Re: Sync Rep Design

On Tue, Jan 4, 2011 at 2:50 PM, Stefan Kaltenbrunner
<stefan(at)kaltenbrunner(dot)cc> wrote:
> On 01/04/2011 07:51 PM, Simon Riggs wrote:
>>
>> On Tue, 2011-01-04 at 10:28 -0800, Josh Berkus wrote:
>>
>>> The relevant question is: which configuration(s) can we have ready for
>>> the next CommitFest and alpha release?
>>
>> Based upon that series of conversations, I've reworked the design so
>> that there is (currently) only a single standby offering sync rep at any
>> one time. Other standbys can request to be sync standbys but they only
>> become the sync standby if the first one fails. Which was simple to do
>> and bridges the challenges of an exactly identified sync standby and the
>> fragility of too closely specifying the config.
>
> ah cool - like that approach for 9.1!

Yeah, I like that idea too, on first blush. I think we should think
it over and see whether we're committing ourselves to any design
decisions we may later regret - what parameters will we need to add
from that point to get all the configs we ultimately want to support?
But it seems a reasonable starting point, and we can argue about the
rest once we have working code.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company