Thread

Re: pg_upgrade instructions involving "rsync --size-only" might lead to standby corruption?

Nikolay Samokhvalov <nik@postgres.ai> — 2023-07-10T20:36:39Z

On Fri, Jul 7, 2023 at 6:31 AM Stephen Frost <sfrost@snowman.net> wrote:

> * Nikolay Samokhvalov (nik@postgres.ai) wrote:
> > But this can happen with anyone who follows the procedure from the docs
> as
> > is and doesn't do any additional steps, because in step 9 "Prepare for
> > standby server upgrades":
> >
> > 1) there is no requirement to follow specific order to shut down the
> nodes
> >    - "Streaming replication and log-shipping standby servers can remain
> > running until a later step" should probably be changed to a
> > requirement-like "keep them running"
>
> Agreed that it would be good to clarify that the primary should be shut
> down first, to make sure everything written by the primary has been
> replicated to all of the replicas.
>

Thanks!

Here is a patch to fix the existing procedure description.

I agree with Andrey – without it, we don't have any good way to upgrade
large clusters in short time. Default rsync mode (without "--size-only")
takes a lot of time too, if the load is heavy.

With these adjustments, can "rsync --size-only" remain in the docs as the
*fast* and safe method to upgrade standbys, or there are still some
concerns related to corruption risks?