Forum OpenACS Q&A: Postgresql timestamp bug

Collapse
1: Postgresql timestamp bug
Posted by Michael Bluett on 11/17/03 01:38 PM
Redhat have posted fixed RPMs for Postgres 7.3.4 and earlier to fix a buffer overflow bug. I can't see any fixed sources on the Postgresql site. Should we be applying patches, installing rpms or are there bug-fixed sources somewhere?
Collapse
2: Re: Postgresql timestamp bug (response to 1)
Posted by Don Baccus on 11/17/03 06:26 PM
Ask the PG people ... that's your best bet unless Lamar Hunt (an OpenACSer who maintains the RPMs for the PG project as well) happens to see the thread.
Collapse
3: Re: Postgresql timestamp bug (response to 1)
Posted by Michael Bluett on 11/30/03 01:12 PM
The OpenACS head install steps page seems to suggest that OpenACS 4.6.3 has been verified with Posgresql 7.3.4? Have people successfully tried Postgresql 7.3.4 (which there is an RPM for) with OpenACS 4.6.3?
Collapse
4: Re: Postgresql timestamp bug (response to 3)
Posted by Jade Rubick on 12/01/03 07:44 PM
I believe it does work with OpenACS 4.6.3

However, upgrading can be a pain. See the FAQ for why, and how to get around it.

Collapse
5: Re: Postgresql timestamp bug (response to 4)
Posted by Andrei Popov on 12/02/03 11:10 AM
There are a few functions that get a bit upset with timestamp/timestamptz difference, but otherwise nothing much.
Collapse
6: Re: Postgresql timestamp bug (response to 1)
Posted by Lamar Owen on 12/03/03 05:37 PM
Who is Lamar Hunt? 😊

Seriously, the PostgreSQL project is getting ready to release version 7.3.5, which, IIRC, fixes this bug.

You can download the Red Hat errata packages and extract the patch yourself easily enough, then you can apply it yourself.  If there is enough demand to do it, I can extract the patch and post it, but it will be late tonight or tomorrow before I will have time.