Just some more info on this
I noticed there is a single haproxy process (nbproc 8) that uses 100% CPU when hitting '1' on 'top' command we see that every minute or two, different core becomes utilized (75%) So according to the CPU maps it jumps between Cores 1,3,5,7 strace: Process 82882 detached % time seconds usecs/call calls errors syscall ------ ----------- ----------- --------- --------- ---------------- 100.00 0.005369 0 1110052 epoll_wait ------ ----------- ----------- --------- --------- ---------------- 100.00 0.005369 1110052 total Thanks ________________________________ From: Haim Ari Sent: Thursday, May 25, 2017 10:23:18 AM To: [email protected] Subject: Haproxy first core 100% Hello, I'll try do describe the issue as clear as possible: We set up an haproxy Cluster on Ubuntu16.04 + pacemaker + corosync We faced an issue where after working for a few hours with single core (the haproxy process was bind to core #1 not #0 - which is handling the IRQ interrupts ) it is 100% utilized. When restarting the haproxy process it drops down to 40% (with same traffic) and after a few hours it gets to 100% again. Then we changed to nbproc 8 and assigned the following cpu mapping: nbproc 8 cpu-map odd 1 3 5 7 cpu-map even 2 4 6 8 I spread the fronted so that HTTP goes to 'odd' and HTTPS goes to 'even' However we are facing the same issue, only that now we see core#5 on 100% Attached sysctl.conf + haproxy.conf Haproxy: HA-Proxy version 1.7.3 2017/02/28 OS: Ubuntu 16.04.2 LTS Kernel: 4.4.0-66-generic Pacemaker: 1.1.14-2 Corosync 2.3.5-3 Any advise on this ? Thank you
