Re: version mismatch message - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: version mismatch message
Date
Msg-id Pine.LNX.4.56.0307311043230.1729@krusty.credativ.de
Whole thread Raw
In response to version mismatch message  ("Christopher Kings-Lynne" <chriskl@familyhealth.com.au>)
Responses Re: version mismatch message  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Christopher Kings-Lynne writes:

> I get this in my logs from our backups.  I ahve explicitly put -i in
> pg_dumpall.  How about we totally suppress this message if -i is supplied,
> because obviously the person knows perfectly well it's proceeding despite
> version mismatch?

I think if we did that, then people would become inclined to use pg_dump
-i automatically.  The reason that the version mismatch check exists is
that we *know* that pg_dump is unable to dump a reasonably complex
database from certain versions correctly.

Nonetheless, I think a pg_dump 7.3.2 dumping a 7.3.3 database should not
raise a version mismatch message at all.  (In general, pg_dump x.y.z1
ought to be able to dump a server x.y.z2 for any z1 and z2.)  Check out
_check_database_version in pg_backup_db.c if you like to fix it.

-- 
Peter Eisentraut   peter_e@gmx.net


pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: patch queue
Next
From: Peter Eisentraut
Date:
Subject: Re: [PATCHES] [PATCH] Re: [pgsql-advocacy] Why READ ONLY transactions?