Magnus Hagander <magnus(at)hagander(dot)net> writes:
> Yes, especially since we discussed it in Stuttgart. I guess it may
> have been during the party...
I remember we talked about it, I didn't remember a patch had reached the list…
> Yes, if it should go in any of the current binaries, initdb would be
> the reasonable place, not pg_ctl ;)
Well, pg_ctl is able to call initdb for users, but yes.
> That said, if we're going to wrap pg_streamrecv into 9.1 (which I
> think we should), then *that* is where it should go.
They should at least cooperate so that you don't need to setup WAL
archiving explicitly while preparing the standby, if at least possible.
http://2ndQuadrant.fr PostgreSQL : Expertise, Formation et Support
In response to
pgsql-hackers by date
|Next:||From: Peter Eisentraut||Date: 2011-01-02 21:45:29|
|Subject: Re: and it's not a bunny rabbit, either|
|Previous:||From: Simon Riggs||Date: 2011-01-02 21:19:46|
|Subject: Re: Sync Rep Design|