. If you are at all unsure when you install,
just take guided partitioning.

That should set up a 512M partition for boot, a 1G partition for swap
and the rest of the disk for /

Unless you *really have* to partition things yourself, putting everything
in one partition will normally work. Once you've done this a few times
and know what you want, then you can partition separate space for /var
or whatever.

GPT and UEFI is the way forward on machines that support them.

All the very best, as ever,

Andrew Cater
(amaca...@debian.org)


Thank Cater!

my other PCs are in those transition period , support both legacy mbr 
and modern gpt

Re: Monthly FAQ for the Debian-user mailing list (last modified 20250404)

[Greg]

On 2025-04-05, Jeffrey Walton  wrote:
> Still missing a topic or discussion of "SOLVED" in the subject.
>

We're all waiting for Gene to put "SOLVED" on his never-ending network
of threads.

But what would it would mean or communicate to future anthropologists
remains yet another puzzle left to the reader.

Re: Memory RAM SO_DIMM

[David Christensen]

On 4/4/25 22:01, William Torrez Corea wrote:

I have enabled swap memory, but if i disable the swap memory the machine is
slow.



I tried running computers without swap and found that they crashed when 
the running programs used too much memory.  Now I allocate 1 GB swap on 
the system drive during installation and enable the System Load Monitor 
applet in the Xfce to watch swap usage during operation.  If the 
programs use up memory and start to swap, I can do something about it.



David

Re: Who: Bookworm v. Trixie

[Titus Newswanger]

On 3/31/25 15:24, gene heskett wrote:
I blame it on the busted bookworm installer. Anything plugged into usb 
triggers it to put orca and brltty in whether you want it or not.  I 
don't own a wired mouse. I did close to 40 installs trying to find a 
way around that but probably 30 of he reinstalls were because it would 
not reboot once I'd shut orca up. Couldn't remove it because of 
dependencies until an update finally fixed the dependencies but now 
I'm stuck with a 30 second to a full minute system freeze while trying 
to create or open a file in my /home/me path. That's very distracting, 
fouls up ones train of thought waiting and waiting for the file 
requestor to open. Once it does open, freezeup is finished, but it 
sure wastes a lot of time. Mouse pointer moves but buttons and 
keyboard are dead.


I've personally had two different situations where my PC behaved that slow:

1: I had my network config mangled (dns not resolving or a mounted share 
folder lost connection with the server)


2: my hard disk/ssd was so wore it took it that long to read/write files...

If neither is your problem then I can't think what next.


I've had where brltty would grab any USB serial converter I plugged in, 
then I couldn't use them for their intended purpose. Since I don't use 
brltty I uninstalled it, fixing that problem.




Cheers, Gene Heskett, CET.


--
Thank You!

Titus Newswanger
Curtiss WI

Re: how to remove debian in bios menu

[Michel Verdier]

On 2025-04-05, Hans wrote:

> Maybe  for someione interesting: As I also have Windows on my drive, there is 
> an entry for Windows. I deleted this, because then I only have the entry 
> "debian".  And this is tarting grub, which got an entry for Windows.
> Dunno, if this is a good way, but it is working for me.

If I remember well it will be added again with each windows update. And
potentially set to the default boot...

Re: how to remove debian in bios menu

[Jeffrey Walton]

On Sat, Apr 5, 2025 at 6:02 AM hlyg  wrote:
>
> i press F12 during Dell boot, a list of options show
>
> under section "UEFI BOOT:" there are 2 items: FreeBSD and debian
>
> but all 3 disks installed use mbr, no wonder they don't work
>
> most probably they are resulted from previous installation on gpt disks
>
> how to remove them?

Use efibootmgr. Gentoo has a nice page on using it.

Jeff

Re: recursively share NFS

[tomas]

On Thu, Mar 20, 2025 at 02:04:45PM +0800, An Liu wrote:
> AFAIK,  
> 
> Exporting a nfs mounted location is possible via nfs-ganesha

I have no experience with this one. Thanks for reminding me :-)

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: Failed unmounting disk mes. on every restart

[J]

>
> I would investigate why my ext4 disks were not being unmounted cleanly
> on shutdown if it were happening here. I suppose my suspicions would
> first fall on any network connections.


Yes, i want to investigate it too, but i don't know where to start. SMART
of the disks is OK, btw.

I have experimented with different suspect-apps, looks like the culprits
are qBitTorrent and Nicotine+ (Soulseek) with access to the problematic
partitions.

Btw, when QB is working (not even "active") i have to click on the Restart
button always twice. I have it on Gnome, but people say the same problem is
on KDE.

So, i want to stress the question - can the failure of a disk unmounting
lead to the loss of information?

Re: OT: Connect two computers with linux with wlan, but without any router

[Henrik Ahlgren]

Hans  writes:

> yes, I already am aware of this, but this I wanted to avoid. It will be then 
> again a new hop, which causes delay (and I suppose, a software router is 
> sklower than a hardware device).

I haven't tried this, but take a look at:

https://wiki.debian.org/WiFi/AdHoc
https://help.ubuntu.com/community/WifiDocs/Adhoc

I suppose it is feasible also to configure the Wi-Fi in "infrastructure
mode", where the other PC functions as an access point without
necessarily performing any routing. While most consumer "routers" act
both as an AP and a router, they are not the same thing.

