From: | Uwe Schroeder <uwe(at)bunspace(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | Bill Moran <wmoran(at)potentialtech(dot)com>, Steeles <steeles(at)gmail(dot)com> |
Subject: | Re: What is better method to backup postgresql DB. |
Date: | 2010-10-27 05:27:40 |
Message-ID: | 201010262227.40713.uwe@bunspace.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
> In response to Steeles <steeles(at)gmail(dot)com>:
> > new to postgresql. need to backup postgresql DB, which way is better to
> > backup DB.
> >
> > from training, I learned that we can backup the whole PGdata and other
> > directories to achieve backup goal, originally I was planned to schedule
> > jobs to use pgdump to backup all DBs.
> >
> > so which way is better for backup or there is other way to backup PGDB.
>
> Use pg_dump or pg_dumpall. If you're at the level that you have to ask
> this question, then you'll have nothing but trouble getting reliable
> backups by backing up directories.
>
> Also, since you're new to PostgreSQL, I _HIGHLY_ recommend that you don't
> assume that you're getting backups until you can demonstrate that you can
> restore them.
Good advise! For all my production systems I do a life backup/replication
using slony plus a nightly pg_dump which is then copied to a remote server.
That way, no matter what happens, I have a backup that's a maximum of 1 day
old (if I have to use the pg_dump) or a few seconds behind (if the replicated
database isn't hit - which sure is also remote and not in the same physical
location)
Whatever you use, always make sure you actually can restore the backup.
From | Date | Subject | |
---|---|---|---|
Next Message | Gabriele Bartolini | 2010-10-27 06:05:47 | Re: DB become enormous with continuos insert and update |
Previous Message | ljb | 2010-10-27 01:11:42 | Re: [GENERAL] Gripe: bytea_output default => data corruption |