Skip site navigation (1) Skip section navigation (2)

Large Functions and Index Rebuild

From: Terence Gannon <tgannon(at)stoneboat(dot)com>
To: "pgsql-general(at)hub(dot)org" <pgsql-general(at)hub(dot)org>
Subject: Large Functions and Index Rebuild
Date: 2000-01-27 17:58:22
Message-ID: 01BF68BE.EFA66E40.tgannon@stoneboat.com (view raw or flat)
Thread:
Lists: pgsql-general
I'm having a lot of trouble loading plpgsql functions longer than 100-150 
lines.  I can nurse them in by VACUUMing, then create the function, 
VACUUMing again, create the next one, but it seems that eventually, I 
corrupt the indices on pg_proc, and I can't even VACUUM without getting 
something like;

NOTICE:  Index pg_proc_prosrc_index: NUMBER OF INDEX' TUPLES (1092) IS NOT 
THE SAME AS HEAP' (1089)
NOTICE:  Index pg_proc_proname_narg_type_index: NUMBER OF INDEX' TUPLES 
(1092) IS NOT THE SAME AS HEAP' (1089)
NOTICE:  Index pg_proc_oid_index: NUMBER OF INDEX' TUPLES (1092) IS NOT THE 
SAME  AS HEAP' (1089)
pqReadData() -- backend closed the channel unexpectedly.
        This probably means the backend terminated abnormally
        before or while processing the request.
We have lost the connection to the backend, so further processing is 
impossible.
  Terminating.

Is there any way I can avoid this, or alternatively, any way of rebuilding 
these indices so that I can carry on?  Many thanx for your help...

Terry Gannon
stoneboat.com

Responses

pgsql-general by date

Next:From: Bruce MomjianDate: 2000-01-27 17:59:00
Subject: Re: [SQL] RE: [GENERAL] Problem with SELECT on large negative INT4
Previous:From: Tom LaneDate: 2000-01-27 17:52:37
Subject: Re: [SQL] RE: [GENERAL] Problem with SELECT on large negative INT4

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group