Hi Howard (long time no see!) - We've never been able to make this work. With 4000+ users, and typically 1,000 of them on at any time this has always been an issue.
Typically the first install we have to have all of the servers down and we do the install on the admin server of our server group. Then we take that one down and repeat the process - you're probably familiar with it. For a while the thinking was we could re-direct the load balancer to only the servers that were up, but we just ran into too many issues, like.... * Load balancer re-programming on the fly in production is tricky business * Some patches blew up if ANY of the AR servers were running when they were installed, even on secondary servers * Users really didn't like the idea of "are we up or not?" At the end of the day we end up doing this stuff on the weekend during the low volume times, and breaking it up into manageable sized chunks when we can. I will say this too - we did come up with a few non-BMC sanctioned idea of how to do this. For example, using unix shell scripts it is now possible for me to take a server, install all the upgrades, etc. Then copy all the files to another server, run a shell script, and be 99% of the way to being done. The shell script renames every instance of server name "A" to server name "B". Then there's some manual cleanup to be done. But it does cut down the install time SUBSTANTIALLY. Unfortunately we are not allowed to use this method in most instances, even though it is (IMHO) more reliable AND gives you consistency between installs across servers. We've all had that day where we're installing the third server or whatever and the DBA's decide that's a GREAT time for them to pull the plug. That scripting option won't work if you're on windows obviously... From: Action Request System discussion list(ARSList) [mailto:arslist@ARSLIST.ORG] On Behalf Of Richter, Howard (CEI - Atlanta) Sent: Wednesday, September 10, 2014 8:06 AM To: arslist@ARSLIST.ORG Subject: High Availability ITSM (7.6.4 or greater) ** All, After needing 3 long duration outages to install the pieces needed for a BMC add on product (which will remain nameless), I wanted to see if there was some method for a High Availability ITSM (7.6.4 or greater) system. Our current system is 3 arservers (1 app, 2 user facing), three mid-tiers and one database. So I am looking at future architectural ideas (when we move to 8.x) to put us in a position, whereby we can give our customers the high availability we need and yet install some of these products, that require restarts and other items that add to unavailability. So looking for ideas. Thanks, hbr [http://ucm.coxenterprises.com/cs/groups/public/documents/www_content/mdaw/mday/~edisp/coxenterprisescolor_jpg.jpg] Howard Richter, Remedy Administrator Email = howard.rich...@coxinc.com<mailto:howard.rich...@coxinc.com> ________________________________ No virus found in this message. Checked by AVG - www.avg.com<http://www.avg.com> Version: 2014.0.4765 / Virus Database: 4015/8175 - Release Date: 09/08/14 _ARSlist: "Where the Answers Are" and have been for 20 years_ _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org "Where the Answers Are, and have been for 20 years"
