Andreas Schmitz <a.schmitz@cityweb.de> writes:
> the vacuum seems to be ok. I discovered that the "cluster" kills the
> data for some reason.
Hmm. I'm not sure whether clustering on a GIST index is really useful
(seeing that there's no linear ordering associated with GIST). But it
shouldn't lose data.
I was able to reproduce a problem as follows: run the tsearch regression
test, then do "cluster wowidx on test_txtidx". This appears to lose
one row:
regression=# select count(*) from test_txtidx;
count
-------
502
(1 row)
regression=# cluster wowidx on test_txtidx;
CLUSTER
regression=# select count(*) from test_txtidx;
count
-------
501
(1 row)
Oleg, Teodor, can you investigate this?
regards, tom lane