RE: BUG #17757: Not honoring huge_pages setting during initdb causes DB crash in Kubernetes

The controllers generally always pull in the latest PostgreSQL.
It is easy to get the latest version with PostgreSQL updated.

Unfortunately, getting a bug fix is a lot harder.
One controller currently holding this defect for over a year with no end in sight.


Found this:
https://github.com/opencontainers/runtime-spec/issues/1050

Looks like a PR exists for it but the solution is invalid.
https://github.com/kailun-qin/runtime-spec/commit/a6505339204535150260d8e4f0bc112628f1fa87


More info:
https://www.postgresql.org/message-id/flat/20200218093240.jd3lgoxmisyl2tt5%40localhost#61c2c7fc3d3dd80512c9130b6967be16


It would be nice if "try" worked as expected.
I totally understand it is not a PostgreSQL issue but any assistance would be very appreciated.


Thanks,
David Angel




Internal Use - Confidential

-----Original Message-----
From: Andres Freund <andres@anarazel.de> 
Sent: Monday, January 23, 2023 3:10 PM
To: Sisson, David
Cc: Tom Lane; Christophe Pettus; Tomas Vondra; pgsql-bugs@lists.postgresql.org; Howell, Stephen
Subject: Re: BUG #17757: Not honoring huge_pages setting during initdb causes DB crash in Kubernetes


[EXTERNAL EMAIL] 

Hi,

On 2023-01-23 20:35:17 +0000, Sisson, David wrote:
> A quick and dirty solution could be to alter initdb to catch the exception and retry using a copy of the sample with "huge_pages=false".
> Would that be acceptable?

This is a kubernetes or postgres-operator bug (setting up the wrong cgroup limit, which the docs explicitly warn against doing). I don't think we want to accumulate workarounds like that in postgres.


> Passing in a config setting into initdb would still require a rebuild of all controllers.
> That could take months to years at best.

Huh. I don't know anything about the controller, but that seems problematic independent of this specific issue. And you'd still need to deploy a new version of postgres to get such changes...


> Internal Use - Confidential

Hardly.

Greetings,

Andres Freund