Re: Should we add debug_parallel_query=regress to CI?

Andres Freund <andres@anarazel.de> writes:
> On 2025-03-05 12:29:15 -0500, Tom Lane wrote:
>> I think this does need to be documented somewhere/somehow, just so
>> that people don't waste time focusing on "it's failing on FreeBSD"
>> when the actual cause is some other thing we happened to load
>> onto that task.

> 0002 is a first draft for one way to do this.

Works for me.

> Of course this still requires somebody analyzing a failure to look at
> cirrus.tasks.yml, but I don't know if we can do a whole lot about that?

If it's far away from the actual task-spec code it probably won't
get maintained very well, so putting it here seems OK to me.

            regards, tom lane