Thank you Barry for a lead-in to my story. In the summer of 1996 there was a regional power failure that cascaded down from Oregon into California and beyond. Both of our data centers went down after UPS was exhausted. We had no generators at the time because--scout's honor--some crazy executive believed that alternative power would cast aspersions on our product; he had blocked all efforts to install backup power. Shortly thereafter we got generators at both data centers. They were tested regularly, even putting the sites on generators on a regular basis to make sure they worked.
Then came the summer of 2008. There was an environmental failure in one of the two utility feeds into one data center. Incredibly we started having trouble with the other feed, this time something in the internal building distribution. At some point one late afternoon, the remaining utility feed went offline. The generator would not start. After ten years the control module had cr*pped out. This was not a giant knife switch we just grab and throw. It was a computer system whose software was hosed up. With literally seconds of UPS remaining, a heroic electrical engineer who happened to be on site yanked a control module out of one rack and shoved it in another. The lights came back on. In neither case, the real outage nor the near miss, did I worry about finding the IODF on re-IPL because every one of our seven plexes has its own copy of SYS1.IODFxx on its own IPLPARM volume. It's only 12 cylinders. We can migrate a new IODF to every plex in the enterprise via batch in 10 minutes. No concern about how to locate and enter tapes into a silo at a more or less dark data center. Why put the family jewels at risk when their TLC is so cheap? . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile [email protected] From: Barry Merrill <[email protected]> To: [email protected], Date: 06/16/2014 01:05 PM Subject: Re: Dataset in Two master catalog Sent by: IBM Mainframe Discussion List <[email protected]> John Deere, in the 70s had two separate power companies provide their "non-interruptible power" for their mainframes, but the DP manager was not able to also add a diesel powered backup because the only vendor with the sufficient engine to meet their power needs used ONLY a Caterpillar diesel engine. After the fourth failure in a couple of years, the DP manager finally won approval from senior company management to install a diesel powered backup, but ONLY because it was housed in a lockee Deere-shed so no one would see there was a Yellow diesel engine inside. Or so I was told. Barry Herbert W. "Barry" Merrill, PhD President-Programmer MXG Software Merrill Consultants 10717 Cromwell Drive Dallas, TX 75229 [email protected] http://www.mxg.com - FAQ has Most Answers [email protected] - invoices/PO/Payment [email protected] - technical tel: 214 351 1966 - expect slow reply, use email fax: 214 350 3694 - prefer email, still works -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Skip Robinson Sent: Monday, June 16, 2014 1:44 PM To: [email protected] Subject: Re: Dataset in Two master catalog I can think of several scenarios, but how about one that we can all relate to--and I'm the electric company: a data center power failure. Sure, what are the odds of a power failure requiring an enterprise-wide IPL *and* missing your one and only IODF file? Anyone? Dr. Murphy? Anyone? . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile [email protected] From: "R.S." <[email protected]> To: [email protected], Date: 06/16/2014 11:03 AM Subject: Re: Dataset in Two master catalog Sent by: IBM Mainframe Discussion List <[email protected]> W dniu 2014-06-16 17:54, Skip Robinson pisze: > Blessed are the pessimists, for they shall take backups. ;-) > > OTOH nothing scares me more than the prospect of depending on stand-alone > restore to IPL the enterprise. For me it's a matter of training & expertise, but ...WHY STANDALONE? We talk about single file, IODF. Very important during IPL, but quite inactive during normal system run (with exception to HCD activities including ACTIVATE processing). Can you enlight us and tell us the prospect of such failure? Regards -- Radoslaw Skorupka Lodz, Poland ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
