Hi,

I have an active / active 2 node cluster running windows
2k and SQL 2k. Both instances will succesfully come
online and run concurrently on the first node but when I
try to run both instances on the other node the SQL server
agent service fails for the second instance. I can take
the SQL server agent on the first instance offline and I
can then succesfully bring the service online for the
second instance. But then I can't start the service on
the first instance. As I've mentioned above everything
runs fine on the other node.

I've tried the suggestions in KB article 322140 and I've
confirmed the servcie accounts have full admin rights on
the domain and locally.

When I run sqlagent -i dev -c -v I get the following error
in the sql server agent error log,

[298] SQLServer Error: 17, SQL Server does not exist or
access denied. [SQLSTATE 08001]

Any help would be appreciated.

Cheers

I think you may be a bit confused ... or you've confused
me

Each SQL 2K virtual server has its own SQL Server agent.
The whole group fails over, not just the SQL Server agent,
so if you have it set so that the SQL agent resource
affects the group, the entire group (SQL IP, SQL Name, SQL
disks, etc.) will switch to the other server.

What do your error logs say (Windows)? What is the
definition of the SQL virtual servers (i.e. running the
fn_virtualservernodes)? Those will certainly provide some
more info.

Check your SQL Server Agent properties from Enterprise Manager. Check all
your security settings and see what is chosen for "local host server" on
the connection tab (should be set to default). Make sure that the SQL Agent
errorlog for each instance is pointing to its own log file (general tab).

If all of the above settings are correct, then check your registry keys to
make sure that the SQL Agent for each instance is pointing to the files in
its own directories.

Cindy Gross, MCDBA, MCSE
http://cindygross.tripod.com
This posting is provided "AS IS" with no warranties, and confers no rights.