Adding some context:
The function which throws the warning is here:
https://searchfox.org/mozilla-central/source/toolkit/xre/glxtest/glxtest.cpp#192
RHEL added libpci as dependency:
https://bugzilla.redhat.com/show_bug.cgi?id=1955338
It was reported on BSD derivatives and Solaris, where
Hi Steven,
indeed that might be part of the issue, since libuuid1 "Provides:
libuuid1t64 (= 2.39.3-9)", hence an exact version of libuuid1t64 which
is not in repos. debootstrap otherwise might be able to solve the
libuuid1t64 dependency by installing libuuid1 only.
Interesting is that
Package: libuuid1t64
Version: 2.39.3-6.1
I am currently unable to generate riscv64 images via debootstrap:
---
I: Extracting libuuid1...
I: Extracting libuuid1t64...
E: Tried to extract package, but file already exists. Exit...
---
Both packages obviously provide the same files. "apt
Package: ifupdown
Version: 0.8.41
On Orange Pi Zero 3, when using ifup to configure the Ethernet adapter,
after doing a (soft) reboot, the interface is not present anymore. With
NetworkManager, this is not the case. It is a bug in the vendor kernel
build/driver for sure, just triggered by
Package: systemd
Version: 254.1-2
since systemd 253 was merged into Sid/unstable and Trixie/testing,
systemd-nspawn fails to boot Sid and Trixie containers with foreign
architectures via qemu-user-static and binfmt:
---
Spawning container rootfs on /root/rootfs.
Press Ctrl-] three times
Package: systemd
Version: 252.12-1
Since Debian Bookworm (systemd 252), the reboot, poweroff and "shutdown
now" commands throw the following error if the dbus service is not running:
---
Failed to connect to bus: No such file or directory
---
and if dbus is running, but
Is it planned or would it be possible to have a patch with the fix for
Bookworm (Deluge v2.0.3)?
Best regards,
Micha
Who do you mean? Should be reported here by OP who faced the issue:
https://github.com/puppetlabs/puppetlabs-mysql/issues
Only the Puppet MySQL module developers will know whether they need
plain text logging for some specific reason and how to hence address
this best.
Bookworm (and hence
The Puppet module requires an update:
https://github.com/search?q=repo%3Apuppetlabs%2Fpuppetlabs-mysql%20%2Fvar%2Flog=code
One option would be to make it create and chown the directory by itself,
if plain text logs are for some reason needed/wanted.
The probably cleaner option would be if it
The choice to not add the xrdp user to the ssl-cert group OOTB decreases
connection security unnecessarily until done manually.
Furthermore, it doesn't make sense to make the ssl-cert package a strict
dependency, if it isn't usable by XRDP without manual setup.
If the opinion here is "if you
Package: wireless-regdb
Version: 2022.04.08-2~deb11u1
I just faced the same cfg80211 error, trying to switch from CRDA to the
kernel builtin feature. It took me a while to find this bug report and
about the need to switch to the upstream version of the database. Thanks
for the pointer, and my
I opened a merge request to solve this:
https://salsa.debian.org/apache-team/apache2/-/merge_requests/36
Best regards,
Micha
Thanks for the link with the contained patch. Bumping this as it still
would be great to see it merged:
---
diff FileDir.cpp FileDir_fix.cpp
569c569
< if (mkdir( name, 0700 ) == 0) bret = true;
---
> if (mkdir( name, 0777 & gbl_umask.mask ) == 0) bret = true;
---
Best regards,
Micha
The issue has been solved upstream with motion 4.5 (probably 4.4
already). Since unstable still serves motion 4.3.2, I hope unstable and
then testing can be migrated to the latest motion release.
Bookworm first freeze starts in January, and it would be a pity to have
motion not included with
Great, also solved here, many thanks!
Best regards,
Micha
Package: bluez-alsa-utils
Version: 3.0.0-2
I'm very happy to finally see BlueALSA provided by the Debian package
repository, many thanks to the maintainers!
It is probably too late for my suggestion, and maybe there is a specific
reason why the service is called differently compared to
Package: whiptail
Version: 0.52.21-5
Since Debian Bookworm, it is not possible anymore to control whiptail
dialogs when being passed through a sudo pipe to the shell, e.g.:
---
echo 'whiptail --msgbox test 20 80' | sudo bash
---
This is often used when executing remote installer
I also just verified that both are installed, ah via chromium-common
indeed as dependencies.
