I found the solution, and it's a bit obscure.
The problem is detailed in MetaLink Note 152868.1
Problem Description
-------------------
You have seen in the v$process view that the number of processes without
a
session are growing up one every 5 minutes, until the processes limit is
reached
and an ORA-20 is shown.
You have tried to debug the process at system level without success,
because you've
seen that the process doesn't exists anymore.
You have stopped all third applications and Oracle Services including
the listener,
but the number of process still is growing up.
Solution Description
--------------------
Open %ORACLE_HOME%/dbs/operf81.log file to check if an ORA-1017 error is
generated
every time Oracle Performance Monitor tries to connect to the database.
If so, the credentials used by Oracle Performance Monitor are out of
date.
Use operfcfg utility or modify appropiate entries in the registry
(HKLM/System\CurrentControlSet\Services\Oracle8\Performance), to update
the
[EMAIL PROTECTED]
Sent by: [EMAIL PROTECTED]
08/14/2002 12:01 PM
Please respond to ORACLE-L
To: Multiple recipients of list ORACLE-L <[EMAIL PROTECTED]>
cc:
Subject: ORA-20 - Maximum Processes Exceeded
Dear list,
I'm running into a strange problem on one of my dev databases.
First, the particulars:
Dell Dual Pentium III
1 Gig Ram
Oracle 8.1.6
The database in question is used for testing and development on
a customized 3rd party app.
This past week I've seen the app max out the number of processes
by logging in with one of it's utilities until there were no processes
left.
This morning something a little different happened. The ORA-20
popped up again, but this time, there were not corresponding
entries in v$session.
Joining v$session and v$process returned only 31 rows, while
there were actually 150 rows in v$process. Very strange.
I then tried to use oradebug to get a processstate dump, and the
first attempt completed very quickly, and no trace file was generated.
e.g.
SVMGR> oradebug setorapid 100
Windows thread id: 664, image: ORACLE.EXE
SVRMGR> oradebug dump processstate 10
SVRMGR> .
I then tried to use oradebug to get a processstate dump again with
a different level, and that never did complete.
SVMGR> oradebug setorapid 100
Windows thread id: 664, image: ORACLE.EXE
SVRMGR> oradebug dump processstate 1
...
A subsequent attempt on another PID with a dump level
of 10 did the same thing, it hung.
At this point I bounced the database, which took a few minutes.
Any ideas on how to go about troubleshooting this?
I may run a periodic job to capture some v$process and v$session info
for post mortem use, but a dump would be helpful to find out what the
orphaned processes are.
Thanks,
Jared
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author:
INET: [EMAIL PROTECTED]
Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
San Diego, California -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
--
Please see the official ORACLE-L FAQ: http://www.orafaq.com
--
Author:
INET: [EMAIL PROTECTED]
Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
San Diego, California -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
