Re: Yet another failure mode in pg_upgrade

On Mon, Sep 3, 2012 at 12:16:04PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce(at)momjian(dot)us> writes:
> > I am working on an additional enhancement that also pulls the live
> > cluster's port number from the postmaster.pid file. I am attaching the
> > part of my patch that was modified to add that feature. This allows
> > live checks without requiring any port numbers to be specified. Let me
> > know if you would like me to email you that merged into your patch, if
> > you want it for 9.2.
>
> That seems like a reasonable idea, but how are we going to go about
> merging the various proposed patches here? Do you want me to commit
> what I've got, and then you can rebase this port change on that?

Sure, that is easy. As you can see, it pulls the port from the pid
file, and it just moves a few functions around, and moves the port
number check later in the process. It looks larger than it is.

> As far as 9.2 vs HEAD goes, I think we should try to converge 9.2 and
> HEAD as closely as possible, just for our own sanity in future
> back-patching. Right now I think the only necessary difference is
> the int64-XLogRecPtr changes in HEAD.

Sure, understand.

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

+ It's impossible for everything to be true. +