Why would you select a database server that didnt have master/master?
By only having a single master, a cluster can really only act as load
balancing and not HA because of the single point of failure. If postgres
was maybe a small project and the feature was expected soon, i would
understand, but its been around quite a while, so I wouldnt expect it
any time soon.
I am a bit surprised by the exim over postfix selection as well, though
not as big of a deal.
Any particular reasons you chose those particular two servers?
-Mark
On 2012-10-15 06:54, Andrew Colin Kissa wrote:
On 15 Oct 2012, at 1:42 PM, Mark Chaney wrote:
Thanks for the invite. Im checking it out. Do you have a web server
preference for Baruwa 2? I know there are options, but I find it best
to use whatever the dev is using. Makes things easier to troubleshoot.
Plus I hope he knows/uses what works best for his project. =P
I am using Nginx/Uwsgi because Nginx is asynchronous and
event-driven, and Uwsgi because it allows scaling horizontally
easily.
If master/master isnt available, how are things setup when clustered
from a postgres standpoint? Do we only have the option of doing a
master and a slave?
Yes you can only do master/slave.
--
www.baruwa.org
_______________________________________________
Keep Baruwa FREE - http://pledgie.com/campaigns/12056
_______________________________________________
Keep Baruwa FREE - http://pledgie.com/campaigns/12056