Cluster Error 1726

[Dinkytoy]

Hi,

I've been having a bit of a nightmare with our primary SQL cluster recently.

Following some firmware upgrades to both servers one has decided that it wanted a divorce and it refused to start the cluster service generating an error each time, nothing useful simply 'Cluster service refused to start' or along those lines.

Having researched this I was able check the registry and cluster service files and all seemed to check out. There were very few temp files also which I removed to check and the clean up also had no effect.

Eventually a colleague and I decided to evict the node and re-add it back in. I tried this, this weekend past with no success.

Every time I now try and add the node back into the cluster I get the following errors in the cluster.log:

00000d18.00000e1c::2010/02/15-12:29:12.888 ERR [NM] Unable to synchronize node information, status 1726.
00000d18.00000e1c::2010/02/15-12:29:12.888 ERR [CS] ClusterInitialize failed 1726
00000d18.00000e1c::2010/02/15-12:29:12.888 ERR [CS] Service Stopped. exit code = 1726

I've checked connectivity and nic binding orders all seems to be ok. Has anyone any suggestions on where to take this next?

 

[itsp1965]

You didn't mention if you are using Win2K/2K3 or 2K8. Sounds like you're experiencing RPC errors. Are you sure you don't have the Windows firewall enabled or have removed any services that reside on RPC traffic?

 

[Dinkytoy]

Oops

It's 2003 ent sp2 both ends .

Firewall is deffinitely off and RPC service checks out on both. I've run through the service list on both nodes and they are the same with the exception of SQL Server and Cluster Service, them only being active on the remaining live server.

 

[Dinkytoy]

We rebuilt the box in the end and got it back into the cluster without a problem.