Here’s my problem.
Node A & B are synchronizing to the wrong host.
Node C is synchronizing to the right host.
This is the behavior I see on the CVMs.
When I do “hostssh ntpq -pn”, the hypervisors (AHV) are reporting the correct ntp server.
How do I bring nodes A & B back inline to what they should be?
I tried manually correcting their ntp.conf files with the correct IP and restarted the ntpd service. No change and eventually the ntp.conf files changed to the wrong settings. Not sure how to wrestle this one to the ground.
@DannyS , can you have a look at this article http://portal.nutanix.com/kb/4519 and try to see if you can try to fix the issue.
Let me know if you need anything else.
Thanks. I figured it out that what I was seeing was right. One node was the leader and the other two followed it.
@DannyS , that sounds amazing. Exactly that’s how it works on a Nutanix cluster. We have an NTP leader who further syncs time with the External NTP server and the rest of the CVMs in the cluster follow that NTP leader.
Anyways you can keep the above article to understand more about how NTP works on Nutanix clusters and fix any issues if they occur in future.
@DannyS how are you doing today. I hope I was able to clear your doubts. Let me know if you need help with anything else or need any sort of clarifications.
Here are a few more NTP related topics and blogs which you can keep for future reference and go through for a better understaiding.