No performance data from command: check_pdh

[RobJames]

So i've just started a new job where i've inherited a Nagios Core setup, and am slowly working my way through many of the outstanding alerts. I've not used it before but am getting up to speed with how it works.

I've noticed that i have a load of alerts across multiple servers all reporting "NSClient - ERROR: No performance data from command: check_pdh".
The one common detail i've noticed is that they come from 2 Exchange 2010 servers. I've also noticed that the performance counters with the error do differ from server to server:

Server 1:
Microsoft Exchange Domain Controller LDAP Searches Timed Out Per Minute
Microsoft Exchange Domain Controller LDAP Search Time
Microsoft Exchange Domain Controller LDAP Read Time
Microsoft Exchange DAG Replication Queue
Microsoft Exchange Information Store Average RPC Latency
Microsoft Exchange DAG Replication Copy Queue Length
Microsoft Exchange Mailbox Submission Queue
Microsoft Exchange Resource Booking Failed
Microsoft Exchange Avg Document Index Time
Microsoft Exchange Hub Transport Servers in Retry
Microsoft Exchange Database Log Record Stalls per Second
Microsoft Exchange Information Store DB Log Write Latency
Microsoft Exchange Database Page Faults per Second
Microsoft Exchange Information Store DB Log Read Latency
Microsoft Exchange Mailbox Assistant CPU Time
Microsoft Exchange Database Log Threads Waiting

Server 2:
Microsoft Exchange Information Store DB Log Write Latency
Microsoft Exchange OAB Download Tasks Queue
Microsoft Exchange Information Store DB Log Read Latency
Microsoft Exchange Database Page Faults per Second
Microsoft Exchange Database Log Threads Waiting
Microsoft Exchange Database Log Record Stalls per Second
Microsoft Exchange Hub Transport Servers in Retry
Microsoft Exchange Mailbox Assistant CPU Time

I 've looked on the servers, and ran a perfmon to see if i can get any performance data from the server for any of the counters listed above and do get results. Most of them are 0 (which is a good thing) but some are returning values. For example DB Page Faults/sec gives me an average value around 241, and Log Threads Waiting averages about 0.3, so I am getting values.

I've run a lodctr /r to reload the performance counter data, but nothing changed.

I've restarted the nsclient service in case it an issue with it hooking into the performance data, again no change.

I've also reset the wmi performance counters with wmiadap /f and wmiadap /r to force the WMI classes to update, and to recreate the perfmon objects, but with no effect.

Suggestions?

Thanks

[kyang]

Can we see both of your cfg files for both Exchange servers? It will help if we see how those services are defined.

Code: Select all

/usr/local/nagios/etc/objects --> Server 1 and Server 2