Sorry for the late reply, I have not been subscribed to this bug. I have
not written earlier because the bug was limited to one location (it
seems in connection with one wireless network made available to all
users and logging in to the network before logging into the user
account), and I wanted
Thanks a lot for your feedback Marcelo; I'm glad everything seems fine now.
Cheers,
Guilherme
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in
Dear Guilherme,
I have a mostly vanilla Ubuntu 14.04.5, right now with kernel
3.13.0-145.
I've just tested the new "204-5ubuntu20.28" packages from "-proposed",
and they run fine, with and without the latest "cgmanager". No problems
detected.
Thank you for your support.
--
You received this
For the folks that observed the recent cgmanager issue with logind:
there's a new systemd version in -proposed; it's version
"204-5ubuntu20.28". I've just tested it with and without cgmanager, in
kernels 4.4.0-119 and 3.13.0-145 (according to LP #1750013), and didn't
observe any constant CPU
Folks that are experiencing this issue: the best way to circumvent it
for now I guess it's downgrade systemd package to version
204-5ubuntu20.26 and remove cgmanager.
To remove cgmanager: "sudo apt-get remove cgmanager"
To downgrade systemd version: "sudo apt-get install systemd-
@serge-hallyn, I've attached output of as requested.
I hadn't subscribed to notification. That's now been fixed. Missed your
request when you posted it.
Happy to do more.
** Attachment added: "trace-cgmanager-ubuntu-14.04.5-00.txt"
Same problem as well strace on cgmanager attached.
** Attachment added: "strace-cgmanger.log"
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1303649/+attachment/5093263/+files/strace-cgmanger.log
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
Strace on systemd-logind as well.
** Attachment added: "strace-systemd-logind.log"
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1303649/+attachment/5093268/+files/strace-systemd-logind.log
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
I was also experiencing cpu overload for cgmanager (up to 99%) then I
disabled it in grub's default (with "cgroup_disable=cpu") but systemd-
logind started eating CPU (80%) instead.
I see nothing strange in the logs, but if I can be of any help by
providing any information please just ask me.
Could you find the pid of cgmanager ( 353 below) and do
Strafe -f -p 353 -o trace.txt
for maybe 5 seconds, ctrl-c it, and attach trace.txt here?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
FWIW I'm experiencing similar symptoms on an ancient desktop that has
been stable, reliable and used every day for years. I too ran dist-
upgrade a couple of days ago. I've no idea what repos I use. That config
hasn't been touched for a long time. Can investigate if the information
might be
I cca two days the same problem. Two Core CPU's all time on 60% and nm-
applet not shown on panel:
PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND
459 root 20 0
Hofer-Temmel Christoph,
I've got the binary packages for the 204-5ubuntu20.25 version from
Launchpad itself. The pages for the x86 versions you need are:
https://launchpad.net/ubuntu/trusty/i386/libpam-systemd/204-5ubuntu20.25
I am having the same issue since this morning (14 march): cgmanager and
systemd-logind both using 60-70% of system resources, gui unusable, login via
console possible.
Looking at /var/log/apt/term.log I see that the update has been installed on 9
march, but did not give any issues until now.
Guilherme,
Yes, I am using 'backports' too.
However, the 'cgmanager' version that was automatically installed was
'cgmanager_0.24-0ubuntu7.5_amd64', and it is still available to me.
Thank you for sorting this!
--
You received this bug notification because you are a member of Ubuntu
Bugs,
Yes, I was using -backports. I dumped a list of everything I had
installed before I reinstalled, and I had cgmanager/trusty-updates
0.24-0ubuntu7.5 installed right before the I wiped my system. I was
still experiencing the problem.
--
You received this bug notification because you are a member
Thanks Dale and Marcelo for your quick report - this is really useful.
The problem seem to be caused due to cgmanager being added as dependency
of systemd in -proposed - this request was clearly explained in LP
#1750013 (not a dup for this, it's another issue with systemd-logind).
