Yes, they have exactly the same specification.
On Sat, Aug 25, 2012 at 2:33 PM, Lonni J Friedman <netllama(at)gmail(dot)com>wrote:
> Are both servers identical hardware?
> On Sat, Aug 25, 2012 at 12:37 AM, Prima Chairunnanda <prima(dot)ch(at)gmail(dot)com>
> > Hi Everyone,
> > I have some questions about the PostgreSQL streaming replication
> feature. I
> > have been able to setup a hot-standby server, and I can see updates on
> > master server being replayed on the standby server.
> > However I notice that once I start making read-only queries, the standby
> > starts lagging behind the master server. After a while, the delay can
> > over 1 minute, even though I set the max_standby_streaming_delay
> > to 15s. I am under the impression that once a WAL record is lagging 15s
> > behind, then it should trump any read-only queries in the system.
> > Below are the values of pg_last_xlog_receive_location and
> > pg_last_xlog_replay_location at several time points
> > now 03:06:16
> > pg_last_xlog_receive_location 0/59A31620
> > pg_last_xlog_replay_location 0/599A6B90
> > now 03:07:18
> > pg_last_xlog_receive_location 0/59A98460
> > pg_last_xlog_replay_location 0/59A1BE58
> > now 03:07:31
> > pg_last_xlog_receive_location 0/59AA8F48
> > pg_last_xlog_replay_location 0/59A22288
> > So as we can see the WAL record at 0/59A31620 was received at 03:06:16,
> > more than a minute later at 03:07:31, it still hasn't been replayed.
> > Am I configuring the wrong thing? Is there an option to increase the
> > priority of the WAL replay once it has been lagging for a certain amount
> > WAL record?
> > Thanks in advance for your help.
> > Prima
In response to
pgsql-novice by date
|Next:||From: Jeremy Wells||Date: 2012-08-26 07:08:36|
|Subject: best index for timestamp field null and not null queries|
|Previous:||From: Bruce Momjian||Date: 2012-08-25 19:31:15|
|Subject: Re: Learning how to write stored procedures|