Re: Missed check for too-many-children in bgworker spawning

Hi,

On 2019-10-09 12:29:18 -0400, Robert Haas wrote:
> I would say rather that if fork() is failing on your system, you have
> a not very stable system.

I don't think that's really true, fwiw. It's often a good idea to turn
on strict memory overcommit accounting, and with that set, it's actually
fairly common to see fork() fail with ENOMEM, even if there's
practically a reasonable amount of resources. Especially with larger
shared buffers and without huge pages, the amount of memory needed for a
postmaster child in the worst case is not insubstantial.

> The fact that parallel query is going to fail is sad, but not as sad
> as the fact that connecting to the database is also going to fail, and
> that logging into the system to try to fix the problem may well fail
> as well.

Well, but parallel query also has to the potential to much more quickly
lead to a lot of new backends being started than you'd get new
connections on an analytics DB.

> Code that tries to make parallel query cope with this situation
> without an error wouldn't often be tested, so it might be buggy, and
> it wouldn't necessarily be a benefit if it did work. I expect many
> people would rather have the query fail and free up slots in the
> system process table than consume precisely all of them and then try
> to execute the query at a slower-than-expected rate.

I concede that you have a point here.

Greetings,

Andres Freund