Re: pg_lzcompress patch for 8.3, 8.2 branch

Tom Lane napsal(a):
> Zdenek Kotala <Zdenek(dot)Kotala(at)Sun(dot)COM> writes:
>> Tom Lane napsal(a):
>>> On the other hand, I remain unconvinced that this problem is severe
>>> enough to justify much backporting work. AFAIK we've only seen one
>>> occurence of a problem to date.
>
>> I know about two occurrence. One was reported on -bug
>> (http://archives.postgresql.org/pgsql-bugs/2008-04/msg00206.php)
>> and second was reported from our customer.
>
> I'm still not impressed. Bear in mind that the patch you are so eager
> to backport has received *zero* field testing, which means there's a
> non-negligible risk that there's something wrong with it.

Our customers uses the patch (version 8.2) on 2TB heavy loaded table which
contains text field with average size ~10kB. He have used it for two months
without any problem. I think it is good field testing. It helped him to fix
corrupted data problems without any random crash or downtime.

> Add on the
> non-negligible risk of messing up something associated with back-porting
> the earlier patch, and consider that back-branch minor releases go out
> with no field testing to speak of (there's the build farm but that's
> about it). You have to seriously question whether the risk is worth
> what is surely an extremely marginal stability improvement.

I don't need it to backport to 8.1 and older. Yeah, It was my eager activity.
I'm happy with 8.3 and 8.2 backport.

thanks Zdenek