Re: Weird spikes in delay for async streaming replication on 9.1

On Thu, 26 Feb 2015 11:06:50 -0500
John Scalia <jayknowsunix(at)gmail(dot)com> wrote:

> Possibly, but that's why WE use a pair of standby servers, not a
> single one, so that all transactions get committed in a timely
> manner. The odds of both standbys failing at the same time are really
> small. Maybe your script should check which is latest WAL segment on
> each system first? That might show that you have a timedelay with
> getting the info to the standby.

That doesn't seem to be a problem. When I run this:

watch -n 0.1 'ps auxww|grep [w]al.*streaming'

on both primary and standby, the ID number after "streaming" increments
on both smoothly; there's no significant pause even when the delay starts
growing.

Regards,

David.