Vivek Khera <vivek at khera.org> writes:
> Well, I'm now convinced that something is wrong.  There's just no way
> in heck it should take 6 days to replicate my large table across two
> machines, even if they are cheap-o IDE disks... :-)

Check to see what activity is going on on the system, and perhaps what
locks are present.

I find it handy to set, in postgresql.conf:
  stats_command_string = true

(You can do this and signal the postmaster with SIGHUP, and expect it
to "do the right thing".)

What you'd expect to see, in pg_stat_activity, is that the system be
running:

   delete * from public.msg_recipients; copy public.msg_recipients
   from stdin;

If you look for the OID number of msg_recipients and of your database,
you can head to $PGDATA/global/[database OID], and expect that

$ ls -lt | head

will have, at the top of the list, either one of the files for that
table, or one of the index files for that table.

It'll be at least marginally reassuring if you see, over a period of a
few seconds, that the files there are growing, indicating that data is
being copied into them.
-- 
"cbbrowne","@","ca.afilias.info"
<http://dev6.int.libertyrms.com/>
Christopher Browne
(416) 673-4124 (land)