From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Robert Treat <xzilla(at)users(dot)sourceforge(dot)net> |
Cc: | "Patrick S(dot) Riedel" <priedel(at)asllc(dot)net>, pgsql-admin(at)postgresql(dot)org |
Subject: | Re: Convert v7.0.2-2c1 DB |
Date: | 2003-02-12 06:11:06 |
Message-ID: | 1192.1045030266@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Robert Treat <xzilla(at)users(dot)sourceforge(dot)net> writes:
> On Tue, 2003-02-11 at 17:27, Patrick S. Riedel wrote:
>> I have a client with a v7.0.2-2c1 database stored on media. The
>> database files were stored raw, not dumped. Their current pgsql version
>> is 7.1; they do not have v7.0.2-2c1 installed anywhere.
> Lordy, that's a fine mess you've got there.
Yup...
> As far as what version of 7.0.x to use, I don't recognize that release
> so it could very well be a release candidate version, but might also be
> a vendor/rpm/debian style version tag.
Any released 7.0.* version should be disk-level-compatible. Given that
this is a 7.0.2-something and not a 7.0-something I don't think you need
to worry that it might be a 7.0 prerelease. So as long as you have the
complete $PGDATA directory tree (not a subset that omits pg_log, for
example) I think you should be able to compile any 7.0.* release and
fire it up against that directory tree. Then pg_dump and away you go.
> In any/all cases, always copy the data over, never do anything on the
> original files.
Check, always keep a pristine copy until you *know* you have a working,
cross-checked, backed-up database in a later version...
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Oliver Elphick | 2003-02-12 06:47:52 | Re: Create Perl Support |
Previous Message | Stephan Szabo | 2003-02-12 04:23:55 | Re: make check: 1 of 89 tests failed. |