Re: [PERFORM] Spurious failure to obtain row lock possible in PG 9.1?

2012-09-25 Thread Albe Laurenz
henk de wit wrote: > I'm using Postgres 9.1 on Debian Lenny and via a Java server (JBoss AS 6.1) I'm executing a simple > "select ... for update" query: > > > SELECT > > importing > > FROM > > customer > > WHERE > > id = :customer_id > > FOR UPDATE NOWAIT > > > Once every 10 to 20 times P

Re: [PERFORM] Postgres becoming slow, only full vacuum fixes it

2012-09-25 Thread Kiriakos Tsourapas
Thank you, I will take this into consideration, since upgrading to 9 will be much harder I assume... On Sep 24, 2012, at 15:23, MirrorX wrote: > i remember having a server with 8.4.4 where we had multiple problems with > autovacuum. > if i am not mistaken there are some bugs related with vacuu

Re: [PERFORM] Postgres becoming slow, only full vacuum fixes it

2012-09-25 Thread Kiriakos Tsourapas
Hi Mark, When the problem appears, vacuuming is not helping. I ran vacuum manually and the problem was still there. Only full vacuum worked. As far as I have understood, autovacuuming is NOT doing FULL vacuum. So, messing around with its values should not help me in any way. Thanks > > Giv

Re: [PERFORM] Postgres becoming slow, only full vacuum fixes it

2012-09-25 Thread Kiriakos Tsourapas
Hi, Suggestion noted. Nevertheless, I cannot imagine what it would help. Actually, the cost_delay makes autovacuum freeze when it takes more time than expected, therefore, having it enabled should help the system. I may try it as a last possible resolution (remember that I have to wait for a c

Re: [PERFORM] Postgres becoming slow, only full vacuum fixes it

2012-09-25 Thread Kevin Grittner
Kiriakos Tsourapas wrote: > When the problem appears, vacuuming is not helping. I ran vacuum > manually and the problem was still there. Only full vacuum worked. > > As far as I have understood, autovacuuming is NOT doing FULL > vacuum. So, messing around with its values should not help me in > a

Re: [PERFORM] Postgres becoming slow, only full vacuum fixes it

2012-09-25 Thread Ondrej Ivanič
Hi, On 25 September 2012 21:10, Kiriakos Tsourapas wrote: > Suggestion noted. > Nevertheless, I cannot imagine what it would help. Actually, the cost_delay > makes autovacuum freeze when it takes more time than expected, therefore, > having it enabled should help the system. Yes, and I think tha