4

My system is a Debian 6.0.3 x86_64 box w/kernel 2.6.32-5-openvz-amd64, it hosts a number of OpenVZ containers. I have recently migrated from LXC/newer kernel from backports repo to the stable OpenVZ kernel to resolve network issues but got another issue - the CPU spends a lot of time processing software interrupts. Here is a Munin graph for CPU time: http://prntscr.com/arjzl.

I added nohz=off and highres=off to the kernel command line but that did not help much. And, the number of interrupts is quite low on this box according to another Munin graph: http://prntscr.com/ark19, so the interrupts seem to be not the cause of this issue. When I run top, zabbix_server is the most CPU consumer but I don't see any relation between zabbix_server and software interrupts.

I'm not sure how to proceed with investigation, this looks like a kernel bug or a kernel module bug to me, but I don't know hot to track this down to the guilty process/module. Maybe there is another kernel command line parameter to tune. What should I try next?

Alex
  • 7,789
  • 4
  • 36
  • 51

1 Answers1

3

zabbix_server is the most CPU consumer but I don't see any relation between zabbix_server and software interrupts.

I wonder how exactly do you try seeing a relation? Have you tried stopping it, for example?

Since this problem has arisen after kernel change, it looks reasonably to assume this is a bug of OVZ 2.6.32-5. I suggest using http://wiki.openvz.org/Download/kernel/rhel6/042stab055.16 if it's fresher than yours current (as it seems).

poige
  • 9,171
  • 2
  • 24
  • 50
  • Hmm good point, I just tried to stop it and si time was back to normal as soon as I did! So they were related closely. But I just can't get the reason of this relation, `zabbix_server` is a userspace thing so it should be a bug definitely. I will try to get a newer OpenVZ implementation but it looks like I will have to compile it myself. – Alex Jun 18 '12 at 17:47
  • 1
    @Alex, You can extract pre-built RPM, they're shipping, to corresponding location and then just use `mkinitcpio` appropriately. That would save hell-a-lot of work. Also, since you've found zabbix was a bugger, you can as well `strace` it to see if there's something weird going on. – poige Jun 18 '12 at 17:53
  • 1
    @Alex, «zabbix_server is a userspace thing so it should be a bug definitely» — http://www.linfo.org/software_interrupt.html : «A software interrupt, also called an exception, is an interrupt that is caused by software, usually by a program in user mode.» – poige Jun 19 '12 at 02:04
  • RedHat kernel solved my problem! – Alex Jun 19 '12 at 06:47
  • @Alex, glad to hear that. Did u use the RPMs or build the kernel by yourself? – poige Jun 19 '12 at 06:50
  • I used this build script: https://github.com/CoolCold/tools/blob/master/openvz/kernel/create-ovz-kernel-for-debian.sh to build .deb packets – Alex Jun 19 '12 at 07:03