cupsd crashes regularly (daily)
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
cups (Ubuntu) |
Fix Released
|
High
|
Till Kamppeter | ||
Quantal |
Fix Released
|
High
|
Till Kamppeter | ||
cups-filters (Ubuntu) |
Fix Released
|
High
|
Till Kamppeter | ||
Quantal |
Invalid
|
Undecided
|
Unassigned |
Bug Description
Every day I get a 'System Problem Detected' dialog that says that cupsd has crashed. The symptom is similar to the one described in bug 1034045.
I expected this would be fixed by cups 1.6.1-0ubuntu11, but I have that version installed and the errors persist. The test case in bug 1034045 does not result in an immediate crash of cupsd. Specifically, this does not produce a crash on my system:
1. Run 'sudo rm -f /var/crash/
2. Run 'sudo service cupsd restart'
3. Verify that a new /var/crash/
[IMPACT]
For CUPS users who activate the "Show printers shared by other systems" in the server settings of system-
[TESTCASE]
Activate "Show printers shared by other systems" in the server settings of system-
cupsctl --remote-printers
I am not sure whether there must be actually a remote printer to trigger the crash. So on another computer in the same local network (or on a virtual machine if you do not have a second computer) create a CUPS queue and share it. Activate "Published shared printers connected to this system" in the server settings of system-
cupsctl --share-printers
on that computer. Now observe for some days. Probably you will get a crash report once a day.
With the proposed package the logrotate script is improved to not take a way log files from CUPS while it is running. It stops CUPS, moves the log files, and after that starts CUPS again. This way CUPS stops crashing during the logrotate once a day, so the system will behave normally now. Printing does not trigger the crash, so print jobs are always executed correctly.
You can test whether crash reports still work with the proposed package by triggering an artificial crash running the command
sudo killall -11 cupsd
Check whether the crash report process of Apport gets triggered, but DO NOT post the generated bug report on Launchpad. Remove /etc/apport/
[Regression Potential]
None. No change in any executable which could cause a regression.
ProblemType: Bug
DistroRelease: Ubuntu 12.10
Package: cups 1.6.1-0ubuntu11
ProcVersionSign
Uname: Linux 3.6.3-030603-
NonfreeKernelMo
ApportVersion: 2.6.1-0ubuntu6
Architecture: amd64
CupsErrorLog:
Date: Mon Dec 3 10:49:54 2012
InstallationDate: Installed on 2010-09-17 (808 days ago)
InstallationMedia: Ubuntu 10.10 "Maverick Meerkat" - Beta amd64 (20100901.1)
KernLog:
Dec 3 10:41:14 steve-laptop kernel: [71466.813974] type=1400 audit(135454927
Dec 3 10:41:14 steve-laptop kernel: [71466.814450] type=1400 audit(135454927
MachineType: Apple Inc. MacBookPro3,1
MarkForUpload: True
Papersize: letter
ProcEnviron:
TERM=xterm
PATH=(custom, no user)
XDG_RUNTIME_
LANG=en_US.UTF-8
SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=
SourcePackage: cups
UpgradeStatus: Upgraded to quantal on 2012-08-07 (118 days ago)
dmi.bios.date: 03/05/08
dmi.bios.vendor: Apple Inc.
dmi.bios.version: MBP31.88Z.
dmi.board.
dmi.board.name: Mac-F4238BC8
dmi.board.vendor: Apple Inc.
dmi.board.version: PVT
dmi.chassis.
dmi.chassis.type: 2
dmi.chassis.vendor: Apple Inc.
dmi.chassis.
dmi.modalias: dmi:bvnAppleInc
dmi.product.name: MacBookPro3,1
dmi.product.
dmi.sys.vendor: Apple Inc.
Changed in cups (Ubuntu): | |
importance: | Undecided → High |
Changed in cups (Ubuntu Quantal): | |
importance: | Undecided → High |
status: | New → Confirmed |
milestone: | none → quantal-updates |
Changed in cups (Ubuntu): | |
status: | Confirmed → Triaged |
tags: | added: patch |
Changed in cups-filters (Ubuntu): | |
status: | New → Triaged |
Changed in cups-filters (Ubuntu Quantal): | |
status: | New → Invalid |
Changed in cups-filters (Ubuntu): | |
importance: | Undecided → High |
assignee: | nobody → Till Kamppeter (till-kamppeter) |
milestone: | none → ubuntu-13.04-feature-freeze |
I reproduced the crash again today, this time with dbus debug symbols installed. The stack trace is attached. The stack is entirely avahi + dbus code, and the signature is very similar to other crashes reported in other applications. Looks like there might be a locking problem in the way cups uses avahi?
This thread describes a similar stack trace, and has some tips about using avahi in a threaded environment: marc.info/ ?l=freedesktop- avahi&m= 122021745832173
http://
Comment #12 of bug 524566 shows a very similar stack (without avahi), and the remainder of the bug has additional analysis.
Bug 1055060 has a crash at the same location within dbus, though the bottom of the stack is much different.