During my test
Serge,
I am not seeing anything like this with 16.04.4. Since I did a full
wipe and reinstall, I didn't expect to. Yes, I definitely saw it with
14.04.5 with "proposed" enabled. And the packages Marcelo indicated
were among the most recently installed. I think you can treat his
system as
Serge,
Yes, this happened to me last friday (March 9th), after the latest
update of my 14.04.5 which had 'proposed' enabled.
I have since reverted to the 204-5ubuntu20.25 version of the three
above-mentioned packages, by manual installation. Had also marked the
same packages on hold to prevent
Hi,
just to get this straight to narrow down scenarios to try to reproduce:
1. Dale with 16.04 you are *not* seeing this, right? You saw it with
14.04 with proposed enabled?
2. Marcelo, you are seeing this with 14.04.5 with proposed enabled?
--
You received this bug notification because you
Marcelo,
Thanks for the information. My system was unresponsive enough that
diagnosing the problem was a huge pain. I decided to back up /home and
do an install of 16.04 from scratch since I'd been intending to do an
upgrade soon anyway.
I was also using the "proposed" repo, so I should have
Solved the issue temporarily for me:
1. Had to manually download and install older packages:
sudo dpkg -i libpam-systemd_204-5ubuntu20.25_amd64.deb libsystemd-
daemon0_204-5ubuntu20.25_amd64.deb systemd-
services_204-5ubuntu20.25_amd64.deb
2. Put those versions on hold:
sudo apt-mark
Continuing my previous comment: after the update I got both 'cgmanager'
and 'systemd-logind' daemons eating 70% of processor time each (in a
multi-core cpu).
Tried to reinstall and reconfigure all recently updated packages, with
no result.
The visible effect is a long delay when logging (either
Same here.
I'm running a 14.04.5 distribution and the problem has appeared after
the last dist-upgrade.
I'm using the 'proposed' repository for Trusty, and a few packages have
been updated recently. I suspect it was caused by the following update:
I don't know whether we've had a regression here, but I started seeing
what looks like this exact same problem this morning.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
** Also affects: lxc (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to lxc in Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in cgmanager_ping_sync()
** Changed in: lxc (Ubuntu)
Status: New = Confirmed
** Changed in: lxc (Ubuntu)
Importance: Undecided = High
** Changed in: systemd (Ubuntu)
Status: New = Confirmed
** Changed in: systemd (Ubuntu)
Importance: Undecided = High
--
You received this bug notification because
Actually the source of the problem is that I was calling
nih_dbus_setup(), which actually sets up the mainloop. It is not needed
- as I thought it was - for making *_sync calls from cgmanager-
client.so. So the right fix is to not call that.
Two comments on this. First, nih_dbus_setup() seems
Right. So to be completely clear, by calling nih_dbus_setup(), nih was
registering a libdbus handler that runs on the client side such that
whenever libdbus deemed necessary, it would attempt to wake up the
clients main loop by calling nih_main_loop_interrupt(). However, since
the client didn't
** Branch linked: lp:ubuntu/trusty-proposed/systemd
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to lxc in Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in cgmanager_ping_sync()
To manage notifications
This bug was fixed in the package lxc - 1.0.3-0ubuntu2
---
lxc (1.0.3-0ubuntu2) trusty; urgency=medium
* Cherry-pick upstream fix for cgmanager integration. (LP: #1303649)
-- Stephane Graber stgra...@ubuntu.com Fri, 11 Apr 2014 12:17:41 -0400
** Changed in: lxc (Ubuntu)
** Changed in: libnih (Ubuntu)
Status: Confirmed = Invalid
** Changed in: cgmanager (Ubuntu)
Status: Confirmed = Invalid
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to lxc in Ubuntu.
