| From: | Andreas <maps(dot)on(at)gmx(dot)net> |
|---|---|
| To: | pgsql-general(at)postgresql(dot)org |
| Subject: | Howto Replication for dummies? |
| Date: | 2012-03-24 22:52:04 |
| Message-ID: | 4F6E5014.9060107@gmx.net |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hi,
I've got a PG 9.1.3 on an OpenSuse 12.1 runnig that lately crashed
because of memory allocation errors probaply due to failing RAMs.
In this case the server died thursday afternoon and at first didn't
comlain when I restartet it.
Later some queries produced "out of memory" errors which is daft as the
box has 16 GB RAM.
Friday I found that pg_dump failes on one of the big tables.
Now I got the advice with a big stick that I should avoid data loss in
the future.
The DB is not exactly huge.
A sql dump is about 1.2 GB big but has about 150 schemas with many
smaller tables and there are 5 big tables that currently hold up to 9
mio lines.
Is there a howto that explains what to do to get a mirror-server that
follows the productive system?
Btw. what would happen to the slave in the event that the main server
messes up a table in a crash as above explained but goes on running
after an automatic ort manual restart?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jasen Betts | 2012-03-25 03:40:55 | Re: Table growing faster than autovacuum can vacuum |
| Previous Message | Andy Colson | 2012-03-24 14:28:25 | Re: plpgsql function to insert or update problem |