http://www.drbd.org/ is something you will likely be interested in.

You build your two boxen, and layer it up to the drbd level:

- real disks
- raidx (optional)
- lvm (optional)
- file system of choice
- drbd

Next use the HA project's heartbeat and stonith for ip take-over in the event 
of a missed heartbeat from a service you are monitoring on the active box.

You still need to sync off-site backups of course, but this handles the high 
availability part with no latency worries (per transaction commits) and 
guaranteed consistency. It's done at the block level and you'd have it in 
synchronous mode using local boxes. I don't know anything else that can do that.

I've never done the drbd part yet, but I've just ordered two more boxes with 
this in mind. I first looked into HA several years ago, but never needed it 
until now.

------------------------
Regards, Paul Evans




-------------------- m2f --------------------

--
http://www.openerp.com/forum/viewtopic.php?p=23925#23925

-------------------- m2f --------------------


_______________________________________________
Tinyerp-users mailing list
http://tiny.be/mailman/listinfo/tinyerp-users

Reply via email to