> I have a rock64 board lying around unused, I'll have to hook it up to
a monitor and see what happens on there with xfce and chromium.
That would be great. While we tested it with installed
I'm not sure about the exact reasons why these users use exFAT. When
aiming to mount a filesystem on Windows as well, NTFS performs just as
bad and vfat has file and filesystem size limitations, so I can imagine
cases where one needs to stick with exFAT.
Some of the SBCs have a quite powerful
I would also love to be able to ship recent Linux versions in all cases,
but as of sadly common practice of many SoC and SBC manufacturers, to
not produce open source hardware/firmware/drivers, and to provide own
outdated kernel versions with driver and firmware blobs instead of
Package: exfat-fuse
Version: 1.3.0-2
Since Debian Bullseye (also tested on Bookworm), the exfat-fuse package
does not provide the (FUSE-based) "exfat" filesystem driver, i.e. the
following commands fail if the exfat kernel module is not present, e.g.
when a different kernel is used, often
Package: chromium
Version: 99.0.4844.74-1~deb11u1
> Which boards did you test with?
We observed the issue on these boards:
- Odroid C2
- Odroid C4
- Odroid N2+
- Radxa ROCK Pi 4
- Radxa Zero
> What desktops were you using? Was this under X or wayland? Does it
make a difference if you run
I made the request upstream as advised:
https://bugzilla.mindrot.org/show_bug.cgi?id=3397
Best regards,
Micha
Package: chromium
Version: 98.0.4758.102-1~deb11u1
Latest Chromium on Debian Bullseye fails to start on aarch64/arm64
systems. Different ones with different kernels were tested, though all
Linux 5.10 variants, using (and generally supporting) modesetting DDX.
Only v98 from security repo is
I tested the new package v5.0.3-1 from Debian unstable on x86_64 systems
with Debian unstable, Bookworm, Bullseye and even Ubuntu Focal (GitHub
actions runners), and it works very well on all of them. I tested the
"ocs-sr savedisk" command, using some different command-line options,
Dear Georges,
many thanks for your quick action. I'll test the new package once it is
available in unstable.
Best regards,
Micha
Package: clonezilla
Version: 3.35.2-3
Dear maintainer,
the last Clonezilla upstream release has been applied to Debian a long
time ago. In the meantime there have been many releases, latest at time
of writing bring v5.0.3:
Package: openssh-server
Version: 1:8.8p1-1
Currently the standalone OpenSSH sftp-server is used as default SFTP
subsystem, set via /etc/ssh/sshd_config. This implies a dependency on
the openssh-sftp-server package and means that every SFTP connection
spawns a new external process, while
Package: chromium
Version: 98.0.4758.102-1~deb11u1
Ah sorry, my bad, the errors do show up on version 98.0.4758.102-1~deb11u1.
So then it is not an upstream issue but a difference between the build
or environment on Bullseye vs Bookworm.
Best regards,
Micha
Package: chromium
Version: 90.0.4430.212-1
Hey guys,
I recognised that Chromium on Bullseye is constantly trying to access
the CPUFreq API, even when it runs within a VM where this API is
expected to be not available. This triggers constant error messages:
---
*** stack smashing detected
Package: firefox-esr
Version: 78.15.0esr-1
I can confirm this to be an issue on Debian Bullseye with v78.15.0esr-1
already.
Would be great if the maintainer could add the libpci3 dependency for
firefox-esr on Bullseye and Bookworm.
Best regards,
Micha
Package: qemu-guest-agent
Version: 1:5.2+dfsg-11+deb11u1
When using qemu-guest-agent within a Proxmox VM running Debian Bullseye,
we recognised that sending the shutdown command from the host only has
any effect when systemd-logind run within the VM and hence dbus is
installed. The latter
Huh, what spam bot made it in here?
However,
@Stefan as I see MS Internet Explorer support being dropped by more and
more projects, do you agree to have the MSIE [2-6] comment blocks simply
removed, together with the mod_setenvif dependency? If so, let me know
and I can send a merge request,
We want to offer OS images for Parallels desktop (virtualizer on macOS).
