Posted 18 June 2012 by Kyle Gleed, Sr. Technical Marketing Architect, VMware
A common complaint I get when troubleshooting host profiles is that the error messages that get generated aren’t very helpful. For example, consider the following error I recently received when trying to apply a host profile to a host named “esx-03a”:
There’s really not much to go on with this error. However, I’ve learned that when troubleshooting host profiles and coming across one of these vague error messages I can usually figure out the cause of the error by looking in the host’s /var/log/syslog.log file. In my case, with the error noted above, I went to the syslog.log file and found the following:
The log shows that the the error was occurring when the host profile was attempting to set the NTP service to “on”. I was able to verify that this was my problem by quickly editing the host profile and setting the service to “off”:
After saving the changes and re-applying the Host Profile to host esx-03a I was able to verify that this did eliminate the error.
Of course the next challenge was to figure out why setting the NTP service to “on” was causing an error? To investigate I tried manually starting the NTP service on the host:
From here the problem became evident – I was trying to start the NTP service without having defined a valid NTP server. Whoops ;) Once I added an NTP server I was able to set the NTP service back to “on” in my host profile and everything worked as expected.
Yes, we do need some better error handling with host profiles, but until then I hope this post will help you in your troubleshooting efforts.
Follow me on twitter @VMwareESXi