I am a newbie, trying to move a pg db from PostgreSQL 7.13 on RedHat
7.3 to PostgreSQL 7.42 on FreeBSD 4.9. The db has one table with 32,000
rows, each containing a blob. The blobs range in size from 2K to about
300K. The entire backup .tar file is just over 3GB.
I first tried to restore using the folloing command:
pg_restore -d athena backup.tar
It ran for a while (creating all the table defs) and the I got the
pg_restore: ERROR: literal carriage return found in data.
So, I had some bad carriage returns. I found an answer at
(http://forums.devshed.com/archive/t-99865), which suggested using sed
to fix the carriage returns. I did that, creating backup2.tar. The
problem is, when I now run the following command (after recreating the
pg_restore -d athena backup2.tar
I get this reply _immediately_:
pg_restore: [archiver] out of memory
This happens so fast that I'm pretty sure the restore hasn't even
*started*. So, I'm not sure how it has run out of memory. This is on a
P4 2.8 GHz system with a 2GB swap and 1GB RAM.
I can restore all the other tables individually with backups to sql
and imports, but the blob table is the most important table in the db.
I've spent about 8 hours trolling through Google and mailing lists. Any
suggestions as to how I can find out more information on my problem?
Thanks in advance for your help,
PS - If I can't get this working myself, I'd be willing to pay someone
to figure it out. What would be a reasonable amount to expect to pay
for this kind of troubleshooting?
pgsql-novice by date
|Next:||From: Tom Lane||Date: 2004-06-04 15:00:43|
|Subject: Re: problem with pg_restore |
|Previous:||From: joseph speigle||Date: 2004-06-04 03:40:43|
|Subject: Re: Cannot insert a duplicate key into unique index|