On Thu, Aug 29, 2013 at 5:32 PM, Steve Thompson smt@vgersoft.com wrote:
CentOS 6.4, x86_64.
ntpd on one of my systems has started consuming 66% of one core, although it appears to be functioning correctly otherwise. No pertinent logs. Of
Did you take a peek at the traffic going to this server?
Since you haven't specified, I figure this is just an ntp client and not an ntp server. Is that right?
Depending on your configuration (network ACLs, iptables on the host, and ntpd.conf config on the host) maybe your server is being used in a denial of service attack.
course, nothing was changed :) I've seen this before many times, but usually the CPU consumption falls back to normal within a day or so, but this has been going on for several weeks now. Stopping and restarting ntpd makes no difference. Reinstalling ntpd makes no difference. The ntpd.conf file is the same as on 200+ other systems, all of which work normally. Anyone seen this before?
You might consider running strace on the ntp process when it decides to ramp and consume CPU cycles.
Steve _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos