Ühel kenal päeval, K, 2006-06-07 kell 17:45, kirjutas Jim C. Nasby:
> Plus, if the only issue here is in fact the long-running transaction for
> vacuum, there's other ways to address that which would be a lot less
> intrusive than doing something like going to 64 bit XIDs. IIRC, in 8.2
> vacuum will start a new transaction every time it fills up
> maintenance_work_mem, so just setting that low could solve the problem
> (at the expense of a heck of a lot of extra IO).
If the aim is to *only* avoid transaction wraparound, then maybe we
could introduce VACUUM FREEZE ONLY; which never removes any old tuples,
but instead just marks them by setting xmin=xmax for them, in addition
to its freezing of live-and-visible-to-all tuples.
This would avoid touching indexes at all and may well be what is desired
for tables with only very little updates/deletes.
--
----------------
Hannu Krosing
Database Architect
Skype Technologies OÜ
Akadeemia tee 21 F, Tallinn, 12618, Estonia
Skype me: callto:hkrosing
Get Skype for free: http://www.skype.com