In startup, init.ora, you scecifiy how many process Oracle may keep running at anyone time. In your case, when you try and log in, Oracle senses you trying to exceed this and returns the error you are seeing.
You are going to have to manually kill at least one process, before you can log in.
The big question is WHY is it occuring. I have this issue currently, we are running a Websphere front end, and webphere keeps the Oracle processes alive if somebody drops out of Websphere, which is a real pain in the butt, because it means websphere has to be manually stopped and restarted periodically in order to release the Oracle processess. You may be having a similer issue.
When you, issue the shutdown immediate and still if you see oracle not available, then it would be the time to just abort the oracle instace through shutdown abort...
SVRMGR> shutdown immediate
ORA-01034: ORACLE not available
SVRMGR> shutdown abort
SVRMGR> startup mount
SVRMGR> recover database
SVRMGR> alter database open;
I once played the game of startup force and got into trouble... Its not worth the trouble that you have to go through to get it back if it goes wrong!!!
This is my personal opinion. I would take a long shot and save time....
Ahhh guys.... it was the number of processes running on the server. It wouldn't mater what shutdown method you used, if the number of processes have maxed out, you wouldn't be able to connect to do any on that until you kill one of the Dedicated Oracle processes in the background..
Bookmarks