| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Sebastien Boisvert <sebastienboisvert(at)yahoo(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Problems backing up |
| Date: | 2010-04-07 17:33:19 |
| Message-ID: | 17759.1270661599@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Sebastien Boisvert <sebastienboisvert(at)yahoo(dot)com> writes:
> ----- Original Message ----
>> From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
>>> [ COPY fails to dump a 138MB bytea column ]
>> I wonder whether you are doing anything that exacerbates
>> the memory requirement, for instance by forcing an encoding conversion to
>> something other than the database's server_encoding.
> Our backups are done with the "-F c" (in addition to the normal user/host/port options). As far as I know that shouldn't be
> triggering any type conversions (which is UTF8 all-around).
No, that wouldn't do it. I'm thinking about forcing client_encoding to
be different from the database encoding, via pg_dump's -E switch, or
indirectly via PGCLIENTENCODING or PGOPTIONS or a user-specific
client_encoding setting (there are some other ways to set
client_encoding too). You could double check that by seeing what
client_encoding is specified in pg_dump's output.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Vick Khera | 2010-04-07 19:27:02 | Re: Solid State Drives with PG |
| Previous Message | Pavel Stehule | 2010-04-07 17:12:18 | Re: count with high allocation |