Re: reindexdb & clusterdb broken against pre-7.3 servers

Michael Paquier <michael(at)paquier(dot)xyz> writes:
> On Mon, May 06, 2019 at 10:23:07PM -0400, Tom Lane wrote:
>> The originally-complained-of breakage exists in all active branches,
>> so is it really OK to commit this only in HEAD?

> I did not think that it would be that critical for back-branches, but
> I don't mind going ahead and remove the code there as well. Are there
> any objections with it?

> Also, wouldn't we want instead to apply on back-branches the first
> patch proposed on this thread which fixes the query generation for
> this pre-7.3 related code?

Given that we pushed out the bad code a year ago and nobody's complained,
I think it's safe to assume that no one is using any supported release
with a pre-7.3 server.

It's reasonable to doubt that this is the only problem the affected
applications would have with such a server, too. I don't see a lot
of point in "fixing" this code unless somebody actually tests that.

regards, tom lane