Re: Hardening pg_upgrade

On Thu, Aug 21, 2014 at 11:43:42AM -0700, David G Johnston wrote:
> Bruce Momjian wrote
> > I had to make an exception for temporary tables because pg_upgrade uses
> > temporary tables to collect schema information. I tried writing the
> > query to use CTEs (second patch), but I would then have to have one
> > query for 8.3, which doesn't support CTEs, and another for 8.4+, plus
> > the CTE query was more complex than I liked. Another idea would be to
> > drop 8.3 support (and remove lots of code to support that), but the
> > recent large increase in the number of people upgrading from 8.4 makes
> > that unattractive. (8.3 did use a different timestamp storage format
> > though.)
>
> Why not tell people on 8.3- that a direct upgrade is not supported but that
> an indirect upgrade to 9.4 or earlier has to be performed first and then
> that can be upgraded to 9.5+ ?

Yes, we could easily do that, and trim down pg_upgrade in the process.
Are people OK with that?

> I'm not clear on how the 8.4 upgrades volume impacts a decision to support
> 8.3- upgrades?

My point is that people aren't doing upgrades just from 9.1 and 9.2, but
often from very old releases. and the end-of-lifed of 8.4 prompted a lot
of people to upgrade. Now, since 8.3 has been end-of-lifed since
February, 2013, we might be able to argue that 8.3 already had a year to
upgrade, so if they now want to upgrade, they have to do it in two
steps.

Anyway, I think we need more opinions on this.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ Everyone has their own god. +