FW: vacuum problems

i have re-named the ssmsact table to ssmsactold, dropped table ssmsact and
associated index, tried vacuuming again and still get the same error. i
then re-created ssmsact table and indexes w/ the SAME NAMES as used
originally, populated it with values from ssmsactold, and checked that all
values were inserted -- they were. but still get the same error message
when vacuuming! help! we need to vacuum!!!

jt

-----Original Message-----
From: JT Kirkpatrick [SMTP:jt-kirkpatrick(at)mpsllc(dot)com]
Sent: Tuesday, December 28, 1999 10:00 AM
To: 'pgsql(dot)admin(at)hub(dot)org'
Subject: vacuum problems

any idea how to correct this problem? system seems to be working fine, but
nightly vacuums give us this error message:

works=> vacuum;
NOTICE: BlowawayRelationBuffers(ssmsact, 1533): block 3016 is referenced
(priva
te 0, last 0, global 1)
pqReadData() -- backend closed the channel unexpectedly.
This probably means the backend terminated abnormally before or
while pr
ocessing the request.
We have lost the connection to the backend, so further processing is
impossible.
Terminating.

any help would be appreciated! postgres 6.4.2, rh linux 5.1 2 w/ kernel
2.2.2

jt kirkpatrick