From: | Klint Gore <kgore4(at)une(dot)edu(dot)au> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | pgsql-general(at)postgreSQL(dot)org |
Subject: | Re: pg_dump - lost synchronization with server: got message type "d", length 6036499 |
Date: | 2008-07-04 00:34:56 |
Message-ID: | 486D7030.3090604@une.edu.au |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Tom Lane wrote:
> Klint Gore <kgore4(at)une(dot)edu(dot)au> writes:
> > Tom Lane wrote:
> >> Would you be willing to send me a "pg_dump -s" (ie, just schema no
> >> data) dump of this DB? Off-list of course.
> >>
> > attached. created with pg_dump 8.3.3 win32 to the 8.3.1 win32 server.
>
> Thanks. When I reload this here and try to pg_dump it, I see no evidence
> of any problem --- pg_dump completes pretty quickly and seems to eat
> no more than a megabyte or so. I'm not sure what to make of that.
> Could there be a Windows-specific memory problem in pg_dump?
>
Any more thoughts on this?
The file being produced when dumping the table in question is about 1
gig when it crashes. In the last check-in for fe-protocol3.c, there's
notes about 1 gig strings so I'm wondering if my data when received by
pg_dump (or output by the 8.3.1 server) goes over? 270m of bytea
converted to \xxx octal strings would make something > 1 gig. But the
270m is spread over 400 rows with nothing bigger than 3m.
If there's no interest in researching this any further, then I'll stop
the server and take a file system copy of the data directory and upgrade
it to 8.3.3 and pray the problem goes away.
klint.
--
Klint Gore
Database Manager
Sheep CRC
A.G.B.U.
University of New England
Armidale NSW 2350
Ph: 02 6773 3789
Fax: 02 6773 3266
EMail: kgore4(at)une(dot)edu(dot)au
From | Date | Subject | |
---|---|---|---|
Next Message | Ron Peterson | 2008-07-04 01:17:50 | Re: LDAP Authentication |
Previous Message | aravind chandu | 2008-07-03 19:14:24 | xml and postgresql |