Re: signal 11 segfaults with parallel workers

On Mon, Jul 31, 2017 at 6:35 AM, Rick Otten <rottenwindfish(at)gmail(dot)com> wrote:
> Ok, I got a core this time at 23:00 when the database went down.
> Here is the basic backtrace:
>
>
> The query that took it down this time (based on the pid reported in the
> stacktrace) does indeed spin out a parallel plan, but it is a simple query.
> I was surprised to see the multicorn library mentioned in this trace, it has
> nothing to do with the multicorn FDW installed on the system.
>

We load all the libraries in parallel workers which are loaded by the
main backend. This is to ensure that master and worker backends have
exactly the same guc's defined in the worker.

> I've run the query several times in the last few minutes and can't get it to
> generate a core again.
>

Did the query take the parallel plan during execution? The above
symptom shows that it should crash if you run the same query after
restarting the server.

--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com