Re: BUG #5157: Hash index not concurrency safe

Jeff Janes <jeff(dot)janes(at)gmail(dot)com> writes:
> On Sun, Nov 1, 2009 at 8:52 AM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> I think we could recover by having the hashgettuple code path
>> re-synchronize by looking for the heap TID it previously returned.

> Can it get pushed to another page (an overflow page)? My quick
> reading of the code suggests it can't get pushed, which makes the fix
> easier.

It can't, which is an important part of the reason why this fix is okay.

> I'll work on a fix for it. But if 8.4.2 is coming out in the next
> couple of weeks and we want the fix to be in it, then we might want
> someone more proficient than me to work on it.

I'm working on it already ... I figure it's my fault for not catching
this while reviewing the patch :-(.

There's no timetable as yet for 8.4.2.

regards, tom lane