Re: BUG #17914: walsenders taking up all memory

Hello,

We use Kafka Connect debezium connectors, which translates to streaming =
OFF. We can tell that streaming is off by querying:
select *
from pg_stat_replication_slots;

And seeing that stream_count is 0.

The way we can tell that walsenders are jumping to 30GB and beyond is
checking in RDS --> select instance --> Go to Monitoring --> Select OS
Process list

Attaching an example screenshot below:
[image: image.png]

Thank you for looking into this!

On Wed, May 3, 2023 at 3:30 AM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
wrote:

> Hi,
>
> On Tue, May 2, 2023 at 5:56 AM PG Bug reporting form
> <noreply(at)postgresql(dot)org> wrote:
> >
> > The following bug has been logged on the website:
> >
> > Bug reference: 17914
> > Logged by: Stan S
> > Email address: stashukny(at)gmail(dot)com
> > PostgreSQL version: 14.3
> > Operating system: AWS RDS - (Red Hat 7.3.1-12), 64-bit
> > Description:
> >
> > Seeing the same issue as reported here:
> >
> https://www.postgresql.org/message-id/CAMnUB3oYugXCBLSkih%2BqNsWQPciEwos6g_AMbnz_peNoxfHwyw%40mail.gmail.com
> >
> > In our case, logical_decoding_work_mem is set to 64 MB, yet walsenders
> can
> > each take up 30 GB of memory, crashing our database.
>
> Thank you for reporting this issue. I have two questions:
>
> * Did you use streaming = off/on on the subscriber?
> * How did you confirm the walsender used 30GB of memory?
>
> Regards,
>
> --
> Masahiko Sawada
> Amazon Web Services: https://aws.amazon.com
>