Thanks for your input guys - I had to put this problem on hold for a few days. In case it helps anyone in a similar boat, this problem seems to have been due to a umask of 0027. Changing to 022 seems to have fixed it. Cheers, Simon Kinsella -----Original Message----- From: pgsql-admin-owner@xxxxxxxxxxxxxx [mailto:pgsql-admin-owner@xxxxxxxxxxxxxx] On Behalf Of Tom Lane Sent: Wednesday, January 31, 2007 4:01 AM To: Simon Kinsella Cc: pgsql-admin@xxxxxxxxxxxxxx Subject: Re: [ADMIN] Cannot initdb for 8.2.1 - "..../postgres.bki" does not exist "Simon Kinsella" <simon@xxxxxxxxxxxxxxxxxxxxx> writes: > \u@cheddar> /usr/local/bin/initdb -D /usr/local/pgsql/data -E UNICODE > initdb: file "/usr/local/share/postgresql/postgres.bki" does not exist > This means you have a corrupted installation or identified the wrong > directory with the invocation option -L. > The .../share/postgresql/ files do exist. Looking at the code, this message is emitted if attempting to stat() the file fails, regardless of what the actual errno is. So I'm wondering about permissions problems, specifically inability to search one of the directories on the path. We should adjust initdb to report the actual kernel errno value, instead of presuming that it's "file not found". regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster
