We have different versions. The most recently patched versions run RHAS 2.1 Update 6 and RHAS 3.0 Update 3. We had opened a case with Redhat for a system running 2.4.9-e.49summit (RH2.1 Update 4 or 5) and we got a response back claiming it's fixed it that version. Thanks, Denny -- Denny Koovakattu Quoting Magnus Andersen <Magnus.Andersen@xxxxxxxxxxxxxxx>: > Denny, > > What kernel version are you running? > > Magnus > > -----Original Message----- > From: Denny Koovakattu [mailto:oracle-l@xxxxxxxxxxxxxx] > Sent: Friday, July 22, 2005 3:13 PM > To: Magnus Andersen > Cc: Oracle-L (E-mail) > Subject: RE: Know anyone? > > > > We had similar problems, but they were caused by kswapd. If the CPU usage > is > mostly system with zero or close to zero CPU being used by the user > processes, > it indicates the kswapd bug. I don't think changing pagecache eliminates the > bug, but just moves the point at which it gets triggered. We also suspect > that > more physical memory increases the chances of encountering the bug. Hugetlb > might help for these systems. Mostly we have been hit by this bug when we > establish few Oracle connections at the same time. We have also had issues > running find or du. It's unpredictable. The last response we got from Redhat > was that it's fixed in the exact kernel version we are running ;) > > Thanks, > Denny > -- > Denny Koovakattu ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. -- //www.freelists.org/webpage/oracle-l