Experiencing high load on some rsyslog instances. Status of threads showed the mainQ thread consumed 50-100% CPU. Change of queue.workerthreads to 2 enabled the second workerthread, but this does not consume any CPU. How are the workerthreads for main queue loaded?
Running on Debian 10 with rsyslog 8.1901.0-1. Top output: top - 08:12:50 up 6 days, 17:12, 2 users, load average: 1.25, 0.50, 0.46 Threads: 14 total, 0 running, 14 sleeping, 0 stopped, 0 zombie %Cpu(s): 14.0 us, 2.9 sy, 0.0 ni, 81.5 id, 0.2 wa, 0.0 hi, 1.4 si, 0.0 st MiB Mem : 7973.2 total, 119.5 free, 341.9 used, 7511.9 buff/cache MiB Swap: 952.0 total, 825.7 free, 126.2 used. 7273.8 avail Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND *30437 lognet 20 0 815648 28612 6320 S 66.1 0.4 0:46.40 rs:main Q:Reg*30438 lognet 20 0 815648 28612 6320 S 7.0 0.4 0:05.58 rs:ESP02 queue: 30439 lognet 20 0 815648 28612 6320 S 6.3 0.4 0:04.58 rs:SIEMEP1 queu 28705 lognet 20 0 815648 28612 6320 S 6.0 0.4 0:04.31 in:imtcp 30440 lognet 20 0 815648 28612 6320 S 1.0 0.4 0:00.40 rs:net-all.log 30461 lognet 20 0 815648 28612 6320 S 1.0 0.4 0:00.56 rs:ESP01-IPS qu ..... *30507 lognet 20 0 815648 28612 6320 S 0.0 0.4 0:00.83 rs:main Q:Reg* Peter _______________________________________________ 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.
