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

Re: GIN fast insert database hang

From: Teodor Sigaev <teodor(at)sigaev(dot)ru>
To: Robert Haas <robertmhaas(at)gmail(dot)com>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: GIN fast insert database hang
Date: 2009-02-12 17:56:58
Message-ID: 499462EA.9040902@sigaev.ru (view raw or flat)
Thread:
Lists: pgsql-hackers
> This freezes the whole system even with autovacuum = off in
> postgresql.conf.  As before, the backends wait on a semop() call.
Fixed. There was a deadlock of LockBufferForCleanup and LockBuffer(SHARE). 
Redesign that place to downgrade  LockBufferForCleanup to LockBuffer(EXCLUSIVE) 
with correction of page's locking during scan of pending list.


> I was able to reproduce the recovery failure this way once as well,
> but that part of the problem seems to be much more erratic.  Most of
Fixed too. I missed comments on XLogInsert:
  * NB: this routine feels free to scribble on the XLogRecData structs,
  * though not on the data they reference.  This is OK since the XLogRecData
  * structs are always just temporaries in the calling code.

and I reused once initialized XLogRecData many times in a loop.
-- 
Teodor Sigaev                                   E-mail: teodor(at)sigaev(dot)ru
                                                    WWW: http://www.sigaev.ru/

Attachment: fast_insert_gin-0.27.gz
Description: application/x-tar (24.3 KB)

In response to

pgsql-hackers by date

Next:From: Bruce MomjianDate: 2009-02-12 18:00:28
Subject: Re: pg_migrator and handling dropped columns
Previous:From: Tom LaneDate: 2009-02-12 17:21:07
Subject: Re: DISCARD ALL failing to acquire locks on pg_listen

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