| From: | Martijn van Oosterhout <kleptog(at)svana(dot)org> |
|---|---|
| To: | "Michael L(dot) Boscia" <mikeboscia(at)gmail(dot)com>, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: pg_dump after transaction id wraparound failure |
| Date: | 2006-05-15 19:01:55 |
| Message-ID: | 20060515190155.GB23811@svana.org |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Mon, May 15, 2006 at 08:48:54PM +0200, Martijn van Oosterhout wrote:
> Vacuum will make anything that disappeared by wraparound in the last
> billion transactions reappear, so a databasewide vacuum should solve
> all your problems, no need to dump...
Also, for future reference. If you have a lot of read-only tables (and
judging by the names of your tables it looks like you might), then
running VACUUM FREEZE on a table will put it in a state where you never
have to vacuum it again (until you do an insert/update).
Put another way, anything in a table at the point you run a VACUUM
FREZE over it is protected from wraparound. Over time all tables will
convert to this state, just an explicit freeze makes it quicker, for a
1.4TB data this might be important.
Have a nice day,
--
Martijn van Oosterhout <kleptog(at)svana(dot)org> http://svana.org/kleptog/
> From each according to his ability. To each according to his ability to litigate.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2006-05-15 19:03:07 | Re: pg_dump after transaction id wraparound failure |
| Previous Message | Scott Marlowe | 2006-05-15 18:53:45 | Re: pg_dump after transaction id wraparound failure |