>>> Bruce Momjian <bruce(at)momjian(dot)us> wrote:
> Simon Riggs wrote:
>> On Wed, 2009-02-25 at 17:50 -0500, Bruce Momjian wrote:
>> > Simon Riggs wrote:
>> > > On Thu, 2009-02-26 at 06:15 +0900, Fujii Masao wrote:
>> > > >
>> > > > Again, I'm not planning to get rid of any existing
>> > > > capabilities
>> > >
>> > > Good
>> > >
>> > > > unless necessary.
>> > >
>> > > That is not a caveat I will accept, a priori.
>> >
>> > What does "accept" mean above? Are you the sole acceptor for
>> > this feature? That is surprising to me.
>> >
>> > You can say you would vote against it but your wording above
>> > seems overly controlling.
>>
>> I would hope my words carry the same weight as others when people
>> speak of what can and cannot be included, when backed by reasonable
>> and logical technical reasons.
>>
>> Some things are important, some not, and I've done my best to use
>> words that indicate my views on that.
>
> I think you need to work on improving your word choice then.
Anything which breaks existing techniques for log shipping, PITR
recovery, or warm standby would present a significant obstacle to
adoption in our environment. Based on off-list emails received when I
mentioned what we do, I know we're not alone.
Some comments have made me nervous on this count, but responses to
questions about it have generally been reassuring. The "unless
necessary" seemed a little ominous. I won't get too excited unless
someone thinks a change is necessary and I see the nature of the
change.
As long as there is some way to ship base backups and WAL files to
multiple targets, not all of which are running any PostgreSQL
software, and the database can be restored from such copies, we're not
looking a more than a few weeks of staff time to adjust our scripts
and procedures.
-Kevin