Saturday, June 30, 2012

Linux Kernel and Leap Seconds

We had several systems at $WORK tonight become somewhat unresponsive.  CPU usage was pegged, and  interactive response times were typically abysmal, sometimes waiting several seconds and/or minutes for keystrokes to be acknowledged.

In addition, some systems were responding well enough, but had very high context switch rates.  The lowest rate I saw was about 500,000 context switches a second, but the highest I saw was over 2.8 million switches a second!

Almost all of the poorly-performing systems were virtual servers, and the performing-okay-but-high-context-switch-rate systems were all physical servers.  I hypothesized that the problem was the same one, but the physicals had more CPU power available to them -- most of our VMs don't have more than 2-3 vCPUs, whereas the physicals have upwards of 16 in some cases, if you count hyperthreads.

As it turns out, it was being caused by some weird (sorry, I'll try to keep from adding any more of this technical jargon...) kernel interaction when it processed the leap second that occurred today.  For more details, I turn you now to the blog entry that helped me narrow the problem down and provided me with a simple fix:

http://blog.mozilla.org/it/2012/06/30/mysql-and-the-leap-second-high-cpu-and-the-fix/

For the record, should this link ever stop working, he said:

The fix is quite simple – simply set the date. Alternatively, you can restart the machine, which also works. Restarting MySQL (or Java, or whatever) does NOT fix the problem. We put the following into puppet to run on all our machines:
$ cat files/bin/leap-second.sh 
#!/bin/bash
# this is a quick-fix to the 6/30/12 leap second bug

if [ ! -f /tmp/leapsecond_2012_06_30 ]
then
/etc/init.d/ntpd stop; date -s "`date`" && /bin/touch /tmp/leapsecond_2012_06_30
fi

His solution was a lot more elegant than mine, which was to simply reboot the system. :)  It was also a lot easier to apply prophylactically to our entire fleet.

3 comments:

  1. Thanks for this writeup; it helped me understand the chaos in our lab this weekend. Yikes!

    ReplyDelete
  2. What about starting ntpd again? We found it was enough to set the date, no need to kick over the daemon.

    ReplyDelete
    Replies
    1. Ah yes, it looks like the source I cited didn't include a restart after running the date command, but we did issue a "/sbin/service ntpd start" after running date.

      Delete