The default NTP servers have been configured for our three XenServer hosts:
I next used the ntpq utility to display a list of NTP peers known to the server:
ntpq -p
Using the command line ntpq -p on each of the XenServer hosts returned the following peers:
[root@xen01 ~]# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== *robert.elnounch 213.251.128.249 2 u 618 1024 377 10.361 -2.166 0.112 -probe.vaygr.net 194.29.130.252 2 u 993 1024 377 23.783 -5.053 0.163 +fra.pionica.com 212.244.36.228 2 u 993 1024 377 10.160 -1.079 0.470 +ntp1.monshouwer 193.79.237.14 2 u 91 1024 377 13.685 -0.979 0.964 LOCAL(0) .LOCL. 10 l 22 64 377 0.000 0.000 0.001 [root@xen02 ~]# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== *LOCAL(0) .LOCL. 10 l 15 64 377 0.000 0.000 0.001 [root@xen03 ~]# ntpq -p remote refid st t when poll reach delay offset jitter ============================================================================== *LOCAL(0) .LOCL. 10 l 59 64 377 0.000 0.000 0.001
Xen02 and Xen03 (Poolmaster) did not return any external peers. I restarted the NTP service using the following command line:
/etc/init.d/ntpd restart
Now when I ran ntpq -p on each of the XenServer hosts, the following peers are now listed.
Ellis Karim's Blog 
Only config ntp resolve u problem with HA-restart-host?
No, none of the changes we made to NTP or to the HA value has resolved the XenServer Pool master rebooting problem.
For the moment we have disbaled HA. This has stopped the server from fencing and we have not experienced any more problems since. The fix? On the 14the May 2012 we plan to perform a XenServer upgrade from Citrix XenServer 5.6 fp1 to XenServer 5.6 SP2 to see if this will fix the problem.