Re: [HACKERS] pg_upgrade incorrectly equates pg_default and database tablespace

2012-04-10 Thread Bruce Momjian
On Thu, Mar 22, 2012 at 02:55:32PM +0200, Ants Aasma wrote: > Hi, > > while working on a support case I stumbled upon a bug in pg_upgrade. > Upgrade fails with "No such file or directory" when a database is > moved to a non-default tablespace and contains a table that is moved > to pg_default. The

Re: [HACKERS] pg_upgrade incorrectly equates pg_default and database tablespace

2012-04-06 Thread Jeff Davis
On Fri, 2012-03-30 at 13:11 -0700, Jeff Davis wrote: > I confirmed this bug upgrading 9.1 to master, and that this patch fixes > it. Thank you for the report! > > Patch looks good to me as well, with one very minor nitpick: the added > comment is missing an apostrophe. > > Bruce, can you take a l

Re: [HACKERS] pg_upgrade incorrectly equates pg_default and database tablespace

2012-03-30 Thread Jeff Davis
On Thu, 2012-03-22 at 14:55 +0200, Ants Aasma wrote: > Hi, > > while working on a support case I stumbled upon a bug in pg_upgrade. > Upgrade fails with "No such file or directory" when a database is > moved to a non-default tablespace and contains a table that is moved > to pg_default. The cause

[HACKERS] pg_upgrade incorrectly equates pg_default and database tablespace

2012-03-25 Thread Ants Aasma
Hi, while working on a support case I stumbled upon a bug in pg_upgrade. Upgrade fails with "No such file or directory" when a database is moved to a non-default tablespace and contains a table that is moved to pg_default. The cause seems to be that the following test incorrectly equates empty spc