So far we couldn't find a way to generate them via CI/CD pipeline on a
Linux host, but need to import an image or appliance (for another
virtualizer) into Parallels on a macOS system to have it doing the
conversion to
Package: qemu-utils
Version: 1:6.1+dfsg-8+b2
Aiming to automate VM image creation, I recognised that on Debian,
qemu-img does not support the Parallels virtualizer hds virtual disk
image format. Generally it seems to be supported by QEMU, like here on
RHEL 7:
Indeed it would be great if this patch could be applied to the Deluge
v2.0.3 package on Debian Bullseye. Here the related pull request, in
case it contains additional information:
https://github.com/deluge-torrent/deluge/pull/271
Best regards,
Micha
Adding another information:
I retested it on Debian Buster and Stretch and the same error prompt
appears when LXDE is not started from the actual screen/console but when
calling "startx" from e.g. an SSH session, which then start LXDE on the
main screen. The solution/workaround is the same.
The problem is still present on Debian Bullseye when simply doing "apt
install lxde" and "startx".
When additionally "apt install lxsession-default-apps", it can be seen
that "lxpolkit" is set as default "Polkit agent" and clearing that field
resolves the issue.
The thing is now that
raspbian is not maintained by Debian, fwiw.
Indirectly it is, since Raspbian uses the unmodified (at least in all
cases I'm aware of) Debian package sources and ships the same updates
shortly after they have been pushed to the Debian (source) repositories
(including debian-security), just
Okay, first I verified that it is an issue on all architectures, even on
Raspberry Pi with the dedicated Raspbian armv6hf package builds.
Then I sadly found that downgrading all systemd source packages to
232-25+deb9u12 does not solve the issue. Also I wasn't able to find
another recent
The LTS team has no interest in fixing this IMHO major usage bug on the
init system?
Since we didn't see this earlier among user user base, chances are high
that it has been caused by the last package update (security repo) from
08 Jul 2021.
I'll try to verify this by pulling an earlier
Package: systemd
Version: 232-25+deb9u13
On current Debian Stretch, the systemctl --now flag, when enabling or
disabling a service, has no effect in those two cases:
1. The service is already enabled respectively disabled: On newer
systemd versions (Buster and Bullseye), the service is still
Hi,
many thanks for your reply and linking that FAQ, this clarifies things.
I also found the linked older bug report about the very same question:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=825945
I personally would prefer more compatibility here, possibly combined
with a flag to
on FAT partitions (and other filesystem
types with no link support) gracefully. A backup could be created
differently if this is seen mandatory.
Best regards,
MichaIng
/MichaIng/DietPi/issues/4690
Best regards,
Micha
Hi Daniel,
that is great news, many thanks for your work and quick response!
Best regards,
Micha
Am 28.05.2021 um 07:21 schrieb Daniel Baumann:
tag 989178 pending
thanks
On 5/27/21 7:24 PM, MichaIng wrote:
I hence hereby request/wish for a backport of Netdata v1.29.3, provided
and tested
Package: netdata
Version: 1.12.0
Severity: wishlist
Hey guys,
while Debian Buster ships Netdata v1.12.0, the upstream project made
large progress and v1.31.0 has been released in the meantime. Especially
the Netdata Cloud feature, introduced with Netdata v1.22.0, is a notable
extension which
I think "all" cannot be used like that.
From the docs:
https://httpd.apache.org/docs/2.4/mod/mod_ssl.html#sslprotocol
"all" is an alias for "+SSLv3 +TLSv1 +...", so while the raw config file
syntax is fine, internally "-all" is probably seen as invalid and
replaced by the default
Am 19.04.2021 um 23:49 schrieb Stefan Fritsch:
Am 18.04.21 um 17:10 schrieb MichaIng:
The dependency is declared in /etc/apache2/mods-available/ssl.load and
is a Debian packaging thing, not relevant to upstream.
Now I see the "comment" in the .load files, thanks for the hint!
I th
Package: apache2
Version: 2.4.46-4
When enabling mod_ssl via "a2enmod ssl" while mod_setenvif is disabled,
it becomes apparent that mod_ssl depends on mod_setenvif. I checked the
configuration file /etc/apache2/mods-available/ssl.conf, but it actually
does not use any mod_setenvif directive.
Package: p7zip
Version: 16.02+dfsg-8
When extracting a 7z archive to an output directory which does not yet
exist, 7zr creates this directory automatically. While for all files and
directories within the archive, the default umask modes are applied, the
parent output directory is created
Sorry for not checking the man page, it is all clear then.
