Re: Postgres v16.4 crashes on segfault when memory >= 16gb

Thank you for the reply. I will try with jit turned off and report back but
also try to collect stack traces from the pods.

On Wed, Aug 21, 2024 at 4:31 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Weslley Braga <wbraga(at)implentio(dot)com> writes:
> > I've been observing for the past month. I'm running PostgreSQL in
> > Kubernetes through the CloudNativePG project. When I bump the pods
> resource
> > configuration to anything above 16gb I notice that complex queries such
> as
> > the one shared below cause the runtime to fail with a segfault see logs
> for
> > details. The same query with a configuration < 16gb memory works ok.
>
> Hmm, does turning jit = off affect it?
>
> > I'm glad to share more information to further help identify the issue.
> I've
> > also extracted core dumps from the error.
>
> The core dumps will be useless to anyone on any other platform or with
> an even slightly different build of Postgres. If you could extract
> stack traces from them, that might be helpful:
>
>
> https://wiki.postgresql.org/wiki/Generating_a_stack_trace_of_a_PostgreSQL_backend
>
> regards, tom lane
>