Re: Failed unmounting disk mes. on every restart

[David Wright]

On Sun 30 Mar 2025 at 17:35:18 (+0300), J wrote:
> Every time i restart the PC i have an error message while rebooting
> process: "[FAILED] failed unmounting *disk-mount-point*..."
> 
> It didn't bother me really, because this message usually just immediately
> disappeared. But last time the *computer could get stuck* on this for a
> minute or two.
> 
> These are *ext4* partitions.

That pause sounds normal to me. If the power was cut to the computer,
then on booting up, I will see an extra line, the first in this pair:

  noah03: recovering journal
  noah03: clean … … …

(noah03 is the filesystem LABEL.)

> Mounting points were made with *Gnome-disks* and in *FSTAB* it looked like
> that
> 
> */dev/disk/by-uuid/*8cd66b97-bde6-4475-b875-8f3928a8b14d
> /mnt/8cd66b97-bde6-4475-b875-8f3928a8b14d *auto
> nosuid,nodev,nofail,x-gvfs-show 0 0*
> 
> Then i remade it manually this way
> 
> *UUID=8cd66b97*-bde6-4475-b875-8f3928a8b14d
> /mnt/8cd66b97-bde6-4475-b875-8f3928a8b14d auto
> nosuid,nodev,nofail,x-gvfs-show 0 0
> 
> Now it doesn't stuck trying to unmount it for a minute, but still i see the
> error warning

I don't think that makes any difference. How long it takes to recover
the journal depends on how much work is involved, so it's variable.
I guess it increases with the time the system has been up and running.

> Of course, as expected *if i close all the apps* and processes which use
> these partitions, *there is no error*.
> 
> But shouldn't PC do the unmount automatically after the apps are closed
> during the reboot process?

Signals are sent on shutdown, but programs have to react to them or
they get killed off after a certain period of time (a few seconds?).
Even then, then system should be able to close files and unmount.

> Does it warns me that it couldn't unmount the disk completely? Or it just
> informs that it couldn't do it from the first time?

AIUI, with ext4, it should normally be able to clean the partition,
hence the second line quoted above.

> Should i be worried? Is it a* must *to close all the apps and/or unmount
> disks manually before reboot?
> 
> Is there a risk of information loss or disk corruption?

I would investigate why my ext4 disks were not being unmounted cleanly
on shutdown if it were happening here. I suppose my suspicions would
first fall on any network connections.

(As you probably know, with FAT disks, it's completely different. You
typically have to chkdsk them yourself (or not bother). It can be
tricky to get scanners, cameras and phones to unmount their sticks
and cards cleanly, so here they often live with their dirty bit set.)

Cheers,
David.

Re: Pls help fixing /boot/efi and GRUB

[Anssi Saari]

David Wright  writes:

>   host!auser 09:57:47 /somewhere/that/is/obnoxiously/long/program-1.2.3$ 
> /bin/su --login
>   Password: 
>   bullseye on /dev/sda5 toto05
>   host 09:57:59 ~# cd /somewhere/that/is/obnoxiously/long/program-1.2.3
>   host 09:58:08 /somewhere/that/is/obnoxiously/long/program-1.2.3# 
>
> where that's a simple cut and paste.

Yes. I don't remember why but at some point in the distant past I got
into the habit of su - and expect to end up in root's home dir and
nowhere else.

Thinking back, it could be argued some crude consoles like Sun's or old
school text terminals might not have copy-paste available. Maybe even
today it could be a thing, don't really know. If my old HP Microserver
gen8 is any indication, server people like to do extremely weird stuff
for remote management.

Re: DHCP and static addresses, nothing to do with Re: Who: Bookworm v.Trixie

[gene heskett]

On 3/31/25 13:55, David Wright wrote:

On Mon 31 Mar 2025 at 11:19:30 (-0400), gene heskett wrote:

The dns problem is separate I guess, but does bring up my other pet
peeve. That is that no one at debian considers the effect on dns to
those of us who have been using hosts files for local dns since back
in the late 90's  I have no  dhcpd setup and rig my lashup so that my
local lookups are first and in the hosts file, if not it the hosts, my
isp's dns gets queried. But every new install changes things around
resolv.conf making that harder and harder to do.

I don't understand. You have a router running dd-wrt, don't you.
https://wiki.dd-wrt.com/wiki/index.php/Static_DHCP
explains how to configure it to hand out static addresses, yet you
maintain that it can't do that, eg:
   https://lists.debian.org/debian-user/2023/12/msg9.html


What makes the debian people treat hosts file users, like 3rd class
users?   Its  easier to setup, needs less maintenance, and Just Works
since my first linux install in '98...  Sure, we can lock NM from
tearing a working net down by making resolv.conf immutable and a real
file. We no longer have to do that with bookworm but from wheezy to
bookworm we did have to protect resolv.conf from NM.  But every time
we mention it, we catch it by giving us what for w/o telling us how to
make it work. That too gets old. Why?

I don't know why you have problems with using /etc/hosts for lookups
on your LAN. I use it here without any problems, and it has to work
because there's no DNS server in my router (too cheap).

