[chef] RE: Re: RE: Re: postgres master/slave replication and failover to hot standby


Chronological Thread 
  • From: David Leaver < >
  • To: " " < >
  • Subject: [chef] RE: Re: RE: Re: postgres master/slave replication and failover to hot standby
  • Date: Mon, 23 Jan 2012 22:41:27 +0000
  • Accept-language: en-NZ, en-US

The native postgresql 9 hot standby / streaming replication takes the place 
of pgpool.
http://wiki.postgresql.org/wiki/Replication,_Clustering,_and_Connection_Pooling#Features_in_the_Core_of_PostgreSQL

Repmgr has commands for doing the failover part and adding new nodes as 
slaves (or setting up a previous master to be a slave etc).
This should be much nicer than doing it manually. (I haven't used repmgr yet, 
setting it up is on my extended todo list).

-----Original Message-----
From: Tim Uckun 
[mailto:
 
Sent: Tuesday, 24 January 2012 11:05 a.m.
To: 

Subject: [chef] Re: RE: Re: postgres master/slave replication and failover to 
hot standby

On Tue, Jan 24, 2012 at 8:38 AM, David Leaver 
< >
 wrote:
> You could take a look at repmgr: 
> https://github.com/greg2ndQuadrant/repmgr


Does this project take the place of pgpool or are you supposed to use in 
conjunction with something like that?



Archive powered by MHonArc 2.6.16.

§