Re: 8.1beta2 pg_dumpall inconsistencies - Mailing list pgsql-hackers

From Tom Lane
Subject Re: 8.1beta2 pg_dumpall inconsistencies
Date
Msg-id 3249.1128265158@sss.pgh.pa.us
Whole thread Raw
In response to 8.1beta2 pg_dumpall inconsistencies  ("Joshua D. Drake" <jd@commandprompt.com>)
List pgsql-hackers
"Joshua D. Drake" <jd@commandprompt.com> writes:
> [jd@jd pgsqldev]$ bin/pg_dumpall -U jd -D -o -p 5500
> ...
> pg_dump: INSERT (-d, -D) and OID (-o) options cannot be used together
> pg_dump: (The INSERT command cannot set OIDs.)
> pg_dumpall: pg_dump failed on database "postgres", exiting

> Why am I getting a partial dump when the options are not compatible?

Because -D and -o are pg_dump options, which pg_dumpall just passes down
without inspection.  I'm not sure that it makes sense to try to
replicate pg_dump's validity checking logic in pg_dumpall; I fear we'd
be more likely to cause problems by letting them get out of sync than
to solve problems by complaining a little sooner.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Martijn van Oosterhout
Date:
Subject: Re: effective SELECT from child tables
Next
From: Martijn van Oosterhout
Date:
Subject: Re: Socket problem using beta2 on Windows-XP