From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | btober(at)seaworthysys(dot)com |
Cc: | pgsql-general(at)postgresql(dot)org, Peter Eisentraut <peter_e(at)gmx(dot)net> |
Subject: | Re: pg_dump and alter database |
Date: | 2003-08-22 16:37:35 |
Message-ID: | 16269.1061570255@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
<btober(at)seaworthysys(dot)com> writes:
> I HAVE used pg_dumpall -g to make a backup of of users and groups, but
> this output does not include the ALTER DATABASE commands.
No, it looks like pg_dumpall dumps ALTER DATABASE operations for a
particular database when it dumps that database.
> Does the ALTER DATABASE command get written only when I do an
> unconditional pg_dumpall, i.e., produce a dump of all databases in the
> cluster? If that is the case, then I do think the feature ought to be
> included in pg_dump,
I think there were a couple of arguments for doing it this way. I can
see a permissions issue for one. pg_dumpall scripts assume they will be
run by superuser, but pg_dump scripts try to avoid that assumption.
Also, a pg_dump script doesn't (and shouldn't, IMHO) assume it knows the
name of the database it's being restored into.
Peter, do you recall any other issues?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Jan Wieck | 2003-08-22 16:49:33 | Re: Single-file DBs WAS: Need concrete "Why Postgres |
Previous Message | Dennis Gearon | 2003-08-22 16:35:52 | Re: Collation rules and multi-lingual databases |