   $ grep hosts: /etc/nsswitch.conf
   hosts:  files mdns4_minimal [NOTFOUND=return] dns

because files doesn't work in bookworm, I had to:

grep hosts: /etc/nsswitch.conf
hosts:  files mdns4_minimal [NOTFOUND=return] dns myhostname
to make the hosts file work


   $

The hosts file is just a reformatted list of the Reserved Addresses
list from the router, with my "fake" domain (.corp) added, and it gets
consulted first.

The main advantage of using my router to hand out static addresses
is that it includes printers, scanners, mobile phones, TVs and set-top
boxes, stuff that I don't want to bother with configuring myself, plus
the fact that it's always powered on.

[ … ]


Sorry to disappoint you but that seems to be working Just Fine, but
once again, you make no attempt to either explain why its wrong, or to
tell us what the right way is other than demanding we waste a week
making dhcpd actually work.

Then why the complaint?

Cheers,
David.

.


Cheers, Gene Heskett, CET.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis

Re: Web server access

[tomas]

On Wed, Apr 02, 2025 at 03:55:08PM +0200, Nicolas George wrote:
> debian-u...@howorth.org.uk (HE12025-04-02):
> > Well, practically it makes no difference. If I send with or without an
> > HTTP version I get the same Bad Request response. And it makes no
> > difference whether I use HTTP/1.0 or HTTP/1.1.
> 
> Does it make a difference if you send CRLF instead of LF, as Tomas
> mentioned? For that, you would need to hit ctrl-enter and see ^M in the
> terminal, each time before you hit enter.

To be fair, I said that most web servers are lenient. The RFCs state
CRLF, though.

I've been immersed in $DAYJOB, so I haven't been paying very close
attention, but my impression was that the problem is solved?

FWIW, my local web server, a lighttpd, responds also with a "400
Bad Request" to a "GET /" without a version. Some randomly tested
hosts "out there" sometimes play along, sometimes not.

And oh, telnet converts the LFs to CRLFs. With nc (which doesn't
translate), LFs alone also elicit a Bad Request (again, on my local
lighttpd instance).

Cheers
-- 
t


signature.asc
Description: PGP signature

SOLVED

[Greg]

On 2025-04-04, Andrew M.A. Cater  wrote:
>
> I hope this is satisfactory to all concerned: if it isn't, please reply
> in a new mail with a meaningful subject.
>
> With every good wish, as ever,
>
> Andy Cater
> (amaca...@debian.org) 
>
>

Re: Documentation for NetworkManager 1.30.0 -- Where?

[Richard Owlett]

On 4/2/25 9:32 AM, Peter Ehlert wrote:

try here:
https://gitlab.gnome.org/GNOME/network-manager-applet


That page tries to load. Evidently doesn't like my configuration of 
SeaMonkey. Doesn't surprise me as I've got a strange configuration.


However Debian's default configuration of Firefox has no apparent 
problem with it.


However that page has no link to operation documentation.
Any suggestions and thanks for trying.



On 4/2/25 06:41, Richard Owlett wrote:

I am running Debian 12.8 with MATE 1.26.0 .
When right-clicking on the Network Manager icon's "About" button it 
identifies itself as version 1.30.0 .
There is a link to http://www.gnome.org/projects/NetworkManager/ which 
results in a 404 error.


I have questions about enabling/disabling WiFi.
To ask in intelligently I must be familiar with the documentation.
Where do I find it?

TIA

Re: increment backup of home dir

[songbird]

Jerome BENOIT wrote:
> Hi,
>
> backup2l is simple and has been reliable for me for years.
>
> Cheers,
> Jerome

  interesting, at first glance it might help me out, but
i don't know for sure.  i'm a bit worried though that the
debian package doesn't look like it is actively maintained.

  i have old backups in tar format.  will backup2l allow
those to be used as a source in such a way that each one
is not duplicated without having to unpack them first?

  ages ago i used bup and it was ok, but i really did not
like the presentation for going back and finding things.
this was a long time ago though and now perhaps it isn't
so bad.  doesn't matter now as i have no old bup anyplace.

  ultimately i really need a way to do backups that will
deduplicate and must be 100% bulletproof and stable.


  songbird

Re: Web server access

[Timothy M Butterworth]

On Tue, Apr 1, 2025 at 8:12 PM Van Snyder  wrote:

> On Tue, 2025-04-01 at 20:00 -0400, Timothy M Butterworth wrote:
>
>
>
> On Tue, Apr 1, 2025 at 7:57 PM Van Snyder 
> wrote:
>
> This might be the wrong forum for this question, but most likely somebody
> can tell me a better place.
>
> I have a web server listening to port 80 (http) and 443 (https).
>
> I can load pages from it from any computer in my house, all behind the
> same router, using its IP number.
>
> I enabled port forwarding in the DMZ in my router for ports 80 and 443.
>
> I can't load pages through my router using its WLAN name or WLAN IP
> number. I get "Unable to connect" from Firefox. or "This site can't be
> reached"  and ERR_ADDRESS_UNREACHABLE from Konqueror.
>
>
> What do you mean by WLAN Name and WLAN IP? Are you using DNS? Are both
> devices connected to the WLAN? Some WLAN Routers have a security setting
> for device isolation that prevents WLAN Clients from networking with each
> other.
>
>
> The WLAN IP number is the router's "outside" number. My server is on the
> "inside" side of the router.
>

Ok so if I understand you correctly then you are attempting to port forward
80 and 443 through the router's WAN Wide Area Network interface to a server
located in the DMZ DeMilitarized Zone. Does the server have Apache ACL's,
IP Tables or TCP wrapper running on it? Can you try to do a port ping or
use telnet to connect to port 80 to test connectivity. ex: `telnet  80`. As you say that the server is on
the inside of your network. Have you tried placing the server in the DMZ?


