Re: recovering from "found xmin ... from before relfrozenxid ..."

On Mon, Jul 13, 2020 at 8:58 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
> > Oh, I hadn't realized that limitation. That would be good to fix. It
> > would be even better, I think, if we could have VACUUM proceed with
> > the rest of vacuuming the table, emitting warnings about each
> > instance, instead of blowing up when it hits the first bad tuple, but
> > I think you may have told me sometime that doing so would be, uh, less
> > than straightforward. We probably should refuse to update
> > relfrozenxid/relminmxid when this is happening, but I *think* it would
> > be better to still proceed with dead tuple cleanup as far as we can,
> > or at least have an option to enable that behavior. I'm not positive
> > about that, but not being able to complete VACUUM at all is a FAR more
> > urgent problem than not being able to freeze, even though in the long
> > run the latter is more severe.
>
> +1 for proceeding in this direction, rather than handing users tools
> that they *will* hurt themselves with.
>
> The more that I think about it, the more I think that the proposed
> functions are tools for wizards only, and so I'm getting hesitant
> about having them in contrib at all.  We lack a better place to
> put them, but that doesn't mean they should be there.

It's not an either/or; it's a both/and. To recover from this problem,
you need to:

1. Be able to tell which tuples are affected.
2. Do something about it.

I think there are a number of strategies that we could pursue around
either of those things, and there are better and worse ways of
accomplishing them, but having one without the other isn't too great.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company