1) The .trcs you are deleting are gold information on your issue. Next time read them to figure out where these processes are coming.
2) It sounds like it's some batch processing scheduled for saturday that floods your database with sessions or processes. Check v$session/v$process at the time the issue is happening to find out where they are coming from and what are they doing.
3) Navigate to the root folder of the file system and run: du -m ./* | sort -n | tail -20 That will give you the top 20 biggest folders along with their total MB size.
4) It could also be the case that some queries and/or jobs are implementing a high degree of parallelism which would also fill up to your maximum processes parameter giving you that error. (depends also on your parallel_max_servers)
By the way, the DB is victim here. It's simply being flooded with sessions/processes by someone or something. It's up to you to find what/who.
What exactly do you know about maintaining & troubleshooting Oracle DB, besides asking for your hand to be held by folks here?
>It has been two(2) weeks now that every Saturday our system encounters maximun 200 process exceeded.
back out the change that was made 2 weeks ago
>Whats else logs do I need to delete at OracleHome?
stop fighting the symptom & stop the errant files from being generated in the first place.
Can I not find it is AWR, ASH, ADDM repors?
Yes you could infer it through a differential-AWR.
Compare two range of snapshots, one from where the issue is happening and a baseline (normal).
In case of sessions:
Compare number of sessions as described in the AWR header session.
In case of PQs:
On instance activity session of the AWR: Compare stat : "queries parallelized"
Watch for the wait events on the problem snapshot that start with PX which would indicate parallel query processing.
Hi Again Al,
I run :
du -m ./* | sort -n | tail -20
But its the folder that is being displayed. Is there a way to display the file itself?