Re: Incomplete dump?

Tom Lane wrote:

>=?ISO-8859-1?Q?Benno_P=FCtz?= <puetz@mpipsykl.mpg.de> writes:
>
>
>>When trying to dump a database for upgrading to the current PSQL version
>>using pg_dump I observed the following:
>>
>>
>
>Which version of pg_dump were you using, exactly?
>
>
>
>>The process seems to have finished without problems, but the resulting
>>dump file does not end in
>>
>>
>
>
>
>>--
>>-- PostgreSQL database dump complete
>>--
>>
>>
>
>
>
>>but rather with a command line (complete, not truncated as might be the
>>case when running out of disk space, which was plenty anyway)
>>
>>
>
>
>
>>Is this an indication of an incomplete dump? If so how could one proceed?
>>
>>
>
>I don't remember which version of pg_dump started adding that trailer.
>If it's an old copy then maybe you're OK.  If it should have a trailer
>and doesn't then you're right to be suspicious.
>
This may well be the reason, The version in question is 7.4.8 (hence my
wish to upgrade) while my backup script worked with a newer (8.0) DB and
was copied over ...
Does anybody know when the trailer was added?

>Could pg_dump have been
>operating under a file-size ulimit that stopped it early?
>
>            regards, tom lane
>
>
>
>
>
"ulimit -a"reports:
core file size        (blocks, -c) 0
data seg size         (kbytes, -d) unlimited
file size             (blocks, -f) unlimited
max locked memory     (kbytes, -l) 32
max memory size       (kbytes, -m) unlimited
open files                    (-n) 1024
pipe size          (512 bytes, -p) 8
stack size            (kbytes, -s) unlimited
cpu time             (seconds, -t) unlimited
max user processes            (-u) 8192
virtual memory        (kbytes, -v) unlimited

so I don't think this to be the problem.

Thanks
    Benno

--
Benno Pütz
Statistische Genetik
Max-Planck-Institut f. Psychiatrie            Tel.: +49-89-30622-222
Kraepelinstr. 10                              Fax : +49-89-30622-601
80804 München, Germany