Lucky me, I am setting up redgate SQL Monitor to monitor SQL Server Instances. First to say – as ever – redgate is developing ingeniously simple tools. So the installation just took minutes, the services on one VM, the repository database on another.
After having this successfully finished i logged into the Web GUI and configured my first target Host where the SQL Server Instance is running that i want to monitor.
As expected everything went fine and the windows metrics appeared like “Disk avg. write time”.
But i was not able to have a look at the SQL Server metrics. So something must have gone wrong. I had a look at Configuration -> Monitoring: “Monitored servers” and found an unconnected machine.
But wait – the first few seconds there was a “Monitoring connected” message, so something must have gone wrong a couple of seconds later on. In column “Actions” i chose the link “Show log”.
Of course i googled “The RPC server is unavailable” along with “0x800706BA”. As I am currently dealing with a Windows Server Failover Cluster Instance I tried the workaround mentioned in Warning Event ID 5605 is Logged in Application log when querying MSCluster namespace through WMI and edited the ClusWMI.mof adding the value FALSE to [RequiresEncryption(FALSE)]. But that did not help.
I also followed the advice from redgate support to check all methods that SQL Monitor uses to connect to the target Host and also the SQL Server Instance.
I also checked the firewall settings allowing tcp communication on port 135 and > 1024. Everything open.
When it came to WMI test using WbemTest I tried to connect to my target Host, but also received the well known error:
So I consulted a colleague of mine to double check if my target Host was available from another VM.
Guess what: It was. So it must have had something to do with the configuration of the two VMs we were trying to connect to the target Host.
She mentioned that she had configured the DNS suffixes in the local search list.
That did the trick. With all my unsuccessful attempts to connect to my target Host I always defined it fqdn in SQL Monitor. But i guess that the WMI security settings on my target Host prevented me from getting the right response.
As the target Host is not in the same domain that my monitoring VM is in, i had to put the IP Address of the very target Host into the hosts file on the monitoring VM in location “C:\Windows\System32\drivers\etc\hosts”
Please note that I just put in the NetBIOS Name of the Server.
That did the trick.
Going back to SQL Monitor, deleting my unsuccessfully configured taret Host and configuring it once again, but this time not fqdn but only with the SERVERNAME it all worked well and finally i got everything up and running.
Happy monitoring to all of you 🙂
Thank you for reading,