Re: BUG #5151: autovacuum process segfaults when max_fsm_pages are too low

[Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>]

Boris Folgmann wrote:
> Currently I'm using postgresql-8.1.15-1.el4s1.1 from the centosplus
> repository for CentOS 4.8. All updates applied, kernel is
> 2.6.9-89.0.11.ELsmp.
> I know that 8.1.18 is out, but haven't found something in the release-notes
> about problems with max_fsm_pages. Anyway, there isn't any more recent RPM
> available at the moment.

I don't think this is related to fsm, actually. Seems like something in
autovacuum, and there was this fix in 8.1.16 (from the release notes):

      Fix crash in autovacuum (Alvaro)

      The crash occurs only after vacuuming a whole database for
anti-transaction-wraparound purposes, which means that it occurs
infrequently and is hard to track down.

So I would suggest upgrading. If that doesn't help, and you can
reproduce the issue, a backtrace from a core dump would be nice.

-- 
  Heikki Linnakangas
  EnterpriseDB   http://www.enterprisedb.com