On Tue, Nov 16, 2010 at 10:25:13PM -0500, Tom Lane wrote:
> 4. The last bit of discussion on -hackers concerned what to do in
> the case where the server got shut down cleanly. If it was shut
> down cleanly, then any data for unlogged tables would have been
> written out from shared buffers ... but did the data make it to disk?
> There's no easy way to know that. In the event of an OS crash or
> power failure shortly after server shutdown, it's possible that
> the unlogged tables would be corrupt.
Aaah, indeed.
> So Robert's initial proposal
> includes truncating unlogged tables at any database startup, even
> if the previous shutdown was clean.
Sounds reasonable.
> Some (including me) are arguing
> that that is unnecessarily strict; but you do have to realize that
> you're taking some risk with data validity
Don't. We've always liked PostgreSQL for that. Or at least
let us point the gun at our feet ourselves (such as with
fsync).
Karsten
--
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346