>
>
>
> I have mapped port 8079 to port 80 in my router. I can't load pages using
> that mapping.
>
> I also map an external port (not 22) to port 22, and I can "ssh" to my
> computer using its WLAN name.
>
> I have cleared browser caches. I have restarted the router. I'm not using
> a proxy server.
>
> This was all working until about three weeks ago. I didn't change the
> firmware in my Linksys.
>
> Any ideas?
>
>
>
> --
> ⢀⣴⠾⠻⢶⣦⠀
> ⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
> ⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
> ⠈⠳⣄⠀⠀
>
>
>

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

Re: OT: Connect two computers with linux with wlan, but without any router

[Stefan Monnier]

> This looks strange for me, as I would think, the AP on the computer
> would also  need some processing time for recognition, correction and
> routing to the host.

Try it!
If you notice an important performance penalty, *then* come back with
the numbers and the details of your setup, so someone can help you.
IME when it comes to wifi communication, routing is the least likely of
the performance problems.


Stefan

Re: DHCP and static addresses, nothing to do with Re: Who:Bookwormv.Trixie

[Michael Stone]

On Wed, Apr 02, 2025 at 10:28:24PM -0500, David Wright wrote:

I don't see the point in leaving it there. If you want to send
something to coyote.coyote.den, why do you want the LAN address
when 127.0.1.1 is just as good. If the line is correct, it does
nothing; if it's incorrect, it can cause harm.


FWIW, I see absolutely no point in making the change. Debian uses 
127.0.1.1 because it's there regardless of the network configuration, 
but in the case of a static IP there's no particular benefit in doing 
things that way. This entire subthread is a red herring as there's 
nothing particular problematic about the configuration in question.

Re: Pls help fixing /boot/efi and GRUB

[tomas]

On Wed, Mar 26, 2025 at 07:48:16AM -0500, Richard Owlett wrote:
> On 3/26/25 6:55 AM, Greg Wooledge wrote:
> > [SNIP]
> > 
> > I normally use "sudo -s", which is the closest sudo approximation to
> > the traditional behvior of "su" (before it was broken in buster).
> > 
> 
> I don't understand the reference to some "brokenness" of "su".
> I've not closely followed this thread so I may be missing context.
> 
> I only use "su" when doing something in MATE terminal on my local machine. I
> do not use any command line options to "su". I just wait for it to ask for
> my root password. I perform a few commands and then close that MATE
> terminal.
> 
> Does this "brokenness" of "su" have any potential effect on my usage?

Read the first item here:

  https://wiki.debian.org/NewInBuster#Changes

...and kids, always read the release notes (I do fail myself in that,
too :)

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: Web server access

[Van Snyder]

On Wed, 2025-04-02 at 01:17 -0400, Timothy M Butterworth wrote:
> I am able to reach The Van Snyder's Web Site using the above IP
> address and URL on port 80 but not 443. I got a certificate error on
> 443. 

I've never before set up a secure server. I followed instructions at a
web page, whose URL I neglected to put into my notes, to set up the
SSL.

I probably did something wrong.

Was there a clue in the error message about what I did wrong?

What page of explanations about how to do it right do you recommend?

Re: Installing old Debian releases

[Thomas Schmitt]

Hi,

keller.st...@gmx.de wrote:
> For comparison, some research and portability tests I'd like
> to install old releases of Debian, i.e. versions 8, 9, 10.
> Are there archives and old repositories to install from?

Old installation and Live ISOs are at
  https://cdimage.debian.org/mirror/cdimage/archive/

The repository
  http://us.cdimage.debian.org/cdimage/snapshot/Debian/
is mentioned in .jigdo files as source of about every package.
Peeking into
  http://us.cdimage.debian.org/cdimage/snapshot/Debian/pool/main/libb/libburn/
i see lots of libburn packages back to year 2008.


Have a nice day :)

Thomas

Re: Who: Bookworm v. Trixie

[Timothy M Butterworth]

On Sun, Mar 30, 2025 at 6:53 PM Timothy M Butterworth <
timothy.m.butterwo...@gmail.com> wrote:

