On Mon, Mar 7, 2016 at 9:25 PM, Craig Ringer wrote:
> pg_rewind will help a lot there if it proves robust enough -
FWIW, some of my colleagues are doing a lot of QE/QA on a HA solution
based on pg_rewind, and it is proving to be quite stable for the
moment, they are having a hard time breaking it (that's the 9.4
version, though it is intentionally close enough to what is in core,
I'm careful about that).
--
Michael