We have been trying to resolve our SQL Server deadlocking issue without any luck. CAS keeps throwing deadlocks. Everything I'm seeing online says that our problem may be SQL Server specific due to the way they do locking and that the fix is what Drew Wills indicated above in 1). However, the weird part is that solution seems to be working just fine for uportal (no deadlocks) but not for CAS. Running jmeter against our test environment was causing a mess...so we pointed it to a mysql database and ran the same test and everything worked just fine...I mean we got a few "Read timed out" errors but no deadlocking exceptions...
On Mon, Jun 14, 2010 at 7:58 AM, Marvin Addison <[email protected]>wrote: > I wanted to update this thread with some further information. We have > recently switched database platforms for our JPA ticket registry from > Oracle to PostgreSQL. The hardware for the two platforms is > significantly different, with PostgreSQL being able to support much > higher transaction throughput as determined by our JMeter load tests. > Despite better overall performance we are seeing increased deadlocks > at a rate of several per day; on Oracle we would see deadlocks > ~1/month. This represents a 100-1000x increase, which is troubling. > We are actively investigating the issue and have some ideas about the > root cause. I'll update this thread again once we know more. > > M > > -- > You are currently subscribed to [email protected] as: > [email protected] > To unsubscribe, change settings or access archives, see > http://www.ja-sig.org/wiki/display/JSG/cas-user > -- Curtis Garman -- You are currently subscribed to [email protected] as: [email protected] To unsubscribe, change settings or access archives, see http://www.ja-sig.org/wiki/display/JSG/cas-user