>
>
> On Sun, Mar 30, 2025 at 6:25 PM Charles Curley <
> charlescur...@charlescurley.com> wrote:
>
>> On bookworm who (GNU coreutils 9.1) operates more or less as I have
>> expected it to operate for several decades: it prints current logins.
>> E.g.:
>>
>> charles@hawk:~$ who
>> charles  tty7 2025-03-30 11:31 (:0)
>> charles  pts/35   2025-03-27 20:13 (192.168.100.47)
>> root pts/36   2025-03-27 21:48 (192.168.100.47)
>> charles@hawk:~$
>>
>> On trixie who (GNU coreutils 9.5) gives me a long list of logins, most
>> of which predate the most recent reboot. "who -u", similarly.
>>
>
> On my Trixie system `who --users` only provides a carriage return and
> prints no information. 'who -a' only prints the last boot time. Seems like
> who is seriously broken.
>
>
>>
>> How do I get only the current logins?
>>
>
Use the w command on Trixie:
tmb@debian-hp:~$ w
04:56:49 up 2 days,  3:26,  2 users,  load average: 0.09, 0.17, 0.17
USER TTY  FROM LOGIN@   IDLE   JCPU   PCPU WHAT
tmb  tty1 -Sat012days  0.07s  0.07s
/usr/bin/startplasma-wayland
tmb   -Sat013:24m  0.00s  2.41s
/usr/lib/systemd/systemd --user



>
>> Also, who on bookworm works fine with no arguments. who on bookworm
>> requires the file to use in order to get any useful output at all,
>> e.g.: "who -Hu /var/log/wtmp". This might be a bug.
>>
>> Finally, I see that bug #798910, "coreutils: /usr/bin/who --lookup does
>> not look up ip addresses in dns", is still outstanding and a bit
>> annoying. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=798910
>>
>> Thank you.
>>
>> --
>> Does anybody read signatures any more?
>>
>> https://charlescurley.com
>> https://charlescurley.com/blog/
>>
>>
>
> --
> ⢀⣴⠾⠻⢶⣦⠀
> ⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
> ⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org/
> ⠈⠳⣄⠀⠀
>


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

Debian repository -- any well documented PDF to HTML converters?

[Richard Owlett]

I've just received a 68 page PDF document. I don't know proper 
terminology but visually it resembles an outline in many of it's items 
are multiple paragraphs. The points are "collapsible". When everything 
is collapsed, the document index is 9 lines.


Due to vision/perception issues I find it more comfortable to work wit 
HTML documents than PDF. As my default OS is Debian 12.8, I used 
Synaptic to search for a conversion utility.

Found that poppler-utils { supplying pdftohtml } was installed.
Used a sparse man-page [ 
https://manpages.debian.org/bookworm/poppler-utils/pdftohtml.1.en.html ] 
to make a generally good test run.


The Poppler Wiki describes the situation as:
There is currently very little documentation for the intenal library. 


Is there a Debian package whose HTML output can collapse sub-points when 
the input PDF does so?


TIA




I has a *problem*. The PDF can collapse sub-points. The HTML *cannot*.
Went looking for more complete documentation than the man-page.

Re: Installation Process

[Jerome BENOIT]

Hi,

On 28/03/2025 18:42, Nicolas George wrote:

Ife Wright (HE12025-03-28):

I want to install debian but I don't understand why I have to erase
everything on my hard disk to do it,I just want to install without erasing
my hard disk


Before to do so, you want to make a backup of your data (basically your HOME 
directory)
on an external support.

Note that if you just want to have a look on how Debian can work on your box,
you want to play with a Debian live-cd.



You do not need to erase anything, the installation process will
overwrite whatever it needs. That includes the root data structures of
most anything that could already been there, so the rest, the data that
will not have been overwritten, will be mostly unusable without a lot of
effort.

Regards,




Cheers,
Jerome

Re: DHCP and static addresses, nothing to do with Re:Who:Bookwormv.Trixie

[gene heskett]

On 4/2/25 23:29, David Wright wrote:

On Wed 02 Apr 2025 at 09:12:24 (-0400), gene heskett wrote:

On 4/2/25 01:28, David Wright wrote:

On Tue 01 Apr 2025 at 04:58:27 (-0400), gene heskett wrote:

On 3/31/25 23:02, David Wright wrote:

On Mon 31 Mar 2025 at 16:35:58 (-0400), gene heskett wrote:

On 3/31/25 13:55, David Wright wrote:

What Brian pointed out in the thread: the lack of 127.0.1.1, the
conventional way in which Debian ensures a host can find its own
name when the network is not up.

I can put it back in, but no one has ever explained why.  And since
it's been gone for 27 years, what name goes with it?
Ack the man page, it should be coyote.coyote.den, but that has been
192.168.71.3 for that same 27 years. [ … ] So
what should I put in there for what the man page says?:
127.0.1.1   thishost.example.org   thishost

127.0.1.1 coyote.coyote.den coyote


Experimenting I find the duplication does not seem to generate an
error, other than I now had to ping itself by address, since the name
is now found at 127.0.1.1 by pings lookup?

That's what you want: as the address is in the 127.0.0.0 network,
pinging it will ping itself, and it gets a reply. It doesn't
require your LAN to be set up, and AIUI it's like localhost
(127.0.0.1) in that it doesn't touch the network hardware.


I'll leave it in the hosts
file as a duplicate,  until I find something that does not work,

I don't see the point in leaving it there. If you want to send
something to coyote.coyote.den, why do you want the LAN address
when 127.0.1.1 is just as good. If the line is correct, it does
nothing; if it's incorrect, it can cause harm.

And coyote's own hosts file can't be seen by other machines trying
to find coyote: they will use their local copy.

