Please refer to the diagram in the link below. While looking at my goals.



  1. Get PostgreSQL 9.1 running on 3 terminals on the 'on site' location.
  2. 'on site' have 'streaming replication' running between these DB's 'on-site' locations.
  3. 'on site' have fail-over to one of the 'on-site' 'slaves'.
  4. between 'master' 'on site' and 'slave data warehouse' I'll just pull log files.

I've read a bunch about Linux but haven't found much about Windows XP. All the machines 'on site' have the same hardware. Unfortunately I can't use linux in the 'on site' environment.

Again been googling but not finding much anyone know of a good tutorial(s) to get this type of setup pieced together!

I'm going to keep looking and if I find anything I'll make sure to post it back here.


  • You cannot use streaming replication between different OS.
    – user1822
    Mar 15, 2012 at 17:22
  • @a_horse_with_no_name - CRAP - working on an edit will keep you posted.
    – cph
    Mar 15, 2012 at 17:28
  • @a_horse_with_no_name -- Ok now just streaming replication between XP units.
    – cph
    Mar 15, 2012 at 17:34
  • Should I be putting this in the ServerStack?
    – cph
    Mar 15, 2012 at 17:38
  • 1
    they were good enough for me to get it running on Windows XP. There is not much OS specific stuff in there that can't be easily translated to the corresponding Windows command (it helps having the GNU tools around though)
    – user1822
    Mar 15, 2012 at 18:41

1 Answer 1


This is an old question but figured I would provide some pointers here. The first place to start, naturally, is with the documentation, which is excellent. It will work exactly as written on any supported OS.

This being said, if you are on Windows, some of the specifics of PostgreSQL are sufficiently UNIXy to be confusing. Its thus helpful to remember what streaming replication is in PostgreSQL, which is a continuous disaster recovery which never completes. So in addition to the replication docs you want to read all about point in time recovery and incremental backups.

The key aspects of replication is the recovery.conf which configures the PostgreSQL system to basically continually recover to now() with logs provided by another server. If you think about it from that perspective, the docs and tutorials make more sense. You do need to know where the data directory of Postgresql is however.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.