Tom,
I think I may have found the error but I can't be sure. I compressed the
pg_dump'd backup file and then samba'd it to a Windows 95 machine in order to
burn it to a CD-ROM. I wonder if Windows added extra line feeds here and
there (although I don't see them when I do a head or tail on the file). If
that's the case, then it is my fault.
-Tony
Tom Lane wrote:
> That is distressing, all right ... and it's not a report we've heard
> before. Can you see any pattern to which tables' contents were saved
> and which were not? I'd wonder about peculiar table names, seldom-
> used column data types, and so forth.
>
All of the tables seemed to be the ones marked ***_proc (e.g.
center_out_proc, ellipse_proc, etc.). These all seemed to be at the end of
the pg_dump. So probably somewhere in the pg_dump a table had an extra
character and screwed up the remaining tables from being written (if I am
correctly understanding how pg_dump works).
>
> Did your indexes get recreated from the db.backup file?
Yes. They get created just after the copy commands. Of course, it would be
nice if they were created first and then the data was copied in. My indicies
have unique keys. There have been times with 6.4.2 where for some reason
(despite having a unique index), I have had two rows in an index. This even
happened when I went to pg_dump the table and rebuild it. I was thinking that
if the index was created first and then the data was copied, then this
probably couldn't occur on a rebuild.
>
>
> Is there any chance that the db.backup file got truncated (say, because
> you ran out of disk space during the dump)?
>
No, this partition is 10 Gigs. I have about 1-2 Gigs left even when the
pg_dump finishes.
>
> If you can, it would be nice to see the db.backup file itself, minus
> data so that it's not too big to email. If you could strip the data
> out and just indicate which tables had data and which not, it should
> amount to only a few K of table-creation commands...
>
> regards, tom lane
Again, the text file is over 2 Gig so I can't seem to find an editor that is
big enough to hold it all in memory (I only have a half a gig of RAM). So it
really is just guesswork. Anything you can think of to strip the data from
this big of a file?
-Tony