I have two Pg 9.0.x servers and I have recently started using streaming replication. In monitoring with nagios, I find that almost every day there comes a point when the standby server experiences a delay in replaying transactions (by a few thousand). Receiving is up to date. and of course this usually happens in the wee hours of the morning. I'm not sure what might be causing that. I do a nightly dump of each database and at first I though maybe that was causing the replay delay if the db was locked against writes during the dump, but the dumps and the delays are not lining up.
There are also times when both are behind by quite a few thousand. Any ideas on what might be causing these or how to try to catch the culprit?
Thanks.
--
Principal Software Engineer
Motorola Mobility Open Source Technologies
Advanced Platforms and Services
