From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Johnson, Shaunn" <SJohnson6(at)bcbsm(dot)com> |
Cc: | "'''pgsql-general(at)postgresql(dot)org ' ' '" <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: how to use pg_resetxlog |
Date: | 2003-10-21 17:39:31 |
Message-ID: | 4625.1066757971@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Johnson, Shaunn" <SJohnson6(at)bcbsm(dot)com> writes:
> --i realize i must be getting rather boring right about
> --now, but i really appreciate the info / guidance.
Somewhere along here you may need to give up and revert to whatever your
most recent backup is :-(.
Taking a second look at the error message:
>> psql: FATAL 1: cannot open pg_attribute: No such file or directory
I realized that this probably isn't an XID issue as I first thought.
Either the pg_attribute file is actually missing (is there a file named
'1249' in the database directory?) or the relfilenode value for it is
clobbered in pg_class. In the latter case you could maybe find and fix
the erroneous data with a hex editor, but one would have to wonder what
else has been clobbered during the disk crash.
Thinking about it, it could be that the clobber is only in the database's
pg_internal.init cache file --- you could try renaming that out of the
way to see if the system can successfully make a new one.
Also, this failure should only affect one database in the cluster ---
can you connect to any other ones?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Dennis Gearon | 2003-10-21 18:23:58 | Re: how to use pg_resetxlog |
Previous Message | Joshua D. Drake | 2003-10-21 17:38:59 | Re: Recomended FS |