DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:X-Mailer:Message-ID:Date:From:Reply-To:Subject:To:MIME-Version:Content-Type;
	b=0AK3ea8PlnYiINrtEjmR9emmONLGzIA5hR90LR9AS5+lh0C53qq1CurnolgbC45F2hEvZc4fjwKc9iG4wLMm/KHSmTpIj3znGdjujIc3hJgktEfYb4VXu9V8xuZbR2HPu6E7C8FbpJ8ztU9eOO21j7S3Hp9sHJuExWEzzLpJcvc=;
Message-ID: <1311699626.15557.YahooMailNeo@web120012.mail.ne1.yahoo.com>
Date: Tue, 26 Jul 2011 10:00:26 -0700 (PDT)
From: A J <s5aly@yahoo.com>
Reply-To: A J <s5aly@yahoo.com>
Subject: Mechanics of streaming replication
To: PG Admin <pgsql-admin@postgresql.org>
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="0-318791354-1311699626=:15557"

--0-318791354-1311699626=:15557
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Hello,=0ATrying to understand the mechanics of streaming replication.=0AI u=
nderstand that the WAL SENDER on primary streams to WAL RECEIVER on standby=
. Also that depending on the settings, completed WAL files on the primary a=
re sent to the archive directory. If needed, the standby will move the file=
s from the archive directory to its wal directory (pg_xlog) and start apply=
ing them to its data files.=0A=0AMy questions are:=0A1. Is it that the rest=
ore command in the recovery.conf only controls when files get moved from ar=
chive to the WAL directory on the standby ? If so, when does the restore co=
mmand get kicked in ? Only at failover or whenever the connection between t=
he primary and the standby is terminated ?=0A2. The archive directory conte=
nts will always lag behind the current WAL on primary by 16MB (unless you t=
weak archive_timeout parameter at the expense of bloating your archive stor=
age). So in a typical setup, you always risk loosing up-to 16MB of committe=
d data at failover. Right ? And I assume there is no way to reconcile this =
16MB automatically when the crashed ex-primary server comes up again. Right=
 ?=0A3. I don't think 'wal_keep_segments' has too much of a practical use. =
It keeps the WALs in the pg_xlog directory of the primary which is local on=
ly to primary. So in case of primary going down, all those WALs will be los=
t to the standby. I guess most people primarily rely on 'archive_command' a=
nd not wal_keep_segments, and use =A0Right ?=0AOr is it that it is preferre=
d to keep wal_keep_segments high enough so that standby need not go to arch=
ive directory every now and then ?=0A4. Does the standby switch-back from r=
eading the archive to streaming once the connection between primary and sec=
ondary is restored ?=0A=0AThanks for any inputs.
--0-318791354-1311699626=:15557
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

<html><body><div style=3D"color:#000; background-color:#fff; font-family:ti=
mes new roman, new york, times, serif;font-size:12pt"><div>Hello,</div><div=
>Trying to understand the mechanics of streaming replication.</div><div>I u=
nderstand that the WAL SENDER on primary streams to WAL RECEIVER on standby=
. Also that depending on the settings, completed WAL files on the primary a=
re sent to the archive directory. If needed, the standby will move the file=
s from the archive directory to its wal directory (pg_xlog) and start apply=
ing them to its data files.</div><div><br></div><div>My questions are:</div=
><div>1. Is it that the restore command in the recovery.conf only controls =
when files get moved from archive to the WAL directory on the standby ? If =
so, when does the restore command get kicked in ? Only at failover or whene=
ver the connection between the primary and the standby is terminated ?</div=
><div>2. The archive directory contents will always lag behind the
 current WAL on primary by 16MB (unless you tweak archive_timeout parameter=
 at the expense of bloating your archive storage). So in a typical setup, y=
ou always risk loosing up-to 16MB of committed data at failover. Right ? An=
d I assume there is no way to reconcile this 16MB automatically when the cr=
ashed ex-primary server comes up again. Right ?</div><div>3. I don't think =
'wal_keep_segments' has too much of a practical use. It keeps the WALs in t=
he pg_xlog directory of the primary which is local only to primary. So in c=
ase of primary going down, all those WALs will be lost to the standby. I gu=
ess most people primarily rely on 'archive_command' and not wal_keep_segmen=
ts, and use &nbsp;Right ?</div><div>Or is it that it is preferred to keep w=
al_keep_segments high enough so that standby need not go to archive directo=
ry every now and then ?</div><div>4. Does the standby switch-back from read=
ing the archive to streaming once the connection between primary and
 secondary is restored ?</div><div><br></div><div>Thanks for any inputs.</d=
iv></div></body></html>
--0-318791354-1311699626=:15557--