We had worked out "In Theory" how we would do it. About 18 months after we opened our second data center, Our Main Data Center needed to shutdown over a long weekend, for 100% shutdown power maintenance.
After the Online's were shutdown Friday Night, We waited for all the DASD & Tape to sync to the "other" site, and shut down the site. (100% Powerdown). We then did our "Plan" (Mostly the DR "real" Plan) and brought Powered Down Center up in the sister site, and ran the normal Batch Production workload. Saturday Morning the Online's came up (2hr later then normal), we ran until Monday afternoon, when we then shut down the Online's (early, Was a holiday and was not a work day but there is limited activity even on S/S/H. Then Ran the Monday Night Batch back in the original DataCenter. We were prepared to keep production running in one location, if there was a issue with the power upgrades until the next weekend (or they fixed it). Our Active - Active Servers switched over without any issues, as they were suppose to. (Except the Voip-PBX) The only issue we noted is the VOIP Did not transfer the active calls to its "sister" VOIP/PBX it hung up on the few calls that were in progress, but all were able to redial the calls and connect thru its Sister/PBX. Few months later we tested the sister site fail over (Test/Dev) to the main site and ran for the weekend, "Just for the Fun of it" In 2012 Hurricane Sandy took out the Powerlines to the Main DataCenter, The DC was running on the Generators but facilities management was not able to get more fuel, and we were in danger of running out of fuel, and having to move the work load to the sister site, We started to make plans for "When" to do the switch, But we were able to get fuel and did not have to do the movement of the workload. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
