10 messages in ru.sysoev.nginxfailover / fault-tolerant configurations
FromSent OnAttachments
alex clemeshaMay 21, 2008 12:22 pm 
J DavisMay 21, 2008 1:24 pm 
CalomelMay 21, 2008 1:34 pm 
alex clemeshaMay 21, 2008 1:41 pm 
J DavisMay 21, 2008 2:28 pm 
Igor ClarkMay 22, 2008 9:35 am 
Pavel GeorgievMay 27, 2008 8:15 am 
Renaud AllardMay 27, 2008 11:58 am 
mikeMay 27, 2008 12:06 pm 
Pavel GeorgievMay 27, 2008 1:49 pm 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:failover / fault-tolerant configurationsActions...
From:Pavel Georgiev (pave@public.gmane.org)
Date:May 27, 2008 8:15:59 am
List:ru.sysoev.nginx

I`m using nginx as a load balances for 2 backend servers that run a php application. I need session awareness so I`ve enabled ip_hash for the upstream, but in case one node fails, all requests that should go to that node get the error_page.

Is there a way to have session awareness without loosing the redundancy that 2 backend servers provide?

If not - it is planned to be implemented soon (as part of ip_hash or another module so that users can have the current functionality as well)? I`m guessing that many people would be happy to have session awareness that redirect requests to another node if the node that should handle the session dies (this will end the session which is better than seeing the error_page).