Thread: Index bloat and the need to REINDEX

Index bloat and the need to REINDEX

From
"Greg Sabino Mullane"
Date:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


Is there a better solution to the "index bloat" problem with regards
to minimizing locking while reducing an ever-growing use of disk
space? In particular, it would be nice if there was a way to force
VACUUM [FULL] to do some of the compression that REINDEX now does.

Right now, we have tables that are updated very frequently, and a
vacuum full just doesn't do much to reduce the disk space, as
the indexes keep growing until a REINDEX is done. Tried on HEAD
as well.

Thanks,
- --
Greg Sabino Mullane greg@endpoint.com  greg@turnstep.com
End Point Corporation
PGP Key: 0x14964AC8 200608151719
http://biglumber.com/x/web?pk=2529DF6AB8F79407E94445B4BC9B906714964AC8
-----BEGIN PGP SIGNATURE-----

iD8DBQFE4k3PvJuQZxSWSsgRAjmHAKCDvKmDe3P7HfCmSwOTtnxpimn2EgCghhSF
bSllNWoYjSU+g3yz23N8jXI=
=0Uxr
-----END PGP SIGNATURE-----




Re: Index bloat and the need to REINDEX

From
Alvaro Herrera
Date:
Greg Sabino Mullane wrote:

> Is there a better solution to the "index bloat" problem with regards
> to minimizing locking while reducing an ever-growing use of disk
> space? In particular, it would be nice if there was a way to force
> VACUUM [FULL] to do some of the compression that REINDEX now does.
> 
> Right now, we have tables that are updated very frequently, and a
> vacuum full just doesn't do much to reduce the disk space, as
> the indexes keep growing until a REINDEX is done. Tried on HEAD
> as well.

Vacuum full doesn't do anything to shrink the indexes -- rather, it's
the other way around, they become more bloated.  Your best bet is to
vacuum (not full) very frequently.

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.