We have a SQL 2000 cluster that is running on two Windows 2003 Enterprise
boxes and everything works fine, except that after the cluster has been
running for a while (24 hrs.), if you fail the cluster over to the failover
node, then when you go to fail it back to the primary node SQL spits an SPN
error, like it needs to re-set the SPN and refused to let the SQL instance
and the SQL agent instance and the SQL fulltext instance all hang. The last
two of course because they are dependent on the SQL instance itself. All
other resources to include the cluster and MSDTC resources fail back to the
primary server just fine. It seems to be just the SQL server instance that
is causing the issues.

After doing hours of research and finding nothing on this, I re-booted the
primary server, and then the SQL server instance as well as all other
resources failed back to the primary server just fine. Does anyone have an
answer as to why this happens? It's not much of a failover cluster if I
can't fail back to the primary server without re-booting it first.

What was the error you were getting?

Rand
This posting is provided "as is" with no warranties and confers no rights.