Re: ERROR: could not find tuple for trigger 37463634

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: mdione(dot)ext(at)orange(dot)com
Cc: "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: ERROR: could not find tuple for trigger 37463634
Date: 2012-03-07 17:11:01
Message-ID: 9861.1331140261@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

<mdione(dot)ext(at)orange(dot)com> writes:
> De: Tom Lane [mailto:tgl(at)sss(dot)pgh(dot)pa(dot)us]
>> You might try reindexing pg_trigger before doing anything more invasive,
>> just in case the tuple is there but it's not being found because of a
>> messed-up index.

> Well, at some point we tried "REINDEX DATABASE pnssi_profiles_bench;"
> and "REINDEX SYSTEM pnssi_profiles_bench;", would that be equivalent?

Yeah, that would have covered it.

> I'm pretty sure after what I have done some objects might still be
> dangling in the db (even more, now that I remember, at some point we did
> "delete from pg_trigger where tgname like '%01_110%';"; yes, I know,
> it's a blunt approach). Is there any way to find those, if they exist?

You could try the queries in the "oidjoins" regression test, which would
verify all the implied foreign key relationships in the system catalogs.
I don't think that covers pg_depend though; you'd need to gin up more
complicated queries if you wanted to look for dangling pg_depend
entries.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Mike Blackwell 2012-03-07 17:22:00 Altering a table with a rowtype column
Previous Message Nicolas Grilly 2012-03-07 16:53:29 Re: Full text search ranking: ordering using index and proximiti ranking with OR queries