The remote procedure call failed and did not execute. (Exception from HRESULT: 0x800706BF)

We monitor 11 Exchange 2007 servers but on one Transport server we frequently receive this error:

--------------------------------------------------

Application Status: Application status is Unknown

Component Status: Component status is Unknown

Component Status Details: Unexpected error occurred. The remote procedure call failed and did not execute. (Exception from HRESULT:0x800706BF)

Component Type: Windows Service Monitor

Monitored Service: MSExchangeTransport

---------------------------------------------------

We have rebooted the saerver and evn added more RAM but still the error remains. Any ideas how we can troubleshoot it?

netstatb.txt

Find more posts tagged with

Accepted answers

All comments

VáclavSáblík

Hello,
Please go ahead and open a support ticket for this. The problem seems more complicated and without diagnostics it's hard to say what is causing the issue.

Thanks,

FormerMember

Thanks. I have now opened a ticket.

MarieB

Hi Graham--

Could you post the case number and any results you get here.

Thanks,

chaxen

I found the following Microsoft article:

support.microsoft.com/default.aspx

Give that a try and update the forum

FormerMember

@ Chaxen - That article refers to Server 2003. The affected server is Server 2008. Thanks anyway.

I have since resolved the issue and here is a summary of what I found along the way:

After:

Running the Solarwinds diags and uploading them to the supports guys.
Increase the SWNetPerfMon.db Timeout Values.
Modify APM Monitors Polling Freq/Timeouts.
Much 'googling'....

I ended up opening a cmd prompt on the affected server and using netstat -b 5 to monitor the connections. I noticed every now and then it would return a massive amount of connections (over 1000) from the Orion server. I ran the same command on the other server that performs the exact same server role and I never had a single instance of multiple Orion connection attempts. I ran the output to a .txt file which I have attached so feel free to take a look and hopefully it may be of some use.

As a test, I went into ‘Manage Assigned Application Monitors’, unmanaged the problematic monitor (Exchange 2007 Edge Transport Role Services - <exchangeserver>) for a period of 1 hour. During that hour I disabled all but 1 Component. For the 1 remain enabled component I clicked ‘Test’ and used TCPView (Sysinternals) to monitor the active connections to <exchangeserver>. As soon as I clicked ‘Test’, the active connections would go through the roof and that was only for a single component! I then waited for the ‘WAITING’ connections to die and changed the Fetching Method from WMI to RPC and clicked ‘Test’ again and watched TCPView. I did not witness the same behaviour and the amount of connections stayed at a low level along with correctly updating the component monitor status.

So for some reason when the Fetching method was set to WMI it would exhaust the connections on the remote server and occasionally cause it to not respond. I have no idea why it sent so many connections but as soon as we changed it over to RPC and found that it worked we didn't spend too long worrying about it!

If anyone wants it the case ref for this is #225751

Cheers,

Graham