Bear in mind that the same holds true for each machine on your LAN,
so the hosts file will be different for each one. My master list,
which I reconcile with the router's DHCP server Reservation List,
is installed onto a system with a line like:
The router is, I believe, running dnsmasq, but is otherwise untouched 
dd-wrt

   # sed -E 
"/^[[:space:]]*192.168.1.[0-9]+[[:space:]]+$HOSTNAME.corp[[:space:]]+$HOSTNAME[#|[[:space:]]|\$]/s/[[:space:]]*([0-9.]+)[[:space:]]+(.*)\$/127.0.1.1\t\2\t#
 \1/" master-list >/etc/hosts

That assumes 1, not 71, corp rather than coyote.den, and it would
fail on your .122 line because the HOSTNAMEs are different for some
reason. (I wrote the pattern to conform to my own expectations.)


Which works for either alias:

gene@coyote:~$ ping bpi51
PING bpi51e5p.coyote.den (192.168.71.122) 56(84) bytes of data.
64 bytes from bpi51e5p.coyote.den (192.168.71.122): icmp_seq=1 ttl=64 
time=0.571 ms

gene@coyote:~$ ping e5p
PING bpi51e5p.coyote.den (192.168.71.122) 56(84) bytes of data.
64 bytes from bpi51e5p.coyote.den (192.168.71.122): icmp_seq=1 ttl=64 
time=0.592 ms


And I'm sitting here, watching that printers web page, generated by 
klipper, both with the webkit in PrusaSlicer on this machine, and with 
firefox on a different xfce4 workspace, no difference in the appearance 
here on this machine, about 100 feet of cat5 away. It Just Works. ;o)



but
it also has no effect on the 30 second gui freeze on opening a file I
own.

I don't see why it would. But I don't have experience with DEs, so
I wouldn't be much help with a problem like that anyway. I also don't
know what your problem with NM and resolv.conf is all about. But
static addressing with a hosts file doesn't have to be 3rd class,
and I don't feel treated as such.

Cheers,
David.


Thank you David.  Take care of #1.

heers, Gene Heskett, CET.

--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis

Re: laptop options

[Russell S.]

>> Which laptop option is friendly with Debian,
>> The purpose is related to work, not game.
>
>System76 has usually good and nice offers:
>
>  https://system76.com/laptops

I can personally vouch for System76. I bought a Darter Pro (darp5) about 6 years
ago and it's still running well. It could probably use a new battery, but other
than that, it's good. I've had Pop_OS! (the default), Arch, Debian, Parrot, and
now FreeBSD on it. It'll even run Windows if you have to.

-- 
this is my clever sig.

Re: site-to-site VPN with credential prompts?

[Erwan David]

On Wed, Mar 26, 2025 at 10:03:36AM CET, to...@tuxteam.de said:
> On Wed, Mar 26, 2025 at 09:41:55AM +0100, Nicolas George wrote:
> > to...@tuxteam.de (HE12025-03-26):
> > > I was once sitting at a $(DAYJOB) where they blocked everything but
> > > 443 (and 80). I tunneled ssh over socat (with TLS, so that the handshake
> > > didn't look suspect, in case their firewall sniffed that). Bonus: I
> > > got to see whether they did MITM, since I made my own server and
> > > client certs.
> > 
> > If behind a BOFH firewall, ssh is usually a lot easier to tunnel to
> > sneak through than a VPN.
> 
> My bet was that 443 is always open because otherwise mid- and hi-
> level mgmt would be on top of the poor admins because they couldn't
> go to their share trading casinos: I won :)

Admins would also have problems to get security updates (and not accessing 
*overflow)


-- 
Erwan David

Re: Unable to install GRUB in dummy

[Andrew M.A. Cater]

On Tue, Apr 01, 2025 at 04:09:31AM +0800, hlyg wrote:
> 
> On 3/31/25 10:50, David Wright wrote:
> > Presumably that error message was from the screen. Have you looked
> > at /var/log/installer/syslog for more expansive error messages?
> > 
> > Cheers,
> > David.
> > 
> Thank Wright! i have solved it on my own
> 
> i am new user of gpt, fatal error is caused by my failure to create efi
> partition
> 

Make it simple for yourself. If you are at all unsure when you install,
just take guided partitioning.

That should set up a 512M partition for boot, a 1G partition for swap
and the rest of the disk for / 

Unless you *really have* to partition things yourself, putting everything
in one partition will normally work. Once you've done this a few times
and know what you want, then you can partition separate space for /var
or whatever.

GPT and UEFI is the way forward on machines that support them.

> it seems both installer and i are quite dumb
> 
> in the end i use mbr, as one of my PCs doesn't support gpt
> 
> btw what is recommended size of efi partition?
> 
> default size is 512M
> 

All the very best, as ever,

Andrew Cater
(amaca...@debian.org)
> 
> 

Re: how to remove debian in bios menu

[hlyg]

Thank Hans! it is really Dell issue. i follow your instruction, it works.

Re: how to remove debian in bios menu

[Hans]

Am Samstag, 5. April 2025, 16:54:24 CEST schrieb hlyg:
> i press F12 during Dell boot, a list of options show
> 
> under section "UEFI BOOT:" there are 2 items: FreeBSD and debian
> 
> but all 3 disks installed use mbr, no wonder they don't work
> 
> most probably they are resulted from previous installation on gpt disks
> 
> how to remove them?

