check_vmware_api.pl inconsistencies

[svrprice]

To start, all referenced vcenter instances are running V7.0.2 Build:17958471 and hosts are all running v7.0.2 Build:17867351. Both XI instances are running 5.8.6.

Running into a couple of different issues on services being monitored under the check_vmware_api.pl plugin. We have been monitoring 4 clusters, 3 vcenter and 27 nodes, for over 4 years with no issues, until about 3 weeks ago, where 5 hosts and one of the vcenter servers all of a sudden started throwing 'CHECK_VMWARE_API.PL CRITICAL - Can't call method "cpuStatusInfo" on an undefined value at /usr/local/nagios/libexec/check_vmware_api.pl line 2387.' on the 'VM Status for VMHost' monitor. All other hosts and vcenter are monitored successfully. No changes or updates were made in the environment for at least 3 weeks prior to initially seeing the issue.

Additionally we are currently in the process of deploying a second XI instance, to split between 2 data centers, which are currently both monitored by a single XI instance, and we are receiving timeout errors on half of the vmware hosts. The hosts which are timing out on the new instance, are all successfully being monitored on the current instance using the same plugin and version. Furthermore the timed out hosts and XI instance are on the same segment with no FW between. We have tried using different versions of the Perl plugin, from 6.5 through 7.0, but see no change.

[ssax]

Please note that you would need a 2nd XI license for the other XI server to be within the license entitlements if it is licensed with the same key:

https://support.nagios.com/kb/article.php?id=145

Please PM me the full output of one of the failing commands from the command line but add this onto the command as well so I can see what it is showing:

Code: Select all

--trace=4 -vvv