Re: ORA-04030: Out of process memory ERROR

No other error. It ends up that Oracle claims that there is a problem with either the BitMap indexes, sort_area_size, or hash_joins. They are working on it now.

Anyway, I got around it by removing all bitmapped indexes ( this alone did not work ). Reducing the value of sort_area_size ( was at 16M, reduced to 4M and the query worked ). The following day, Oracle called back and told me to set HASH_JOIN_ENABLED=FALSE. Since I had already got my query to work, I dont know if this would have made any difference.

As a reply to the earlier posts, we were watching the memory usage of the AIX box, it was OK. We had also re-booted the machine and I was the only user ( AIX or ORACLE ) on the entire box.

Thanks for the input.

Michael.

In article <3793ECD5.4EA11C3A_at_sprynet.com>,   Steve Perry <sperry_at_sprynet.com> wrote:
> What's the other error? I've seen the 4030 followed by a ora-07324. In
fact
> I'm using 7.3.4 on aix 4.3.2 and I got one last month. We bumped up
the
> ulimit and I wrote a script to monitor the application process's
memory
> usage (ps vg). Last week it struck again. What I found was that the
shadow
> process for the application leveled out at 50 meg or so for 2 weeks.
All of
> a sudden, it started growing by leaps and bounds until we got the
error at
> 200 meg. I looked into it at Oracle's metalink has some info on it. My
> problem seems to be some type of memory leak. They're looking into it
now.
> If the process has been running for extended periods of time, try
restarting
> it. I don't know if this is the same problem you're having, but it may
be
> something you can look into.
>
> Steve

Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't. Received on Tue Jul 20 1999 - 12:08:00 CDT