I addressed this as feature request upstream.
Best regards,
Micha
Package: alsa-utils
Version: 1.1.8-2
Little addition: The flag is ignored only in combination with the "init"
command, while with e.g. "save" the error message (and exit code) is
indeed omitted.
It might hence be intended the way it is, but I would suggest to expand
the effect of
Package: alsa-utils
Version: 1.2.4-1
And: 1.1.8-2
Hey guys,
I recognised that alsactl is ignoring the "--ignore" flag:
---
# alsactl --help | grep ignore
-g,--ignore ignore 'No soundcards found' error
# alsactl init
alsactl: init:1759: No soundcards found...
# alsactl -g init
Many thanks for your efforts Daniel, sadly this remains an issue.
It would be great to allow some collaboration on a public repository,
especially on such an essential package that bash is. It might save
everyone's time when it was possible to base a discussion on an actual
MR/PR instead of
Package: alsa-utils
Version: 1.1.8-2
Severity: important
In short:
Please release 1.1.8-3 it to Debian Buster!
1.1.8-3 indeed solves the issue from what I can see for Debian Bullseye.
Also the original issue has been solved, the currently available
alsa-utils package (on Bullseye) does not
Package: apt-show-versions
Version: 0.22.11
I just recognised the same with xz-compressed list files.
I agree that best is probably to rely on APT's internal cache
commands/functions instead of accessing the list files directly, so no
additional extraction needs to be implemented and
Since two months have been passed, may I ask about the status of this
request? The requested debdiff has been attached above and these patches
fix potential and actual security issues in backend applications which
can be hard to debug or even recognise (meaning a security issue exists
without
Hey guys,
we face the same issue whenever a ready-to-run image has been created
and booted from a different device.
From what we found, the Debian installer (or the grub install/config
scripts) stores the hardware ID of the drive to debconf database, if one
is available. Of course on a
Package: htpdate
Version: 1.2.2-3
Hi guys,
the issue has btw not been solved from what I can see and test, but is
still present on Buster backports as well as on Bullseye and Sid.
The reason is "InaccessibleDirectories" option in the systemd unit
"/lib/systemd/system/htpdate.service". For
, line
786, in resolve
raise DistributionNotFound(req, requirers)
pkg_resources.DistributionNotFound: The 'requests' distribution was not
found and is required by cloudprint
Best regards,
MichaIng
with a
systemd unit (and no sysvinit service), hence expects systemd to be
present and logs to systemd-journald in the first place.
And finally tests prove that systemd-journald as only syslog daemon does
not break cloudprint binary or service.
Best regards,
MichaIng
Package: wakeonlan
Version: 0.41-12
Issue still present in current version. As long as wakeonlan uses the
mentioned perl functions for its very basic functionality, it
practically depends on the netbase package.
Best regards,
MichaIng
Package: libapache2-mod-php7.0
Version: 7.0.33-0+deb9u7
Hey guys,
I wanted to report an issue that we faced some time ago:
https://github.com/MichaIng/DietPi/issues/2797
The postrm script calls "php_invoke dismod 7.0 apache2 $mod" to disable
the modules, but this fails if either
it.
Best regards,
MichaIng
Am 28.12.2019 um 21:37 schrieb Doug:
Most people don't use vmtouch as a service, instead as an interactive
command-line tool. If you use the daemon functionality in the deb, can
you please send a pull request to our github repo?
https://github.com/hoytech/vmtouch
to sbin seems to be the worse alternative solution.
Best regards,
MichaIng
epos and matches per
naming the recommendation "gstreamer1.0-plugins-good" and second
suggestion "gstreamer1.0-plugins-bad".
Nothing serious but fix should be straight forward.
Best regards,
MichaIng
is always possible with this option as expected and as well
the issue does not appear on selection fields of --menu and --checklist,
which have their own scroll bars, when required, which show up fine.
Screenshots can be found here:
https://github.com/MichaIng/DietPi/issues/2947
Package: tigervnc-standalone-server
Version: 1.9.0+dfsg-3
When executing tigervncserver, while the netbase package is not
installed, it throws the following error message multiple times:
--
Use of uninitialized value $proto in socket at /usr/bin/tigervncserver
--
This is due
69 matches
Mail list logo
