Actually I think I may be able to answer my own question.  I think my sysAdmin added something to the Linux box and caused the problem. I started to get swap errors with error messages about system drive out of space.  This shouldn't be a problem since there's only 6gigs out of 90+gigs of space on the thing.  I also noticed that if I ran it a few times, I got different numbers in the error message:  could not write to output file (X != 97).

So while the error message wasn't totally apparent, it made sense after I realized the space problem.

>
>
>Patrick Hatcher

>
>"Jim C. Nasby" <jim@nasby.net>
>06/30/2003 08:11 PM EST
>
> To: Patrick Hatcher <PHatcher@macys.com>
> cc: pgsql-general@postgresql.org
> bcc:
> Subject: Re: [GENERAL] Pg_Dumpall error?
>
>
>

>Forgive the stupid question, since I'm completely unfamiliar with the
>logging changes Tom is busy working on, but would it be worthwhile to
>have error messages/asserts include a function/filename where the assert
>failed? (I'm assuming this is an assert failure). If the assert code
>could be easily changed to add that info, it might be helpful to those
>not intimately familiar with the code.
>
>Just a thought.
>
>On Mon, Jun 30, 2003 at 03:03:52PM -0700, Patrick Hatcher wrote:
>> I've done numerous dumps, but I've never seen this type of error before:
>>
>> pg_dumpall > datamart.out
>> pg_dump: [archiver] could not write to output file (36 != 97)
>> pg_dumpall: pg_dump failed on bdc_oz, exiting
>>
>> Any ideas?
>> I'm running Pg 7.3.2
>>
>> TIA
>> Patrick Hatcher
>>
>>
>>
>>
>> ---------------------------(end of broadcast)---------------------------
>> TIP 7: don't forget to increase your free space map settings
>>
>
>--
>Jim C. Nasby, Database Consultant                  jim@nasby.net
>Member: Triangle Fraternity, Sports Car Club of America
>Give your computer some brain candy! www.distributed.net Team #1828
>
>Windows: "Where do you want to go today?"
>Linux: "Where do you want to go tomorrow?"
>FreeBSD: "Are you guys coming, or what?"