This content has been marked as final. Show 9 replies
Dedicated or shared server connection?
Using SID, INSTANCE_NAME or SERVICE_NAME?
Is load balancing used?
SCAN listener being connected to?
Best post the entire TNS connection string used.
Then there are network issues that can also be contributing or causing the performance problem.
Firewall(s) between client and server?
How many hops, and what is the latency per hop (use tracert for overviw and ping with TTL set for analysis)?
Other potential issues includes hostname resolution. Look at the DNS config of the client. How fast does it resolve a hostname to IP?
Is reverse DNS lookup used at server? How fast does it get the hostname for an IP?
Thanks for your Help!
As I mentioned the environment is cluster with OFS.
Two DB's A and B running on node 1 and 2.
A---->running on Server 1
B----->running on Server 2
When A is moved to Node 2 and B is moved to node 1. The connection works perfect. I can connect in a second.
But When i move them back on their respective servers I have issue with DatabaseA on Server1.
DatabaseB on Server2 works normal without any issues.
Very Strange Behaviour!!!
I've seen something similar before, but the sessions wouldn't connect at all. Check that your diagnostic dest for ANY DB running on that same server is not pointing to a non existent location. Also, is your listener process running as the Windows local system account? The database service account and listener should be running as the same account.
I would also check through all of the db parameters and makes sure that nothing is pointing to a location that no longer exists, or get's removed temporarily. This is what was causing my issue, but as I say I couldn't connect at all.
If not, can you try tracing the connection to see where the time is.