memcached failover

[Paul B. Henson]

On Wed, Aug 03, 2016 at 02:57:54PM -0400, Jorj Bauer wrote:
> With our testing so far, the whole concept is moot - even with 
> replication between memcached nodes, SSO doesn't work across nodes.

Popping into the middle of this, but are you talking about clustering
the idp? Have you looked at Unicon's hazelcast storage backend? We've
been using that for a few months and it's been working beautifully. It
handles replication and node failover perfectly as far as I can tell,
specifically in the case you describe where you take nodes down one at a
time for maintenance...


-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.cpp.edu/~henson/
Operating Systems and Network Analyst  |  henson at cpp.edu
California State Polytechnic University  |  Pomona CA 91768