What are the steps for upgrade of a primary/PITR standby pair to the
latest 8.3 patchset, 8.3.6-8.3.10? I don't see this in the docs.
Should there be something added to 15.4. Upgrading?

My guess is:
1. turn off the standby config in primary
2. upgrade primary
3. rebuild standby

That way you don't loose the standby if something goes in the toilet.
Is that right?

--
Sent via pgsql-admin mailing list (pgsql-admin [at] postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

--=-2wyVDBJ5GnDsnBjjf6oT
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, 2010-05-11 at 09:16 -0400, Ray Stell wrote:
> What are the steps for upgrade of a primary/PITR standby pair to the
> latest 8.3 patchset, 8.3.6-8.3.10? I don't see this in the docs.
> Should there be something added to 15.4. Upgrading?
>
> My guess is:
> 1. turn off the standby config in primary
> 2. upgrade primary
> 3. rebuild standby

You need to upgrade standby first, restart it. Then, upgrade master and
restart it. No need to turn off standby config.

Regards,
--
Devrim G=C3=9CND=C3=9CZ
PostgreSQL Dan=C4=B1=C5=9Fman=C4=B1/Consultant, Red Hat Certified Engineer
PostgreSQL RPM Repository: http://yum.pgrpms.org
Community: devrim~PostgreSQL.org, devrim.gunduz~linux.org.tr
http://www.gunduz.org Twitter: http://twitter.com/devrimgunduz

--=-2wyVDBJ5GnDsnBjjf6oT
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEABECAAYFAkvpW60ACgkQtl86P3SPfQ4ufgCfaThGzAWJvcwpWBZHp1eE QT6b
faYAnRx91RIMkxPDtca5QrGx3dRaKOKI
=sfQU
-----END PGP SIGNATURE-----

--=-2wyVDBJ5GnDsnBjjf6oT--

On Tue, May 11, 2010 at 04:29:18PM +0300, Devrim G?ND?Z wrote:
> On Tue, 2010-05-11 at 09:16 -0400, Ray Stell wrote:
> > What are the steps for upgrade of a primary/PITR standby pair to the
> > latest 8.3 patchset, 8.3.6-8.3.10? I don't see this in the docs.
> > Should there be something added to 15.4. Upgrading?
> >
> > My guess is:
> > 1. turn off the standby config in primary
> > 2. upgrade primary
> > 3. rebuild standby
>
> You need to upgrade standby first, restart it. Then, upgrade master and
> restart it. No need to turn off standby config.

do you know why might be going south?

the standby is applying WAL:
----------------------------
,30357,,2010-05-12 17:04:33.927 EDT,4beb17d2.7695,5,2010-05-12 17:04:18 EDT,0, LOG: automatic recovery in progress
,30357,,2010-05-12 17:04:34.534 EDT,4beb17d2.7695,6,2010-05-12 17:04:18 EDT,0, LOG: redo starts at 0/567D982C
,30357,,2010-05-12 17:04:34.630 EDT,4beb17d2.7695,7,2010-05-12 17:04:18 EDT,0, LOG: restored log file "000000020000000000000057" from archive
,30357,,2010-05-12 17:04:35.315 EDT,4beb17d2.7695,8,2010-05-12 17:04:18 EDT,0, LOG: restored log file "000000020000000000000058" from archive
....

,30357,,2010-05-12 17:05:21.555 EDT,4beb17d2.7695,23,2010-05-12 17:04:18 EDT,0, LOG: restored log file "000000020000000000000067" from archive
,30357,,2010-05-12 17:05:22.408 EDT,4beb17d2.7695,24,2010-05-12 17:04:18 EDT,0, LOG: restored log file "000000020000000000000068" from archive


shutdown standby:
-----------------
,30355,,2010-05-13 08:59:49.547 EDT,4beb17d2.7693,1,2010-05-12 17:04:18 EDT,0, LOG: received fast shutdown request
,30357,,2010-05-13 08:59:49.547 EDT,4beb17d2.7695,25,2010-05-12 17:04:18 EDT,0, FATAL: could not restore file "000000020000000000000069" from archive: return code 15
,30355,,2010-05-13 08:59:49.551 EDT,4beb17d2.7693,2,2010-05-12 17:04:18 EDT,0, LOG: startup process (PID 30357) exited with exit code 1
,30355,,2010-05-13 08:59:49.551 EDT,4beb17d2.7693,3,2010-05-12 17:04:18 EDT,0, LOG: aborting startup due to startup process failure


flip to 8.3.10 and restart standby:
-----------------------------------
,3440,,2010-05-13 09:06:35.720 EDT,4bebf95b.d70,1,2010-05-13 09:06:35 EDT,0, LOG: database system was interrupted while in recovery at log time 2010-05-12 17:01:05 EDT
,3440,,2010-05-13 09:06:35.720 EDT,4bebf95b.d70,2,2010-05-13 09:06:35 EDT,0, HINT: If this has occurred more than once some data might be corrupted and you might need to choose an earlier recovery target.
,3440,,2010-05-13 09:06:35.720 EDT,4bebf95b.d70,3,2010-05-13 09:06:35 EDT,0, LOG: starting archive recovery
,3440,,2010-05-13 09:06:35.732 EDT,4bebf95b.d70,4,2010-05-13 09:06:35 EDT,0, LOG: restore_command = '/usr/local/pgsql/bin/pg_standby /data/pgsql/wals/alerts_oamp %f %p %r'
sh: /usr/local/pgsql/bin/pg_standby: No such file or directory
,3440,,2010-05-13 09:06:35.734 EDT,4bebf95b.d70,5,2010-05-13 09:06:35 EDT,0, FATAL: could not restore file "00000002.history" from archive: return code 32512
,3438,,2010-05-13 09:06:35.735 EDT,4bebf95b.d6e,1,2010-05-13 09:06:35 EDT,0, LOG: startup process (PID 3440) exited with exit code 1
,3438,,2010-05-13 09:06:35.735 EDT,4bebf95b.d6e,2,2010-05-13 09:06:35 EDT,0, LOG: aborting startup due to startup process failure

--
Sent via pgsql-admin mailing list (pgsql-admin [at] postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Ray Stell <stellr [at] cns.vt.edu> wrote:
> ,3440,,2010-05-13 09:06:35.734 EDT,4bebf95b.d70,5,2010-05-13
> 09:06:35 EDT,0,FATAL: could not restore file "00000002.history"
> from archive: return code 32512

Return code 32512? I think you'd be OK if your recovery script
returned 1 when it didn't find this file.

-Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin [at] postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

On Thu, May 13, 2010 at 09:31:27AM -0500, Kevin Grittner wrote:
> Ray Stell <stellr [at] cns.vt.edu> wrote:
> > ,3440,,2010-05-13 09:06:35.734 EDT,4bebf95b.d70,5,2010-05-13
> > 09:06:35 EDT,0,FATAL: could not restore file "00000002.history"
> > from archive: return code 32512
>
> Return code 32512? I think you'd be OK if your recovery script
> returned 1 when it didn't find this file.


Tom wrote last year:

Hmm ... 32512 is 0x7F00, which I think means exit(127), which is
generally what the shell returns when it can't find the program
it's supposed to execute.

> sh: /usr/lib/postgresql/8.2/bin/pg_standby: not found
> sh: /usr/lib/postgresql/8.2/bin/pg_standby: not found
> sh: /usr/lib/postgresql/8.2/bin/pg_standby: not found
> sh: /usr/lib/postgresql/8.2/bin/pg_standby: not found

duh. measure twice, cut once.

--
Sent via pgsql-admin mailing list (pgsql-admin [at] postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin