Given Valgrind trace, the leak has nothing to do with mmelasticsearch.
This is some evaluation in RainerScript. Could you update to 7.4.5 to
make sure the leak persists?
--
Pavel Levshin
25.10.2013 18:06, Grégoire Seux:
thanks for your answer, I have attached the configuration on the ticket
-----Original Message-----
From: [email protected] [mailto:rsyslog-
[email protected]] On Behalf Of Radu Gheorghe
Sent: vendredi 25 octobre 2013 15:53
To: rsyslog-users
Subject: Re: [rsyslog] omelasticsearch plausible memory leak
Hi Gregoire,
Please tell more, because I'm using exactly this setup :)
It's been running on a machine for 8 days now and it must have been
indexing a few million messages (don't hit me for not knowing the exact
number, I don't use impstats on it yet).
Still, htop says it takes 7.3MB reserved and 703MB virtual. But I'm not
worried about the virtual one, because for all running apps it adds up to
more than 4GB, and I've only got some 800M actually used (free -m) and no
swap.
Can you share your conf? And how you determine the memory usage?
Best regards,
Radu
2013/10/25 Gregoire Seux <[email protected]>
Hello,
I have noticed an always increasing memory use when using
omelasticsearch
on rsyslog 7.4.4.
Even once the action queues are full, the ram increases goes on,
reaching 30GB in about 10minutes (21M messages enqueued).
A valgrind launched with --leak-check=full shows that:
==56893== 137,545,752 bytes in 898,534 blocks are definitely lost in
loss record 141 of 141
==56893== at 0x4C279EE: malloc (vg_replace_malloc.c:270)
==56893== by 0x5AFE3B9: es_newStr (in /usr/lib64/libestr.so.0.0.0)
==56893== by 0x5AFE6FE: es_newStrFromCStr (in
/usr/lib64/libestr.so.0.0.0)
==56893== by 0x13EA7D: msgGetMsgVarNew (in /sbin/rsyslogd)
==56893== by 0x134E40: cnfGetVar (in /sbin/rsyslogd)
==56893== by 0x12FBD3: cnfexprEval (in /sbin/rsyslogd)
==56893== by 0x12EE97: ??? (in /sbin/rsyslogd)
==56893== by 0x12FC03: cnfexprEval (in /sbin/rsyslogd)
==56893== by 0x12FB58: cnfexprEval (in /sbin/rsyslogd)
==56893== by 0x12FC83: cnfexprEval (in /sbin/rsyslogd)
==56893== by 0x13085F: cnfexprEvalBool (in /sbin/rsyslogd)
==56893== by 0x15156F: ??? (in /sbin/rsyslogd
Does any one experienced that before ?
I have filed a ticket (http://bugzilla.adiscon.com/show_bug.cgi?id=486)
to help adding information about the issue.
--
Grégoire
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad
of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you
DON'T LIKE THAT.
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you
DON'T LIKE THAT.
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE
THAT.
_______________________________________________
rsyslog mailing list
http://lists.adiscon.net/mailman/listinfo/rsyslog
http://www.rsyslog.com/professional-services/
What's up with rsyslog? Follow https://twitter.com/rgerhards
NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of
sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE
THAT.
