Re: [PoC] Federated Authn/z with OAUTHBEARER - Mailing list pgsql-hackers

From Daniel Gustafsson
Subject Re: [PoC] Federated Authn/z with OAUTHBEARER
Date
Msg-id FA245325-A552-4302-BCF6-4AFF494E3B0D@yesql.se
Whole thread Raw
In response to Re: [PoC] Federated Authn/z with OAUTHBEARER  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
> On 21 Feb 2025, at 18:18, Andres Freund <andres@anarazel.de> wrote:

> One question about interruptability. The docs say:
> ....

> Afaict the only real way to do safely across platforms is to never call
> blocking functions, e.g. by using non-blocking sockets and waiting for
> readiness using latches.

Fair point, we'll work on a proposed new wording for this.

> And a second one about supporting !CURL_VERSION_ASYNCHDNS:
>
> Is it a good idea to support that?

We could block building instead of the current warning, but that's at best what
we can do.  I spent some time skimming over package definitions for the major
distributions ans OS's and couldn't find any that use sync dns.

--
Daniel Gustafsson




pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: TAP test started using meson, can get a tcp port already used by another test.
Next
From: Robert Haas
Date:
Subject: Re: explain analyze rows=%.0f