Hi
Maybe you must reset the WAL's
( http://www.postgresql.org/docs/7.3/interactive/app-pgresetxlog.html )
after restoring from tarball if postgres doesn't start.
Am Donnerstag, den 30.06.2005, 07:34 +0200 schrieb jehan:
So I must one way or another run a 7.3, restore the file from the
tarball
Empric tests have shown that read rrequests seem to still get proceeded
whilst write requests get queued until all dirty pages have been written.
That's however just testing and looking. What actually happens is a secret.
|-Original Message-
|From: KÖPFERL Robert
|Sent: Montag, 27.
Actually, I haven't
done such thing, yet. However the intended method or a possible way is
this:
Manage
schema-snapshots and put them in VCS
Keep a file of
changes with version stamps (also in VCS). Document each change to the db (so
SQL commands you issue to change its struct). So one can
Thanks. This works fine on Linux, but I could't get it to work on
FreeBSD 5.3 - any ideas?
On 24.06.2005 16:55, Peter Eisentraut wrote:
Use initdb --locale=de_DE.utf8 and that should be all.
---(end of broadcast)---
TIP 7: don't forget to
OK, I'am not yet at restarting postgres .. but if at get pb then I check
that , thanks !
For now , How can I tell from the bare file the mapping between a
database name and the number appearing in /var/lib/pgsql/base directory
I have :
pgsql/data/base/1/
pgsql/data/base/16975/
Hi,
try this:
psql -t -d yourdb -c SELECT datid FROM pg_stat_database WHERE
datname='yourdb';
http://www.postgresql.org/docs/8.0/static/monitoring-stats.html
Greetings,
Martin
Am Donnerstag, den 30.06.2005, 12:57 +0200 schrieb jehan-free:
OK, I'am not yet at restarting postgres .. but if at
Am Donnerstag, 30. Juni 2005 11:58 schrieb Hannes Dorbath:
Thanks. This works fine on Linux, but I could't get it to work on
FreeBSD 5.3 - any ideas?
I hear that FreeBSD doesn't support Unicode, so you're probably out of luck
for now.
--
Peter Eisentraut
Hannes Dorbath [EMAIL PROTECTED] writes:
I hear that FreeBSD doesn't support Unicode, so you're probably out of luck
for now.
Hm, but the locales are there, it's just named de_DE.UTF-8 instead of
de_DE.utf8 on FreeBSD. InitDB crashs with the following:
creating template1 database in
Just turned syslog logging on on a Linux box, and the queries are coming
through like:
^Iincident, building, location, category, remark, building_access,^M
with the control characters in them ... it doesn't do this on my FreeBSD
box, so figure I'm mis-configuring something in syslog itself
I'm currently testing PITR features of 8.0 release. I followed the steps
given by the documentation and thus deleted the WAL files in pg_xlog
after having restored the database.
But when I put the recovery.conf file and try to start postgres, it
fails. The log file says :
2005-06-30 17:57:50
Romain,
This is a problem I ran into during testing as well. It seems that your base
backup must also include the in use wal file. IN your case:
0001000B. So, I think the restore documentation is slightly
misleading. In my experience, what you really need to do is remove
Does the same thing here, which is kind of annoying when you log queries for
debugging and can't copy/paste them easily.
I think it's a linux syslog bug/feature.
On Thursday 30 June 2005 07:32 am, Marc G. Fournier wrote:
Just turned syslog logging on on a Linux box, and the queries are coming
very good, that worked fine :-)
I restored the files from tar, started a postgresql 7.3 on an old redhat
9 ! pg_dump my database, psql it back to my postgresql 7.4 on my
production RHEL4 server .
still a small pb, I seem to have lost authentification. (although
pg_hba.conf was restore also)
ezpublish_db-# ALTER USER ezpublish SET PASSWORD secret;
ERROR: syntax error at or near $ at character 1
I wonder why you have ezpublish_db-# instead of ezpublish_db=#? I just
noticed it, and to me it happens usually when something's been carried
over from the previous line. My 2 pence...
Jeff Frost [EMAIL PROTECTED] writes:
When I tested with our system, it would not attempt to retrieve that first
WAL
via the restore command. All the ones after that it was happy to do it for,
but if that one did not exist, it told me to get lost.
Would you be more specific please?
On Thu, 30 Jun 2005, Tom Lane wrote:
Would you be more specific please? Looking at the code, I can't see a
reason that the first one would be treated differently from others.
What *exactly* did you get?
Unfortunately, I do not have my output saved, but here's basically what
happened:
I
Jeff Frost [EMAIL PROTECTED] writes:
I'll have to retest that particular scenario and verify that something
else was not causing the problem when I have a chance now that I have
a fully functional and scripted out base backup and recover solution;
however, it did not give me any of the errors
17 matches
Mail list logo
