Bug 448640

Summary:	rsyslogd takes CPU all the to 99% to 100%
Product:	[Fedora] Fedora	Reporter:	sangu <sangu.fedora>
Component:	rsyslog	Assignee:	Peter Vrabec <pvrabec>
Status:	CLOSED RAWHIDE	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	high	Docs Contact:
Priority:	low
Version:	rawhide	CC:	corcodel.marian, d.bz-redhat, katzj, pierre-bugzilla, theinric
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2008-06-02 11:42:44 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description sangu 2008-05-28 00:24:45 UTC

Description of problem:
rsyslogd takes CPU all the to 99% to 100%

see : https://www.redhat.com/archives/fedora-test-list/2008-May/msg00861.html
Version-Release number of selected component (if applicable):
3.19.3-1.fc10

How reproducible:
always

Steps to Reproduce:
1. Update rsyslog-3.19.3-1.fc10
2. Reboot
3.
  
Actual results:


Expected results:


Additional info:

Comment 1 Pierre Ossman 2008-05-28 07:03:38 UTC

+1

It's the third thread (fourth created judging from the pids) that's the culprit.
No syscalls made by it. gdb backtrace:

#0  0xb8078ba2 in Syslog () from /usr/lib/rsyslog/imklog.so
#1  0xb8079376 in ?? () from /usr/lib/rsyslog/imklog.so
#2  0xb8079450 in klogLogKMsg () from /usr/lib/rsyslog/imklog.so
#3  0xb80789d4 in ?? () from /usr/lib/rsyslog/imklog.so
#4  0x0805954e in ?? ()
#5  0x435e451f in start_thread () from /lib/libpthread.so.0
#6  0x4351abde in clone () from /lib/libc.so.6

There is no activity in dmesg.

Comment 2 Peter Vrabec 2008-05-28 08:50:55 UTC

problem should be solved in rsyslog-3.19.4-1.fc10. Let me know, please. thnx.

Comment 3 Didier 2008-11-14 08:58:13 UTC

Seeing this again with latest rsyslog-3.21.3-4.fc10.i386 .

Comment 4 Tomas Heinrich 2008-11-18 15:29:14 UTC

This bug has been fixed and is not present in rsyslog-3.21.3-4, so what you're experiencing is probably unrelated to it.
I haven't seen it in a quick test run. Do you have a reliable way to reproduce it?
Have you done any changes to your configuration file?

Comment 5 Didier 2008-11-18 16:11:40 UTC

No changes to the configuration file.

To complicate the issue, I'm only seeing it on one of my 3 F-10 installations, about 50% of the time.

Would you fancy an strace log ?
If you rather prefer giving gdb instructions, I'll try to follow them too (rsyslog-debuginfo required, I guess ?).

Comment 6 Tomas Heinrich 2008-11-18 17:46:10 UTC

Any additional information would be useful.
strace log and if possible a backtrace from gdb - debuginfo is needed.

Does it always appear at any precise time after boot?

Comment 7 Didier 2008-11-18 20:18:17 UTC

I will have a look at that installation the next few days (it's a station which is not very often rebooted).

Concerning the gdb backtrace : I asked for instructions because I am not closely acquainted with gdb.  :)

Comment 8 Tomas Heinrich 2008-11-19 14:48:13 UTC

Sorry for being so brief about the instructions, here they are:
  # gdb rsyslogd <pid>
  gdb may complain about missing debuginfo,
  e.g. "Missing separate debuginfos, use: debuginfo-install gcc.i386 zlib.i386"

  (gdb) bt
  This will display the backtrace.