7.2.2.x memory leak ... possible solution
From: Bill Beaton <willyb_at_cadvision.com>
Date: 1996/01/14
Message-ID: <4d9t5k$lf0_at_huey.cadvision.com>#1/1
Date: 1996/01/14
Message-ID: <4d9t5k$lf0_at_huey.cadvision.com>#1/1
We applied the ORACLE supplied fix for our problems with ORA-4031 on Friday evening at 1900 MST (bug326110), and have now been running torture tests, as well as our normal client load since. NO 4031 errors, or ORA-0600, or any other problems.
With not flushing the shared pool every 300 seconds, performance is at last hitting the levels that we expected when we upgraded to 7.2. The one known outstanding bug is still that MTS doesn't work, but for now, the swapping level is tolerable. If the system keeps performing at current levels until Tuesday nite, then I suppose I'll have to admit that ORACLE has actually found and properly repaired the memory leak.
Tom Bellavi, this may also be the fix that you need.
-- Names: Bill Beaton beatonb_at_cadvision.com Bill_Beaton_at_calgary.qc-data.com Phone: (403) 295-3254 (Home) (403) 266-8622 (work)Received on Sun Jan 14 1996 - 00:00:00 CET