On Tue, Feb 26, 2008 at 10:20 PM, Phillip Smith
> > > Do we think this is a Postgres problem, a Linux problem or a problem
> > > specific to my hardware setup? Was I wrong to think that I should be
> > > able to stream directly from pg_dump to /dev/st0? I would have
> > > thought it *should* work, but maybe I was wrong in the first place
> > > with that?
> > If you can dd the file onto your tape drive, then it's some weird
> interaction between
> > pg_dump and your system I'd think. Could the be some maximum size that
> you can buffer
> > through pipes / redirects on your machine?
> > I'd test to see if cat pgdumpfile.sql > /dev/st0 works or not. If it
> fails at the
> > same approximate size, then it's something to do with redirection. If >
> tar works but
> > redirects fail, then the problem isn't with postgresql. i.e. do something
> similar to
> > what you're doing with pgsql and see which things fail and which ones
> It appears to be me :(
> [postgres(at)stirling ~]$ cat /tmp/dbname080225.sql > /dev/st0
> cat: write error: Invalid argument
> It ran for a good 30 minutes, then died with that.
I wonder what it's meaning by invalid arg? Is something in the .sql
file somehow coming across as an argument? Can you cat the sql file
to /dev/null successfully?
In response to
pgsql-admin by date
|Next:||From: Tom Lane||Date: 2008-02-27 05:34:37|
|Subject: Re: How to reindex when unable to open relation? |
|Previous:||From: Darren Reed||Date: 2008-02-27 05:21:23|
|Subject: Re: How to reindex when unable to open relation?|