Today I did some more tests and investigations.
First of all I moved to the new mantic ISO image (20230928) that improved
things quite a lot!
The installation (on z/VM so far) is again very snappy and quick,
incl. the enablement of a DASD device in the zDev activation screen.
At the end of the installation, before hitting 'Reboot now' I went to
the console and had a look at top to see if udev related processes are
active, but I couldn't identify any:
top - 10:47:51 up 9 min, 2 users, load average: 0.44, 0.41, 0.20
Tasks: 137 total, 1 running, 136 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3988.9 total, 214.3 free, 1957.6 used, 3233.2 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 2031.3 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
15026 root 20 0 8788 4864 2816 R 0.7 0.1 0:00.28 top
1439 root 20 0 129824 74216 21376 S 0.3 1.8 0:03.76
/snap/subiquity/5183/usr/bin/python3.10 /snap/subiquity/518+
1441 root 20 0 129788 74312 21376 S 0.3 1.8 0:03.57
/snap/subiquity/5183/usr/bin/python3.10 /snap/subiquity/518+
2381 root 20 0 12216 5760 4864 S 0.3 0.1 0:00.10 sudo
snap run subiquity
2383 root 20 0 206204 77680 21632 S 0.3 1.9 0:04.96
/snap/subiquity/5183/usr/bin/python3.10 -m subiquity
1 root 20 0 103300 13676 8940 S 0.0 0.3 0:03.49
/sbin/init ---
2 root 20 0 0 0 0 S 0.0 0.0 0:00.00
[kthreadd]
3 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 [rcu_gp]
4 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
[rcu_par_gp]
5 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
[slub_flushwq]
6 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 [netns]
8 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
[kworker/0:0H-events_highpri]
9 root 20 0 0 0 0 I 0.0 0.0 0:00.00
[kworker/0:1-cgroup_destroy]
10 root 20 0 0 0 0 I 0.0 0.0 0:01.11
[kworker/u128:0-events_unbound]
11 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
[mm_percpu_wq]
12 root 20 0 0 0 0 I 0.0 0.0 0:00.00
[rcu_tasks_rude_kthread]
13 root 20 0 0 0 0 I 0.0 0.0 0:00.00
[rcu_tasks_trace_kthread]
Then, having the post-install reboot completed, and looking at top, I
can observe the udev activities:
top - 10:55:26 up 6 min, 1 user, load average: 2.16, 1.75, 0.84
Tasks: 108 total, 5 running, 103 sleeping, 0 stopped, 0 zombie
%Cpu(s): 18.1 us, 22.7 sy, 0.0 ni, 48.8 id, 8.1 wa, 0.6 hi, 0.8 si, 0.9 st
MiB Mem : 3988.9 total, 3274.2 free, 271.4 used, 539.0 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 3717.5 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1177 root 20 0 24884 5724 2560 R 33.2 0.1 1:50.24
(udev-worker)
1 root 20 0 168592 12912 8432 R 20.6 0.3 1:07.93
/sbin/init
690 root 20 0 467964 13112 10752 D 15.6 0.3 0:52.91
/usr/libexec/udisks2/udisksd
660 message+ 20 0 9328 4480 3840 S 14.6 0.1 0:52.68
@dbus-daemon --system --address=systemd: --nofork --nopidfile -+
16285 ubuntu 20 0 18004 9984 8448 S 10.6 0.2 0:33.06
/lib/systemd/systemd --user
385 root 20 0 24764 7492 4676 S 9.3 0.2 0:30.40
/lib/systemd/systemd-udevd
373 root rt 0 288412 26368 8064 S 6.6 0.6 0:21.50
/sbin/multipathd -d -s
686 root 20 0 16096 7424 6528 S 4.0 0.2 0:13.20
/lib/systemd/systemd-logind
681 root 20 0 2066332 34480 21248 S 1.7 0.8 0:05.48
/usr/lib/snapd/snapd
14 root 20 0 0 0 0 S 0.7 0.0 0:01.90
[ksoftirqd/0]
20 root 20 0 0 0 0 S 0.7 0.0 0:01.87
[ksoftirqd/1]
25 root 20 0 0 0 0 S 0.7 0.0 0:01.27
[ksoftirqd/2]
333 root 19 -1 56736 23936 23040 S 0.7 0.6 0:03.42
/lib/systemd/systemd-journald
30 root 20 0 0 0 0 S 0.3 0.0 0:01.32
[ksoftirqd/3]
104 root 0 -20 0 0 0 I 0.3 0.0 0:00.55
[kworker/1:1H-kblockd]
138 root 0 -20 0 0 0 I 0.3 0.0 0:00.39
[kworker/2:1H-kblockd]
256 root 20 0 0 0 0 S 0.3 0.0 0:00.55
[jbd2/dasda1-8]
Did the same (on the exact same system) with jammy 22.04.3 - nothing
suspicious about udev, also not after post-inst reboot.
