1 Reply Latest reply: Feb 10, 2012 11:50 AM by Rob Shepler RSS

    Just joined

    Robert Bachmann

      My name is Bob Bachmann, I work at Aetna Inc in Hartford Connecticut. We are in the process of establishing an on premise version of jive with a 4.5 pilot and plans to get to version 5 in teh coming months. Any info on best practices for hosting would be great. I am also interested in discussing establishment of high availability/Fail over kind of  environment and wondering if anyone else has ideas on multi data center strategies.

        • Re: Just joined
          Rob Shepler

          Hey Bob... good luck with your pilot.  And go Huskies!!!  I grew up in South Windsor and will forever be a UConn fan.  :-)

           

          We have a couple JIVE instances hosted at National Instruments today (internal and external community).  In both cases we use a 2 node cluster which has load balancing.  However, as it stands we can handle all traffic on a single node - at least for a small period of time.  This has been very helpful as we can remove one server from cluster for maintenance / troubleshooting / restarts giving minimal user disruption.  At DB level... we are in-process of migrating to Oracle RAC... which gives us some flexibility / redundance at DB layer.

           

          One of our instances is on JIVE 5 and there are several new services running for this version (as compared to 4.5).  Today we are running cache, activity, and doc conversion services on separate VMs.  Exception - We have coupled cache and doc conversion on same server.  Had to play with sizing and doc conversion configs to get them to play nice on same box.