No problem. One more question, is there a way to find out, without going through a test install, and from release notes etc. for 8.3 if the database needs migration from 8.2 to 8.3 or not. Regards Dhaval On Nov 14, 2007 10:44 AM, Bruce Momjian <bruce@xxxxxxxxxx> wrote: > Dhaval Shah wrote: > > I am on 8.2 production and it will be difficult to upgrade to 8.3. Is > > it possible to backport the "%r" fix from 8.3 to 8.2? > > You need to troll through the CVS archives to find that patch and try to > apply it to 8.2. This feature will not be backpatched because we don't > backpatch features to previous branches. > > --------------------------------------------------------------------------- > > > > > > > Regards > > Dhaval > > > > On Nov 13, 2007 11:26 PM, Simon Riggs <simon@xxxxxxxxxxxxxxx> wrote: > > > On Tue, 2007-11-13 at 00:07 -0500, Greg Smith wrote: > > > > On Mon, 12 Nov 2007, Mason Hale wrote: > > > > > > > > > After the wal segment file is copied by the restore_command script, is > > > > > it safe to delete it from my archive? > > > > > > > > While I believe you can toss them immediately, > > > > > > This is almost never possible. The last WAL file that must be kept > > > should be sufficient to allow recovery to restart from the last > > > restartpoint. So a variable number of WAL files needs to be kept, not 1, > > > not 2 and certainly never 0. > > > > > > pg_standby with 8.2 provides a -k option to allow keeping last N files, > > > whereas 8.3 passes the %r parameter to show the filename of the last > > > file that must be kept. > > > > > > > you should considering > > > > keeping those around for a bit regardless as an additional layer of > > > > disaster recovery resources. I try to avoid deleting them until a new > > > > base backup is made, because if you have the last backup and all the > > > > archived segments it gives you another potential way to rebuild the > > > > database in case of a large disaster damages both the primary and the > > > > secondary. You can never have too many ways to try and recover from such > > > > a situation. > > > > > > Agreed > > > > > > -- > > > Simon Riggs > > > 2ndQuadrant http://www.2ndQuadrant.com > > > > > > > > > ---------------------------(end of broadcast)--------------------------- > > > TIP 9: In versions below 8.0, the planner will ignore your desire to > > > choose an index scan if your joining column's datatypes do not > > > match > > > > > > > > > > > -- > > Dhaval Shah > > > > ---------------------------(end of broadcast)--------------------------- > > TIP 6: explain analyze is your friend > > -- > Bruce Momjian <bruce@xxxxxxxxxx> http://momjian.us > EnterpriseDB http://postgres.enterprisedb.com > > + If your life is a hard drive, Christ can be your backup. + > -- Dhaval Shah ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster