On 23 jun 2009, at 12.14, "Marc G. Fournier" <scrappy@hub.org> wrote:
> On Tue, 23 Jun 2009, Magnus Hagander wrote:
>
>> Marc G. Fournier wrote:
>>> On Tue, 16 Jun 2009, Dave Page wrote:
>>>
>>>>> Ultimately it seems the failure to fix this problem is just
>>>>> because
>>>>> Marc is the only one able to do it, and he's not around enough.
>>>>
>>>> No, the real problem is having our mail infrastructure tied into
>>>> hub.org as a number of us have said before.
>>>
>>> Now that I have those two new servers in place, if someone wants to
>>> setup spamassassin on that VPS itself to run instead of going
>>> through
>>> the global spamassassin daemon, I'm cool with that ... I haven't
>>> moved
>>> mail over to the new servers yet, but that just takes 15-20
>>> minutes of
>>> downtime to make happen ... could do that tomorrow evening ...
>>
>> Just to be absolutely clear on what you're saying here.
>>
>> Are you saying that you are now OK with de-coupling the
>> postgresql.org
>> mail from hub.org making it a community managed service?
>>
>> Or are you saying that you're ok with running a *second* instance of
>> spamassassin, and still keep mail going through hub.org relays?
>
> I'm okay with a dedicated instance of spamassassin running in mail.postgresql.org
> so that it can be more tailored / configured to mark spam going
> through to the moderators ...
So you're still not accepting that the mail services are a community
service and resource. Ok, sorry to hear that, but good to know.
In that case, the the statement by Dave still stands - this does not
fix the actual issue. It does paint over one crack a bit, but it does
nothing to fix the structural problem.
/Magnus
