Hints about how to debug pg_dump problem?

Hi,

I'm trying to move a database from a development machine onto a production
machine.  I seem to get good output from my pg_dump command, but when I read
the resulting files, I can't get them to import everything.  (FWIW, I'm
moving from a Debian potato/woody combo onto a fresh Debian potato install
with just PostgreSQL updated to the 7.0.2 deb package.)

What I don't understand is the order in which a dump file is processed.  I
thought it should be processed pretty much in the order it appears, right?
After all, if you do

    $ psql -e my_database < my_dumpfile

you should just get a straight-ahead reading of the dumpfile, no?  Watching
the output, however, things are not getting created even though there are no
complaints from psql (about that data).

For instance, if I have data from table1 which appears in the dump file
before data from table2, and I'm getting a parse error on table2, shouldn't
the data from table1 show up (assuming that nothing in it depends on the
data from table2 being there) after the error?

The read eventually dies with a parse error.  I'm sure there must be
something wrong in my source database that I'm getting this, but I don't
even know where to begin tracking down the problem.  I've dumped and re-read
this database before, but I've added the data that is causing the problem
since then.  Obviously, that's what the problem is; I just can't see how the
file is getting read.  Any hints as to where to start would be much
appreciated.

A

--
Andrew Sullivan                                      Computer Services
<sullivana@bpl.on.ca>                        Burlington Public Library
+1 905 639 3611 x158                                   2331 New Street
                                   Burlington, Ontario, Canada L7R 1J4