I have a DELL Latitude, dual boot (Debian and Windows), should have same BIOS 
than yours.

Go into the BIOS-setup, then look General -> Boot Sequence , on the right side 
you can delete or deativate all unneeded boot entries.

My bootentry is "debian", should be your active one.

Maybe  for someione interesting: As I also have Windows on my drive, there is 
an entry for Windows. I deleted this, because then I only have the entry 
"debian".  And this is tarting grub, which got an entry for Windows.
Dunno, if this is a good way, but it is working for me.

Hope, this helps.

Best 

Hans

Re: OT: Connect two computers with linux with wlan, but without any router

[Joe]

On Sun, 30 Mar 2025 15:40:07 +0200
Hans  wrote:


> 
> Then use NGINX with RTMP-module listening on its standard port and
> streaming with RTMP from Computer A to Computer B to the standard
> port. 
> 
> Everything without any AP or router between.
> 
> The stream can then be made visible with VLC or OBS on Computer B.
>

If it's for RTMP, then the H.26n coding and decoding will cause far more
delay than the networking. VLC running on an average desktop PC seems to
impose about half a second of extra delay compared to a hardware H.26n
decoder.

-- 
Joe

Re: Ethernet interfaces ignoring 'Connect Automatically' setting on Debian 12 (Bookworm)

[Lee]

On Fri, Apr 4, 2025 at 10:17 PM  wrote:
>
> On Thu, Apr 4, 2025 at 11:56 PM Lee wrote:
>
> > Can you try it with Make available to other users ON
>
> Toggling "Make available to other users" to ON solves the problem! First I 
> tested this with the built-in Ethernet adapter. After this was successful I 
> configured this to ON for the other adapters and rebooted the device. Now all 
> the devices honor the "Connect Automatically" setting. Thank you!
>
> I think there is still the question of why this setting must be ON for this 
> to work properly.

I was guessing that you originally configured the network as root and
then logged in as someone else when you wanted to do whatever it was.
So it seemed reasonable that the network wasn't available to anyone
other than root if "Make available to other users" was a no

But like I said - I was guessing.  I haven't looked at the code, so I
don't know.

Regards
Lee

Re: "solved" in message subject

[Max Nikulin]

On 05/04/2025 05:01, Andrew M.A. Cater wrote:

On Fri, Apr 04, 2025 at 10:22:47AM +0700, Max Nikulin wrote:

Again when reading mail, if subject is changed almost completely: "Old" to
"New (was: Old)" to "New" with "(was: ...)" stripped by e.g. Thunderbird or
Emacs; then the thread is split into 3 conversations, threading headers are
not respected.


Bother - is that the fault of the MUA?


I do not think Thunderbird and GNUS do a wrong thing stripping "(was: 
...)" part. (Perhaps they may add a button to restore original variant. 
I have in mind something similar to forgotten attachment warning in 
Thunderbird.) I believe that it is Gmail fault that "New" is split from 
"New (was: Old)". I admit that it is not well known aspect that "(was: 
...)" is treated in a special way by some mailers.


It is not bad per se that thread is split in some cases. However it 
would be better to have cross links between both parts to switch to the 
counterpart and "undo" button to override heuristics.



"Solved" is never removed from subject, making it close to useless if the
message caused continuation of discussion. In this sense messengers and web
forums with their likes and thanks may be better to mark useful messages in
long threads.


This is the real problem: threads here go on for months and years.


Perhaps MUA should strip "solved" subject prefix or suffix from reply 
(offering user to restore it) similar to "(was: ...)" part. I expect 
that in Thunderbird it may be done by an add-on.



(As a side note: The web interface does benefit from [SOLVED] as being visible.


I do not mind that it is useful. It may be recommended to users of mail 
application that do not break threads behind the scene. Unfortunately 
additional suggestion to remove "solved", if they are going to continue 
thread, will be mostly ignored.


P.S. Ideally, "solved" is state of the original question that should 
expressed as a link to the solution. Actual state may be changed after 
both messages have been sent. I can imagine an additional HTTP service 
that allows users to add marks like "solved" an to vote for useful 
messages without SMTP noise. This extended state would be visible in web 
archive and through MUA plugins. Subscribers who do not want to deal 
with new fancy features would continue to use mailing lists in the 
traditional way.


P.P.S. I hope, thread splitting heuristics will be drastically improved 
in coming years by large language models running on both sides: sender 
and recipient. More powerful tools will be able to distill archived 
threads to provide summary with links to useful messages.

Re: how to remove debian in bios menu

[Max Nikulin]

On 05/04/2025 21:54, hlyg wrote:

i press F12 during Dell boot, a list of options show
under section "UEFI BOOT:" there are 2 items: FreeBSD and debian
but all 3 disks installed use mbr, no wonder they don't work
[...]
how to remove them?


I am curious which way you boot Debian if UEFI entry does not work. I do 
not expect to see UEFI options in compatibility (BIOS) mode. Maybe UI of 
Dell UEFI is just more flexible than HP where switching between UEFI and 
compatibility mode requires reboot.