On mantic I've noticed a flood of warning messages like:
root@hwe0003:~# udisksctl monitor
** (udisksctl monitor:424911): WARNING **: 11:14:33.100:
(udisksctl.c:2811):monitor_on_interface_proxy_properties_changed: runtime check
failed: (g_strv_length ((gchar **) invalidated_properties) == 0)
^C
(endless repeating messages)
which I think fits to your observations, Dan.
I've search for any already known udisks2 issue upstream, but couldn't
find anything that fits ...
Interestingly I get everything back to normal with just a 'systemd-udevd'
restart:
# systemctl restart systemd-udevd
top - 12:16:45 up 20 min, 3 users, load average: 0.00, 0.33, 0.72
Tasks: 110 total, 1 running, 109 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 3988.9 total, 3260.3 free, 274.9 used, 549.6 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 3714.0 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
166794 root 20 0 11784 5248 3200 R 0.7 0.1 0:00.65 top
1 root 20 0 103052 13040 8560 S 0.0 0.3 1:49.42 systemd
2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kthreadd
3 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 rcu_gp
4 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
rcu_par_gp
5 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
slub_flushwq
6 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 netns
8 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
kworker/0:0H-events_highpri
9 root 20 0 0 0 0 I 0.0 0.0 0:00.39
kworker/0:1-rcu_par_gp
11 root 0 -20 0 0 0 I 0.0 0.0 0:00.00
mm_percpu_wq
12 root 20 0 0 0 0 I 0.0 0.0 0:00.00
rcu_tasks_rude_kthread
13 root 20 0 0 0 0 I 0.0 0.0 0:00.00
rcu_tasks_trace_kthread
14 root 20 0 0 0 0 S 0.0 0.0 0:02.99
ksoftirqd/0
15 root 20 0 0 0 0 I 0.0 0.0 0:00.16
rcu_sched
16 root rt 0 0 0 0 S 0.0 0.0 0:00.00
migration/0
17 root 20 0 0 0 0 S 0.0 0.0 0:00.00 cpuhp/0
18 root 20 0 0 0 0 S 0.0 0.0 0:00.00 cpuhp/1
After a reboot, I'm hitting the same situation again:
top - 12:23:24 up 1 min, 1 user, load average: 2.48, 0.86, 0.31
Tasks: 138 total, 4 running, 134 sleeping, 0 stopped, 0 zombie
%Cpu(s): 19.7 us, 24.1 sy, 0.0 ni, 47.8 id, 6.0 wa, 0.6 hi, 0.9 si, 1.0 st
MiB Mem : 3988.9 total, 3527.3 free, 285.1 used, 268.7 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 3703.8 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
375 root 20 0 25096 5860 2816 R 33.6 0.1 0:30.19
(udev-worker)
1 root 20 0 101752 11760 8560 S 20.9 0.3 0:19.96
/sbin/init
591 root 20 0 467924 13060 10752 R 17.9 0.3 0:16.32
/usr/libexec/udisks2/udisksd
558 message+ 20 0 9316 4608 3840 S 16.9 0.1 0:15.39
@dbus-daemon --system --address=systemd: --nofork --nopid+
12185 ubuntu 20 0 17996 9984 8448 S 12.3 0.2 0:07.06
/lib/systemd/systemd --user
373 root 20 0 24116 6784 4608 S 9.6 0.2 0:09.17
/lib/systemd/systemd-udevd
358 root rt 0 288412 26624 8192 S 6.6 0.7 0:06.34
/sbin/multipathd -d -s
589 root 20 0 16268 7424 6528 S 4.3 0.2 0:04.15
/lib/systemd/systemd-logind
585 root 20 0 1918612 31388 20352 S 2.0 0.8 0:01.89
/usr/lib/snapd/snapd
324 root 19 -1 40344 13880 13112 S 1.0 0.3 0:01.04
/lib/systemd/systemd-journald
