On 12/1/2005 6:13 PM, Jim Archer wrote:

> Hi Christopher and thanks very mush.
> 
> I guess what I'm really wondering is how much longer you would expect (and 
> of course I understand that things change) is how much longer 7.4.x would 
> be supported.  I was thinking if 7.3 -> 7.4 was a big change but 7.4 -> 
> newer was not it might be a while longer.  So that's why my question was 
> badly asked...

I would assume that 7.4 will be supported by the Postgres project at 
least for another year. But you won't see me betting money on 2 years.


Jan


> 
> 
> --On Thursday, December 01, 2005 4:27 PM -0500 Christopher Browne 
> <cbbrowne at ca.afilias.info> wrote:
> 
>> Jim Archer wrote:
>>
>>> Hi Christopher...
>>>
>>> --On Wednesday, November 30, 2005 6:16 PM -0500 Christopher Browne
>>> <cbbrowne at ca.afilias.info> wrote:
>>>
>>>>> Will older versions of Postgres continue to be supported by Slony?
>>>>
>>>>
>>>> Until we have compelling reason to break off compatibility.
>>>>
>>>> I'd like to drop 7.3, as its namespace handling and name quoting
>>>> handling is too often painfully different from the later releases.  But
>>>> nothing has been *so* compelling as to make it really worthwhile to
>>>> do so.
>>>
>>>
>>> Thanks for the reply.  The reason I ask is because the packaged
>>> version of Postgres on the currently stable Debian release (Sarge
>>> Debian 1:3.3.5-12) is version 7.4.7.  It is possible to run a newer
>>> Postgres on Sarge of course, but doing so can complicate maintenance
>>> and future upgrads of the operating system.  Debian is pretty popular
>>> due to its stability, which of course often leave it a bit behind the
>>> curve.
>>>
>>> Is supporting all the 7.x series releases difficult, or just 7.3.x?
>>
>> It's somewhat 7.3.x; there were considerable changes between 7.3 and 7.4
>> in terms of:
>>  a) Name space handling
>>  b) Quoting of names
>>
>> Rod Taylor reported on having hacked Slony-I into a form that let him do
>> an upgrade from 7.2 to 7.4; he only kept it working in that form long
>> enough to get the upgrade done, at which point he killed off the "scary
>> hack" and dropped out the 7.2 nodes.  It seems to me that that is a
>> highly appropriate way to treat 7.3, as well; if you're on 7.3, today,
>> you should be planning to upgrade to something *WAY* newer, likely 8.1,
>> and plan to use Slony-I as a way to get off of it.
>>
>> I'll disagree a little bit with Jan in one regard...  It is desirable
>> for Slony-I to support some old versions of PostgreSQL longer than PGDG
>> does in order to give people on those versions an upgrade path.  But to
>> be sure, once there is a compelling reason (and to my mind, it should be
>> one of functionality that we want that isn't supportible in 7.3) to drop
>> 7.3, we will do so.
>>
>> But as I mentioned in the earlier email, there IS an upgrade path, by
>> using the "mature" versions of Slony-I that might not still be
>> supported, but that will still work, to get from a "no longer supported"
>> version of PostgreSQL to one that is supported (both by PGDG and by the
>> Slony-I team).  So I won't feel any IMMENSE guilt whatever is the point
>> in time that Slony-I desupports 7.3...
>>
>>
> 
> 
> 
> 
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at gborg.postgresql.org
> http://gborg.postgresql.org/mailman/listinfo/slony1-general


-- 
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me.                                  #
#================================================== JanWieck at Yahoo.com #