Re: Why is citext/regress failing on hamerkop?

On 2024-05-12 Su 01:34, Tom Lane wrote:
> BTW, I've also been wondering why hamerkop has been failing
> isolation-check in the 12 and 13 branches for the last six months
> or so. It is surely unrelated to this issue, and it looks like
> it must be due to some platform change rather than anything we
> committed at the time.

Possibly. It looks like this might be the issue:

+Connection 2 failed: could not initiate GSSAPI security context: Unspecified GSS failure. Minor code may provide more information: Credential cache is empty
+FATAL: sorry, too many clients already

There are several questions here, including:

1. why isn't it failing on later branches?
2. why isn't it failing on drongo (which has more modern compiler and OS)?

I think we'll need the help of the animal owner to dig into the issue.

> I'm not planning on looking into that question myself, but really
> somebody ought to. Or is Windows just as dead as AIX, in terms of
> anybody being willing to put effort into supporting it?
>

Well, this is more or less where I came in back in about 2002 :-) I've
been trying to help support it ever since, mainly motivated by stubborn
persistence than anything else. Still, I agree that the lack of support
for the Windows port from Microsoft over the years has been more than
disappointing.

cheers

andrew

--
Andrew Dunstan
EDB:https://www.enterprisedb.com