Your message dated Thu, 12 Dec 2013 03:12:28 +0000
with message-id <1386817948.12186.100.ca...@deadeye.wl.decadent.org.uk>
and subject line Re: userspace apps using epoll_wait have their memory corrupted
has caused the Debian Bug report #728026,
regarding userspace apps using epoll_wait have their memory corrupted
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
728026: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=728026
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: systemd
Version: 204-3
Severity: grave
Justification: no more able to shutdown system except by turning off power
Dear Systemd Maintainers,
first sorry for the late bug report. The following happened the first
time on 13th of September 2013 (i.e. shortly after the 204-3 upload
while being booted with 204-2 initially IIRC, hence reporting against
204-3), but occurred only once at that time. In the meanwhile it though
happened at least two more times with 204-5 (last time on 2nd of October
-- I've stopped playing with systemd at that point), so despite I don't
know how to reproduce it deterministicly, it's definitely reproducable
for me after a while. Michael Stapelberg told me you haven't heard about
such an issue so far, so I guess this bug report is still interesting
for you despite being late.
If I start my EeePC with init=/bin/systemd via grub2's commandline
editor, after a few days (i.e. at least one suspend to RAM since boot)
no command which needs to talk to /run/systemd/private works anymore,
i.e. no service starting or stopping and especially no more shutting
down the system properly:
The following was from the first occurrence with 204-3 (commands and
prompt typed manually because I was in a hurry and was just able to
document the most important stuff by adding "> note.txt 2>&1" to the
commands and then turned off the machine hard by pressing the power
button for 5 seconds:
~ # service miredo stop
Failed to get D-Bus connection: Failed to connect to socket
/run/systemd/private: Connection refused
~ # ps auxww | fgrep systemd
root 1 0.0 0.1 6184 3704 ? Ss 12:00 0:05
/lib/systemd/systemd --system --deserialize 19
message+ 1133 0.1 0.0 4060 2036 ? Ss 12:00 0:53
/usr/bin/dbus-daemon --system --address=systemd: --nofork --nopidfile
--systemd-activation
root 5659 0.0 0.0 3932 1684 ? Ss 14:41 0:00
/lib/systemd/systemd-logind
root 5661 0.0 0.3 26656 6936 ? Ss 14:41 0:06
/lib/systemd/systemd-journald
root 18447 0.0 0.0 11864 1820 ? Ss 14:30 0:00
/lib/systemd/systemd-udevd
root 25398 0.0 0.0 3908 640 pts/0 S+ 23:50 0:00 fgrep
--color=auto systemd
~ # shutdown -h now
shutdown: timeout opening/writing control channel /run/initctl
~ # telinit 0
init: timeout opening/writing control channel /run/initctl
~ #
The last time it happened (2nd of October 2013) I had more time to
gather some information. Running systemd-related processes:
~ # ps auxww | fgrep systemd
root 1 0.0 0.1 6276 2992 ? Ss Sep28 0:12 /bin/systemd
root 260 0.0 1.0 183424 21440 ? Ss Sep28 0:23
/lib/systemd/systemd-journald
root 269 0.0 0.0 11804 1936 ? Ss Sep28 0:01
/lib/systemd/systemd-udevd
root 1009 0.0 0.0 3932 1600 ? Ss Sep28 0:00
/lib/systemd/systemd-logind
message+ 1014 0.1 0.0 3948 1800 ? Ss Sep28 9:21
/usr/bin/dbus-daemon --system --address=systemd: --nofork --nopidfile
--systemd-activation
root 29423 0.0 0.0 0 0 ? Zs 00:38 0:00
[systemd-sleep] <defunct>
root 32008 0.0 0.0 3908 648 pts/8 S+ 19:31 0:00 fgrep
--color=auto system
~ #
I can provide straces for the running dbus-daemon process (pid 1014) by
private e-mail for the following no more working actions:
* service miredo stop
* laptop lid close
* shutdown -h now
If you have more specific ideas what command or which process to strace
would help you in debugging this issue, I'll give init=/bin/systemd
another try at next required (re-)boot.
(This report has been written while normal sysvinit was used.)
-- Package-specific info:
--------------
systemd-delta:
--------------
0 overridden configuration files found.
--------------
Contents of /var/lib/systemd/deb-systemd-helper-enabled:
--------------
==> /var/lib/systemd/deb-systemd-helper-enabled/miredo.service.dsh-also <==
/etc/systemd/system/multi-user.target.wants/miredo.service
==>
/var/lib/systemd/deb-systemd-helper-enabled/multi-user.target.wants/miredo.service
<==
==>
/var/lib/systemd/deb-systemd-helper-enabled/accounts-daemon.service.dsh-also <==
/etc/systemd/system/graphical.target.wants/accounts-daemon.service
==>
/var/lib/systemd/deb-systemd-helper-enabled/graphical.target.wants/accounts-daemon.service
<==
-- System Information:
Debian Release: jessie/sid
APT prefers unstable
APT policy: (990, 'unstable'), (900, 'testing'), (899,
'testing-proposed-updates'), (600, 'stable'), (500, 'proposed-updates'), (200,
'experimental')
Architecture: i386 (i686)
Kernel: Linux 3.11-1-686-pae (SMP w/2 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages systemd depends on:
ii initscripts 2.88dsf-43
ii libacl1 2.2.52-1
ii libaudit1 1:2.3.2-2
ii libc6 2.17-93
ii libcap2 1:2.22-1.2
ii libcryptsetup4 2:1.6.1-1
ii libdbus-1-3 1.7.6-2
ii libgcrypt11 1.5.3-2
ii libkmod2 9-3
ii liblzma5 5.1.1alpha+20120614-2
ii libpam0g 1.1.3-10
ii libselinux1 2.1.13-3
ii libsystemd-daemon0 204-5
ii libsystemd-journal0 204-5
ii libsystemd-login0 204-5
ii libudev1 204-5
ii libwrap0 7.6.q-24
ii udev 204-5
ii util-linux 2.20.1-5.5
Versions of packages systemd recommends:
ii libpam-systemd 204-5
Versions of packages systemd suggests:
ii systemd-ui 2-2
-- no debconf information
--- End Message ---
--- Begin Message ---
Version: 3.11.8-1
On Sat, 2013-11-02 at 17:31 +0100, Michael Stapelberg wrote:
> Axel Beckert <a...@debian.org> writes:
> > Kernel: Linux 3.11-1-686-pae (SMP w/2 CPU cores)
> The following kernel change causes the issue you are describing:
> http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=1c441e9
>
> Apparently, only 32-bit machines running Linux ≥ 3.10 are
> affected. Rolling back to an older kernel version or manually reverting
> the commit I pointed to above fixes the issue.
>
> I am reassigning this bug to src:linux in the hope that the maintainers
> can apply a patch rolling back the commit in question. As described in
> https://bugzilla.kernel.org/show_bug.cgi?id=61781, upstream has reverted
> it for 3.12, too.
This was also reverted in 3.11.8.
Ben.
--
Ben Hutchings
Knowledge is power. France is bacon.
signature.asc
Description: This is a digitally signed message part
--- End Message ---