UEFI almost certainly can boot from mbr (DOS) partition, otherwise it 
will be impossible to boot from USB pen drive. Perhaps it is possible to 
make that option working instead of removing it. I admit that specific 
implementation of UEFI may be an obstacle.

Re: DHCP and static addresses, nothing to do with Re: Who:Bookwormv.Trixie

[Jeffrey Walton]

On Thu, Apr 3, 2025 at 6:55 AM Greg Wooledge  wrote:
>
> On Wed, Apr 02, 2025 at 22:28:24 -0500, David Wright wrote:
> > 127.0.1.1 coyote.coyote.den coyote
> > [...]
> > I don't see the point in leaving it there. If you want to send
> > something to coyote.coyote.den, why do you want the LAN address
> > when 127.0.1.1 is just as good. If the line is correct, it does
> > nothing; if it's incorrect, it can cause harm.
>
> I disagree with you here.  The 127.0.1.1 address is a placeholder put
> there by the installer for the more common case where a machine doesn't
> have a fixed LAN IP address.  Most home or workplace computers these
> days will get their addresses from DHCP without a reservation, so their
> internal addresses may vary.
>
> 127.0.1.1 is used when a fixed LAN IP address isn't available.  But if
> a fixed LAN IP address *is* assigned, that should be used instead.

According to hosts(5) man page, EXAMPLE section
():

   # The following lines are desirable for IPv4 capable hosts
   127.0.0.1   localhost

   # 127.0.1.1 is often used for the FQDN of the machine
   127.0.1.1   thishost.example.org   thishost
   192.168.1.10foo.example.orgfoo
   192.168.1.13bar.example.orgbar
   146.82.138.7master.debian.org  master
   209.237.226.90  www.opensource.org

   # The following lines are desirable for IPv6 capable hosts
   ::1 localhost ip6-localhost ip6-loopback
   ff02::1 ip6-allnodes
   ff02::2 ip6-allrouters

I think the idea is, software can always use 127.0.1.1 to find the
host's fully qualified domain name, without the need to know real IP
address. (And what to do with multihomed hosts?)

> In Gene's case, where all the addressing is manually assigned and static,
> using the traditional approach (192.168.x.y coyote.coyote.den coyote)
> is actually preferred.  It allows a single /etc/hosts file to be
> copied across all computers on the LAN without needing to modify it
> on each host.

Jeff

Re: Instalación

[tomas]

On Sun, Mar 23, 2025 at 04:14:15PM +0100, Julio Gil Garcia wrote:
> Buenas tardes, me llamo Julio:

¡Hola, Julio!

Primero: esta lista es en inglés: la mayoría de la gente aquí no
podrá ayudarte. Si prefieres ccomunicar en castellano, hay una
lista sobre Debian en este lenguaje:

  debian-user-span...@lists.debian.org

en el Web:

  https://lists.debian.org/debian-user-spanish/

En cuanto a tu problema...

> Resulta que soy nuevo en Linux y llevo días intentando, infructuosamente,
> instalar Debian, en un portátil, sin quitar el Windows 11 que tiene
> instalado. Es decir organizándolo con arranque doble.
> Es un ordenador chino, marca XIAOMA (Intel i7-8500Y CPU 1.50GHz - 1.10GHz;
> 16 GB de RAM, con disco SSD donde ya preparé una partición de 1TB para
> instalar Debian)
> He seguido las instrucciones para bajar el instalador para Debian 12,
> comprobarlo y preparar un USB (Kingston nuevo de 64GB) con Rufus.
> Tras conectar por cable el ordenador a Internet, el programa inicia, pero
> cuando inicio la instalación, tras elegir idioma, teclado y región, unas
> veces me sale:
> 
> *Se produjo un error al leer los datos del disco extraíble..*

... mi primera sospecha es el stick. La segunda, una copia incompleta
del instalador. Intenta repetir la copia con el Rufus, y si el problema
persiste, inténtalo con otro stick.

No conozco Rufus, pero quizás ofrezca una posibilidad de verificar
la copia en el stick?

Saludos, y suerte
-- 
tomás


signature.asc
Description: PGP signature

Re: Unable to install GRUB in dummy

[hlyg]

Thank Wright!

with more than 20 years of experience in installing and using debian, do 
i have to read installation guide?


it is said that installing debian is as easy as pressing Enter

it is dumb to report fatal error after all configuration and copying

after 30 years of development, installer isn't fool-proof for me

your advanced technique of booting very old laptop is too complex to me

perhaps trixie is about to become stable, i will install it from scratch

Re: X client from Wayland to X11 through ssh

[Eben King]

On 4/4/25 17:00, Eben King wrote:



Also it suspends the OS after a few minutes, so I
gotta find out where that's controlled.


/etc/gdm3/greeter.dconf-defaults looks to be a likely candidate, as in

https://unix.stackexchange.com/questions/748759/disabling-suspend-etc-on-debian-12

I changed it like so:

# Automatic suspend
# =
[org/gnome/settings-daemon/plugins/power]
# - Time inactive in seconds before suspending with AC power
#   1200=20 minutes, 0=never
sleep-inactive-ac-timeout=0

and restarted gdm:

root@alexandria:~# systemctl restart gdm3

so I'll see if that did the trick.