Re: R: How robust is postgre in case of power failure?
От | Robert Treat |
---|---|
Тема | Re: R: How robust is postgre in case of power failure? |
Дата | |
Msg-id | 1026252689.8096.53.camel@camel обсуждение исходный текст |
Ответ на | Re: R: How robust is postgre in case of power failure? (Naomi Walker <nwalker@eldocomp.com>) |
Список | pgsql-admin |
Yes, that's the idea, so that even in an abrupt power failure you should be protected against data loss. It wont save any transactions in progress and don't forget that abrupt power loss could corrupt your file system itself, so a ups would still be nice to have for a more graceful shutdown. If you feel the need for more detail, dig around the mailing list archives searching for 'fsync'. Robert Treat On Tue, 2002-07-09 at 12:47, Naomi Walker wrote: > At 05:23 PM 7/9/02 +0200, Stefano Coletta wrote: > >I had problems when a power failure occurred on old postgres > >installations. I had to use pg_check to fix problems. > > > >Now I have installed an UPS just to make a clean shutdown of the system. I > >strongly suggest you to purchase an UPS, just to make a clean shutdown. > > I agree a UPS/generators are good things, but don't the WAL's protect the > integrity of the data? > > ---------------------------(end of broadcast)--------------------------- > TIP 2: you can get off all lists at once with the unregister command > (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
В списке pgsql-admin по дате отправления: