Re: SSI freezing bug

On 2013-10-07 17:07:16 +0300, Heikki Linnakangas wrote:
> On 07.10.2013 16:58, Andres Freund wrote:
> >On 2013-10-07 06:44:19 -0700, Kevin Grittner wrote:
> >>Patch attached. Any objections to applying that Real Soon Now?
> >>(When, exactly is the deadline to make today's minor release
> >>cut-off?)
> >
> >Maybe it's overly careful, but I personally slightly vote for applying
> >it after the backbranch releases. The current behaviour doesn't have any
> >harsh consequences and mostly reproduceable in artifical scenarios and
> >the logic here is complex enough that we might miss something.
>
> Well, it's fairly harsh if the feature isn't working as advertised.

Well, you need to have a predicate lock on a tuple that's getting frozen
(i.e. hasn't been written to for 50mio+ xids) and you need to have an
update during the time that predicate lock is held. That's not too
likely without explicit VACUUM FREEZEs interleaved there somewhere.

> >A day just doesn't leave much time to noticing any issues.
>
> It's less than ideal, I agree, but it doesn't seem better to me to just
> leave the bug unfixed for another minor release. Even if we sit on this for
> another few months, I don't think we'll get any meaningful amount of extra
> testing on it.

Yea, maybe. Although HEAD does get some testing...

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services