RE: PG 9.6 Slow inserts with long-lasting LWLocks

: pgsql-performa...@postgresql.org Subject: Re: PG 9.6 Slow inserts with long-lasting LWLocks Sporadic insert slowness could be due to lock delays (locktype=extend) due to many concurrent connections trying to insert into the same table at the same time. Each insert request may result in an extend

Re: PG 9.6 Slow inserts with long-lasting LWLocks

ut I can't catch any evidence telling that system run out of 1GB RAID controller cache on writes. Please assist in understanding meaning and nature of long semop calls appearances. -- Regards, Pavel Suderevsky *From: *Pavel Suderevsky <mailto:psuderev...@gmail.com> *Sent: *Monday, Ja

RE: PG 9.6 Slow inserts with long-lasting LWLocks

on_locked=0 sleep_on_page=0 get_request=0 sleep_on_buffer=0 14:33:55 TOTAL=7 wait_transaction_locked=0 sleep_on_page=0 get_request=0 sleep_on_buffer=0 I understand that RAID-6 is not the best option, but I can't catch any evidence telling that system run out of 1GB RAID controller cache on

PG 9.6 Slow inserts with long-lasting LWLocks

Hi, I have an issue with sporadic slow insert operations with query duration more than 1 sec while it takes about 50ms in average. Configuration: OS: Centos 7.2.151 PostgreSQL: 9.6.3 CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz Memory: totalusedfree shared buf