Re: ora-12500

jskishor_at_yahoo.co.in wrote:
> hi all, this is my first mail to this group , i have a peculiar problem
> of ora-12500. My database is on oracle 9.0.1.1.1 and have dedicated
> server.my operating system is windows 2003 server.
>
> the problem arises at 337 connections, i set my process to 500, my RAM
> is 2gb, job_queue_process=5.
> initially i thought,this error could be because of low physical memory
> and made it 4gb RAM, the change here is the problem arises at 454
> connections while i set my processes to 700,job_queue_process=20.
>
> so, i changed the
> shared_pool_size=100m,db_cache_size=704m,sga_max_size=1200m, i started
> up the second listener also but of no avail.
>
> please help me
>
> kishore
>

Is the 12500 error the one and only error you are getting? (often there is a sequence of cascading errors, usually the first one is the most useful to follow up on). Posting any other relevant info from your listener.log file will help.

Does the fact that you have such precise numbers where the problem happens (337/500 processes, 454/700 processes) mean that you are doing some reproducible testing where you rapidly connect and disconnect?

The only other time I have seen a pattern like this, it had to do with the frequency of service updates to the listener by PMON, which you cannot directly control (they generally range from every 1 to 10 minutes). For example, I have seen the service handler block with a 12516 error because the listener doesn't know about sessions that have disconnected from the database until PMON updates it. In other words, there can be a difference between the information held in the listener and actual number of connections to the database. If this is your situation, as you have seen raising the PROCESSES parameter is the only way to address it.

-Mark Bole Received on Thu Jan 13 2005 - 10:52:08 CST