Markus Schiltknecht <markus@bluegap.ch> writes:
> Alvaro Herrera wrote:
>>> For Postgres-R, I'm currently questioning if I shouldn't merge the
>>> replication manager process with the postmaster. Of course, that would
>>> violate the "postmaster does not touch shared memory" constraint.
>>
>> I suggest you don't. Reliability from Postmaster is very important.
> Yes, so? As long as I can't restart the replication manager, but
> operation of the whole DBMS relies on it, I have to take the postmaster
> dows as soon as it detects a crashed replication manager.
No, you're missing the point. If the postmaster goes down there's no
hope of automatic recovery from the situation. If the replication
manager is separate, and it crashes, then the postmaster can kill all
the backends and auto-restart the whole thing. This architecture has
served us very well for years and I think you're making a serious
mistake to change it.
regards, tom lane