Home > Homelab, Random VMware ramblings > vCenter / PSC appliance fails to configure NTP

vCenter / PSC appliance fails to configure NTP

September 7th, 2016 Leave a comment Go to comments

I have been rebuilding my lab this week and ran into some issues when trying to configure NTP on the PSC and VCSA that were freshly deployed on 6.0U2.

The error I got can be seen below, which was not extremely helpful unfortunately.

error

Google does not have many results for that error message, neither does the VMware Knowledge Base. Since I don’t want to rely on the internal clock in the NUCs and sync via VMware tools the next step is to find out what is causing the error message.

The configuration of the NTP servers is done using the VAMI interface which logs into the following file:

/var/log/vmware/applmgmt/vami.log

2016-09-06T17:18:53.250 [27124]INFO:root:id=ntpUpdate, action=submit
2016-09-06T17:18:53.250 [27124]INFO:root:Mode = NTP
2016-09-06T17:18:53.250 [27132]ERROR:vmware.appliance.timesync.utils:IP validation failed 0.europe.pool.ntp.org {‘0.europe.pool.ntp.org’ does not appear to be an IPv4 or IPv6 address}
2016-09-06T17:18:53.250 [27132]WARNING:vmware.appliance.timesync.utils:Hostname label (0) starts with a digit.
2016-09-06T17:18:53.250 [27132]ERROR:vmware.appliance.timesync.utils:IP validation failed 2.europe.pool.ntp.org {‘2.europe.pool.ntp.org’ does not appear to be an IPv4 or IPv6 address}
2016-09-06T17:18:53.250 [27132]WARNING:vmware.appliance.timesync.utils:Hostname label (2) starts with a digit.
2016-09-06T17:18:53.250 [27132]ERROR:vmware.appliance.timesync.utils:IP validation failed 2.ie.pool.ntp.org {‘2.ie.pool.ntp.org’ does not appear to be an IPv4 or IPv6 address}
2016-09-06T17:18:53.250 [27132]WARNING:vmware.appliance.timesync.utils:Hostname label (2) starts with a digit.
2016-09-06T17:19:13.250 [27124]ERROR:root:Exception {Pint API call timed out.}
Traceback (most recent call last):
File “/usr/lib/applmgmt/base/py/vmware/vherd/base/detwist.py”, line 245, in invokePintMethod
result = methodProcess.getResult()
File “/usr/lib/applmgmt/base/py/vmware/vherd/base/detwist.py”, line 185, in getResult
raise Exception(“Pint API call timed out.”)
Exception: Pint API call timed out.
2016-09-06T17:19:13.250 [27124]ERROR:root:Exception:
Traceback (most recent call last):
File “/opt/vmware/share/htdocs/vami/backend/vami.py”, line 51, in execute
method(response, locale, action, input)
File “/opt/vmware/share/htdocs/vami/backend/manage-time.py”, line 121, in ntpUpdate
raise Exception(“%s” % GetErrorMsg(data))
Exception: Internal Error

2016-09-06T17:19:13.250 [27124]ERROR:root:Error processing request: Internal Error
2016-09-06T17:19:13.250 [27124]INFO:root:Response :
<?xml version=”1.0″?>
<response><locale>en-US</locale><requestid>ntpUpdate</requestid><status><statusCode>failure</statusCode><statusMessage>Internal Error</statusMessage></status></response>

 

So we are failing for a timeout and getting messages that something does not seem to be a valid ip address. Well it is true that I have not entered IP addresses but rather host names. A quick check determines then that on both appliances DNS indeed was not working correctly. I could resolve internal domain names just fine but external names were no joy.

Since I manage the DNS server for my own lab as well I took a look there next, and sure I had forgot to put in DNS forwarders. Easy fix I thought until I noticed that the normal quick and dirty forwarders which I tend to use (8.8.8.8 and 8.8.4.4 aka Google DNS servers) seemed to have some issues. The root hints were also not working at that time.

forwarders

It took me a while to figure out what was happening as I was to lazy to utilize Wireshark at that point, but the Google DNS servers were able to be pinged fine and for testing purposes I disabled the internal Windows firewall without any difference.

As you can see in the above picture though 3 servers are now added that are working and validated fine. It seems that a change was made in the network where I do not have control over to block DNS traffic from certain server apparently. As soon as I added some network internal DNS servers everything worked fine which includes the successful setup of NTP through the VAMI interface on both appliances.

ntp

Categories: Homelab, Random VMware ramblings Tags:
  1. No comments yet.
  1. No trackbacks yet.