14 root 20 0 0 0 0 S 0.7 0.0 0:00.45
[ksoftirqd/0]
20 root 20 0 0 0 0 S 0.3 0.0 0:00.43
[ksoftirqd/1]
25 root 20 0 0 0 0 S 0.3 0.0 0:00.47
[ksoftirqd/2]
30 root 20 0 0 0 0 S 0.3 0.0 0:00.43
[ksoftirqd/3]
36 root 20 0 0 0 0 I 0.3 0.0 0:00.10
[kworker/0:2-rcu_gp]
61 root 20 0 0 0 0 I 0.3 0.0 0:00.17
[kworker/u128:2-flush-94:0]
71 root 20 0 0 0 0 I 0.3 0.0 0:00.07
[kworker/3:1-events]
Restart (or stop / start) of "systemd-udevd" helps again.
I'm wondering now:
- what is different during the installation that prevents udev from going crazy?
(is systemd-udevd for example restarted during the installation at some point
in time?)
- and why is the initial start of systemd-udevd causing a different behavior
than after it's its restart? (is there s/t missing when it's started at boot
time?)
--
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to udisks2 in Ubuntu.
https://bugs.launchpad.net/bugs/2037569
Title:
udev issues with mantic beta
Status in Ubuntu on IBM z Systems:
Confirmed
Status in systemd package in Ubuntu:
Confirmed
Status in udisks2 package in Ubuntu:
Confirmed
Bug description:
While installing mantic beta (on s390x, LPAR and z/VM - but this might not be
architecture specific) I faced issues with udev.
In my installation I've updated the installer to "edge/lp-2009141" (subiquity
22.02.2+git1762.1b1ee6f4 5164).
During my installations I first noticed bad response times in case of
dealing with devices (like enabling new devices with chzdev). chzdev
is used during the installation, hence the installation procedure is
also affected by this. (I mainly notice this issue in case of DASD
ECKD disk enablements.)
But even after after a successful (but due to this issue less snappier)
installation, means after the post-install reboot, in the installed system I
can find several udev related processes, like:
69448 root 20 0 31280 11944 2560 S 39.2 0.0 2:51.67
(udev-worker)
509 root 20 0 31276 13812 4600 S 20.6 0.0 2:07.76
systemd-udevd
893 root 20 0 469016 13544 10496 R 17.3 0.0 1:43.53
udisksd
1 root 20 0 168664 12748 8396 S 16.3 0.0 1:40.47
systemd
which is not only unusual, but (as one can see) they consume quite some
resources.
Even the remote ssh into that system is impacted by this high load.
So far I only see this in mantic.
I tried 20.04.3 as well as lunar, but both do not seem to be affected by this
udev problem.
I neither face the bad response on device enablement, nor can see any udev
related processes still running after post-install-reboot in the installed
system.
(Sometimes I could also see a growing log file 'syslog').
I cannot say yet what is causing this, but since I see 'systemd-udevd'
as prominent process in top, I'll first of all mark this as affecting
systemd-udevd (or systemd).
I've attached the outcome of some more investigations I did ...
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu-z-systems/+bug/2037569/+subscriptions
--
Mailing list: https://launchpad.net/~desktop-packages
Post to : desktop-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~desktop-packages
More help : https://help.launchpad.net/ListHelp
