| From: | "Phoenix Kiula" <phoenix(dot)kiula(at)gmail(dot)com> |
|---|---|
| To: | "Scott Marlowe" <scott(dot)marlowe(at)gmail(dot)com> |
| Cc: | "PG-General Mailing List" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Vacuum Full is *hideously* slow! |
| Date: | 2008-11-15 22:05:40 |
| Message-ID: | e373d31e0811151405m4513e6a0nc2b493d4847d4d54@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Thanks Scott. Could be my I/O as I am on 15k Raptor SATA drives with
RAID 1 only.
Anyway, reindexing happened fast (12 minutes) and things are now humming along.
How long should cluster take on a db that's about 5.5GB in size? Is it
worth doing on a production db? The db is running fast now anyway, so
the only reason I am even thinking about this is because of the
results of a VACUUM ANALYZE, which give me this:
---
INFO: "links": found 5427 removable, 8876599 nonremovable row
versions in 483102 pages
DETAIL: 96 dead row versions cannot be removed yet.
There were 2135991 unused item pointers.
235 pages contain useful free space.
---
My semi-intelligent question: in the output above, that figure of
"483,102" pages. Should this in any way affect my setting for
max_fsm_pages variable in pg.conf?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Enrico Pirozzi | 2008-11-15 22:27:24 | unsigned int type |
| Previous Message | Scott Marlowe | 2008-11-15 21:51:05 | Re: PostgreSQL Windows x64 Build |