Camaleón <noela...@gmail.com> writes: > On Sat, 08 Sep 2012 13:31:16 +0200, Mauro wrote: > >> Hello I've two server with debian squeeze and in cluster with >> heartbeat+pacemaker. >> They run ntpd for time synchronize. >> I've noticed some ntpd crashes in random days and random hour. > > Does restarting the service works? > >> Logs don't say why. >> Can you suggest what can I do to know the reasons for the crash. Thank >> you. > > Mmmm... there's a bug report¹ to request ntpd debugging flag is turned > on (which I think is a must) but in the meantime I guess you will have > to recompile the package from Debian sources and toggle this parameter > on if you want to get an insightful trace. > > ¹http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=643954
I too get random crashes with ntpd in squeeze (1:4.2.6.p2+dfsg-1+b1) on a Xen dom0 host. Frequency is about every two months. It prompted me to install monit² which has nice support for probing NTP protocol and restarting the service automatically - not a perfect solution since the clock jumps forward an hour during the breakage, upsetting lots of software (dovecot and RRD graphs mostly). There is a bug report³ that defines our situation exactly: ntpd exits randomly with no messages in syslog. Thanks Camaleón for the idea to rebuild with debug - I'll try that next time around :) ² http://packages.debian.org/squeeze/monit ³ http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=594071 -- Gerald Turner Email: gtur...@unzane.com JID: gtur...@unzane.com GPG: 0xFA8CD6D5 21D9 B2E8 7FE7 F19E 5F7D 4D0C 3FA0 810F FA8C D6D5
pgp0dBgIhC9eN.pgp
Description: PGP signature