This bug was fixed in the package systemd - 204-5ubuntu19
---
systemd (204-5ubuntu19) trusty; urgency=medium
* cgmanager: don't call nih_dbus_setup() since we won't use the
nih_mainloop(). (LP: #1303649)
-- Serge Hallyn serge.hal...@ubuntu.com Fri, 11 Apr 2014 10:17:04
The problem here is that applications linked to the cgmanager client
library are not calling nih_main_loop(). That's a problem because the
nih-dbus code calls nih_main_loop_interrupt() which adds a byte to the
main loop interrupt pipe. After 65536 bytes have been added to that
pipe, the next call
** Also affects: lxc (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in cgmanager_ping_sync()
To manage
** Changed in: lxc (Ubuntu)
Status: New = Confirmed
** Changed in: lxc (Ubuntu)
Importance: Undecided = High
** Changed in: systemd (Ubuntu)
Status: New = Confirmed
** Changed in: systemd (Ubuntu)
Importance: Undecided = High
--
You received this bug notification because
Actually the source of the problem is that I was calling
nih_dbus_setup(), which actually sets up the mainloop. It is not needed
- as I thought it was - for making *_sync calls from cgmanager-
client.so. So the right fix is to not call that.
Two comments on this. First, nih_dbus_setup() seems
Right. So to be completely clear, by calling nih_dbus_setup(), nih was
registering a libdbus handler that runs on the client side such that
whenever libdbus deemed necessary, it would attempt to wake up the
clients main loop by calling nih_main_loop_interrupt(). However, since
the client didn't
** Branch linked: lp:ubuntu/trusty-proposed/systemd
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in cgmanager_ping_sync()
To manage notifications about this
This bug was fixed in the package lxc - 1.0.3-0ubuntu2
---
lxc (1.0.3-0ubuntu2) trusty; urgency=medium
* Cherry-pick upstream fix for cgmanager integration. (LP: #1303649)
-- Stephane Graber stgra...@ubuntu.com Fri, 11 Apr 2014 12:17:41 -0400
** Changed in: lxc (Ubuntu)
** Changed in: libnih (Ubuntu)
Status: Confirmed = Invalid
** Changed in: cgmanager (Ubuntu)
Status: Confirmed = Invalid
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
This bug was fixed in the package systemd - 204-5ubuntu19
---
systemd (204-5ubuntu19) trusty; urgency=medium
* cgmanager: don't call nih_dbus_setup() since we won't use the
nih_mainloop(). (LP: #1303649)
-- Serge Hallyn serge.hal...@ubuntu.com Fri, 11 Apr 2014 10:17:04
The stacktrace suggests we are in libcgmanager, so adding a cgmanager
task.
** Also affects: cgmanager (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to cgmanager in Ubuntu.
It turns out the write() is not a part of the dbus transaction with
cgmanager, but actually a part of libnih's mainloop exiting code:
(gdb) where
#0 0x7f70ea509700 in __write_nocancel () at
../sysdeps/unix/syscall-template.S:81
#1 0x7f70eae47377 in nih_main_loop_interrupt
** Summary changed:
- systemd-logind high cpu consumption
+ systemd-logind spins in cgmanager_ping_sync()
** Changed in: systemd (Ubuntu)
Assignee: (unassigned) = Stéphane Graber (stgraber)
** Changed in: systemd (Ubuntu)
Status: Incomplete = New
--
You received this bug
Backtrace with debug symbols installed
(gdb) bt full
#0 0x7fad4e680700 in __write_nocancel () at
../sysdeps/unix/syscall-template.S:81
No locals.
#1 0x7fad4efbe377 in nih_main_loop_interrupt () from
/lib/x86_64-linux-gnu/libnih.so.1
No symbol table info available.
#2
The stacktrace suggests we are in libcgmanager, so adding a cgmanager
task.
** Also affects: cgmanager (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
It turns out the write() is not a part of the dbus transaction with
cgmanager, but actually a part of libnih's mainloop exiting code:
(gdb) where
#0 0x7f70ea509700 in __write_nocancel () at
../sysdeps/unix/syscall-template.S:81
#1 0x7f70eae47377 in nih_main_loop_interrupt
This patch seems to prevent logins from hanging and systemd-logind from
taking all cpu time.
It probably should not be blindly applied as there remains the question:
why is the interrupt buffer filling up.
** Patch added: libnih-eagain.diff
** Tags added: patch
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1303649
Title:
systemd-logind spins in cgmanager_ping_sync()
To manage notifications about this bug go to:
50 matches
Mail list logo