== Summary ==
On my computer with an ancient Nvidia chipset (Geforce 7025/nForce 630a), running `sysctl suspend` (or suspending from the Gnome menu) causes the system to start suspending, but it freezes halfway, leaving fans and hard drives spinning. There’s no way to resume from this frozen state apart from forcing a reboot (with a hardware reset button/poweroff).

== Steps to reproduce ==
* boot with modeset enabled
* run `sysctl suspend`

== What I expect to happen ==
For the system to suspend, shutting down all fans and hard drives.

== What happens ==
The system begins to suspend, but freezes halfway, leaving the display on and fans and hard drives spinning, but the keyboard unresponsive.

== Workaround ==
Disable kernel modesetting by adding ”nomodeset” to the kernel commandline.

== Affected kernels ==
Prior to upgrades the system was running HWE kernel 5.15.0, so I tried the 5.15 series, and found that I could now suspend and wake the machine again just as before. I worked my way up the versions:

* 5.15.50: unaffected
* 5.15.165: unaffected
* 5.19.17: unaffected
* 6.4.0: unaffected
* 6.6.0: unaffected
* 6.6.48: unaffected
* 6.7.0: first to fail

I also tried the current newest mainline kernel 6.10.7, and the issue is still present there.

== Background ==
I have an old desktop machine now functioning as a NAS, and yesterday I upgraded it from Ubuntu 20.04 first to 22.04, and then all the way up to 24.04. The upgrade went smoothly, and this is the only issue I’ve come across since.

In the BIOS settings of the affected machine there are three ”suspend mode” alternatives to choose from: ”S1 (POS) only”, ”S3 only” and ”Auto”. I’ve always had it on ”Auto”, but with this issue I also tested both ”S1 only” and ”S3 only”, with no effect.

The issue is also present when booting from the installation media (USB) into a live environment.

I’ve previously upgraded my laptop to 24.04, and there suspending still works as it did before the ugprade, so this is probably hardware-specific; the laptop is a modern one with all-Intel hardware.

Googling around, I could smelled hints of this being once again related to the troublesome Nvidia chipset, so I tried nomodeset with the stock 6.8.0 kernel (6.8.0-41 currently) and voilà! Suspend and wake were working again.

Well, except for the display, which stayed black. But I couldn’t say if this was the way it was before, because the NAS is normally running headless.

== System information ==
Ubuntu 23.10, up to date before today’s updates.

== Steps to reproduce (WIP) ==
0. have a group of packages that are kept back from updating, plus some that aren’t (screenshot demonstration attached) below
1. open update-manager
2. (try to) open the kept group, click on their (unchecked) checkboxes
3. do the same for some of the other, non-kept packages
4. repeat 2.-3. for a while

== What happens ==
Update-manager becomes unresponsive, then eventually segfaults:

update-manager[3292]: segfault at 30 ip 00007fc0718b4c41 sp 00007ffc2cca6a68 error 4 in libgobject-2.0.so.0.7800.0[7fc071889000+34000] likely on CPU 4 (core 0, socket 0)

== Other info ==
I’ve marked the steps above WIP because these are the conditions under which I found this today, but I haven’t yet tried narrowing them down, because of course I have little control over which updates are available. (For instance, the kept packages might just be a red herring.)

I initially tried to report the crash file with ubuntu-bug, but despite asking preliminary questions, in the end it never opened a browser to complete the report. I’ll try apport-collect next, but if that fails too, I’ll just attach the crash file here manually.

In fact the man page seems to be completely out of sync with reality. None of minage, maxage, minsize and maxsize appear to work at all despite being documented there. Minage is read at least (according to debug output), but then still ignored. Specifying minsize or maxsize has no effect, logs are rotated when they hit 1 M regardless. At least the plain `size` directive does appear to work as documented.

The man page for logrotate says ”Each configuration file can set global options (local definitions override global ones, and later definitions override earlier ones)”. This is inconsistent with how it actually behaves.

== Steps to reproduce ==
$ logrotate –version
logrotate 3.14.0

Default mail command: /usr/bin/mail
Default compress command: /bin/gzip
Default uncompress command: /bin/gunzip
Default compress extension: .gz
Default state file path: /var/lib/logrotate/status
ACL support: yes
SELinux support: yes
$ mkdir /tmp/logrotest
$ cd /tmp/logrotest
$ touch test.log
$ touch other.log
$ cat >logrotate.conf
/tmp/logrotest/*.log {
minsize 5M
}

/tmp/logrotest/test.log {
minsize 1M
}
$ logrotate –state /tmp/logrostate logrotate.conf

== What happens ==
error: logrotate.conf:5 duplicate log entry for /tmp/logrotest/test.log

== What I expect to happen instead ==
No error, test.log being processed with its specific, later-defined directives.

@Julian: I’d argue that if a simple accidental click somewhere can cause this, that’s a loaded gun pointed at the user’s toe, and hence a bug in the installer.

@Brian: I’ll try the 22.04 images. I’ve kept trying with 20.04.5, but have yet to find another instance of this occurring, so it’s obviously not easy to trigger.

I’m not particularly troubled by this issue (and the original reporter doesn’t seem to be either), so feel free to adjust ’Importance’ accordingly.

installer.tar.xz Edit (396.6 KiB, application/x-tar)

@Olivier: Sure. This VM was just for testing purposes, so there shouldn’t be anything confidential here.

Screenshot from 2022-09-26 21-01-16.png Edit (159.6 KiB, image/png)

Looks like the deb line for updates is missing entirely from sources.list in this install; only the deb-src is there.

@Steve I did not.

Screenshot from 2022-09-26 19-36-11.png Edit (149.8 KiB, image/png)

@Simon Here goes, although now it just shows both at 9.7, and 9.9 not yet available. Before I downgraded libc6 it obviously showed libc6 at 9.9.

I also tried to reproduce the issue in a fresh VM, but those kept getting 9.9 for all the packages, so installing build-essential caused no issues.

I don’t have in-depth knowledge about phased updates, but this looked like as if downloading the updates during installation was not yet affected by phased updates (and hence got libc 9.9), but, after the installation was completed, the system got cast into a ”still at 9.7” group of phased updates, and so libc6-dev (et al) could not be installed.

Just got hit by the same issue in a VM after a fresh install of 20.04.5. I chose ”Minimal installation”, ticked ”Download updates while installing”, and installed the remaining updates post-install.

After that the first thing I tried to do was to install build-essential, which failed, because the version of libc6 was out of sync with the rest of the libc6* packages: libc6 was already at 2.31-0ubuntu9.9, while e.g. libc6-dev was still only available as 2.31-0ubuntu9.7.

I tried switching from my local archive (fi.archive.ubuntu.com) to the main one (archive.ubuntu.com), but that made no difference. I had to downgrade libc6 to 2.31-0ubuntu9.7 to be able to install build-essential.