We began to experience RELP performance degradation of our rsyslog forwarders
over the weekend. No system parameters have been changed recently, although we
are always increasing the number of reporting clients (we may have hit some
threshold).
We have a Brocade SLB in front of the forwarders, but no changes there either.
There is no visible bottleneck in the forwarders (load of ~1 on 2 cores, swap
unused, no network drops), but strace shows a great deal of
22948 sendto(65, "1120 rsp 92 200 OK\nrelp_version="..., 105, 0, NULL, 0
<unfinished ...>
22963 <... sendto resumed> ) = 319
22961 <... futex resumed> ) = 0
22960 sendto(59, "16679316 syslog 204 <142>2015-04"..., 225, 0, NULL, 0
<unfinished ...>
22959 sendto(60, "16678730 syslog 188 <142>2015-04"..., 209, 0, NULL, 0
<unfinished ...>
22957 <... futex resumed> ) = 1
22949 <... syscall_299 resumed> ) = 0x5
22948 <... sendto resumed> ) = -1 EAGAIN (Resource temporarily
unavailable)
We are currently running 7.6.3 and are working on an upgrade but are hoping
there may be a system configuration change that could improve things until
that's accomplished. What other information might help troubleshot this?
Thanks,
Matt
_______________________________________________
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.
