STL Time Machine Report #42 - Friday 02 April 1999 (1999-04-02) Last weekend was work time for me, watching a couple of Mainframe boot-ups to verify that some application changes took effect and worked properly. Three of my production CICS 4.1 regions were upgraded to LE/MVS runtime libraries, leaving eight still to be upgraded when the third-party database gets LE. The IPL's included some hardware upgrades for faster engines. Although we're in very good shape, we're not ready yet. The Database LE upgrade is a couple of weeks behind schedule. April will be a very busy month for me. Our mission-critical applications cannot be changed after May 1, without a note from God. The actual cutoff appears to be Monday, April 5, for the additional Y2k paperwork. Of course, some non-Y2k enhancements are already scheduled for after May 1, but they have a note from God! So we're going to be very busy on Monday installing almost the last changes we will be allowed to make in 1999. We will still be doing some new development, but that cannot be installed until after rollover. We started our Time Machine testing in January, 1998, so we will have been testing for two full years when the rollover arrives. MAINFRAME ROLLOVER CONSIDERATIONS There was an interesting thread recently on bit.listserv.ibm-main about rollover plans. Several IBM mainframe shops are planning to either power down their big iron, or quiesce all applications, which means stopping all active jobs. There are a couple of good reasons why they want to do this. In some cases, they're worried about unreliable electricity, so they want to completely power down. The other reason why some shops are going the quiesce route, one that I hadn't considered, is that large databases like DB2 have automatic journaling and recovery systems. They don't want to take an unexpected outage and have an automatic recovery run against a journal file that has transactions time stamped on both sides of the rollover. That's a pretty good reason. It's not easy to test automatic recovery in that environment. Just something else to think about... WHERE ARE THE FAILURES? We still haven't seen a big Y2k failure take down a fortune 500 corporation. I don't think we will, and here's why. No company is going to publicize a problem as long as they believe they have it under control. If a batch job blows sky-high and you can somehow get it to run two days later, the general public will never know that somebody had to put in some overtime. What I think may be more likely is Cory's description of the Oxford Health problem. The unprepared company struggles along, but is eventually overwhelmed by financial problems caused by systems that have gone nuts. If this scenario plays out, the big failures might not show up until late in 2000 or even 2001. Still, I won't be flying on New Year's Eve. It will be a regular work day for me. Hope I get a plush hotel room. MISCELLANEOUS I bought a new PC a couple of weeks ago, so now I have enough memory to run Java easily. I visited Cory's Chartroom last Tuesday, and ran into several c.s.y2k regulars who haven't posted recently. I recommend it: http://www.kiyoinc.com/y2kchat.html 9:00 PM to 11:00 PM EST. The PHM is stuck with a warehouse full of brand-new time clocks that won't work with Windows NT. Ooops, how do I expense this? There's still time to visit http://www.DefendYourPrivacy.com/ Finally got one of those new funny quarters for Pennsylvania. We've had one goose fatality on the parking lot, and it looks like the nest may have been raided by coyotes. I didn't think we had any coyotes around here. The trees are budding, spring is coming, and time is running short. As of 1999-04-02, my countdown now reads: 7 days until 1999-04-09 (IBM Julian date 99-099) 94 days until 1999-07-05 (First Fail Date) 160 days until 1999-09-09 (9/9/99, another date) 274 days until 2000-01-01 (Rollover) Previous Year 2000 Time Machine Reports are available at: http://home.att.net/~arnold.trembley/tmr.htm STANDARD DISCLAIMER: I am NOT an official corporate spokesperson. My opinions should not be held against my benevolent employer. -- Arnold Trembley http://home.att.net/~arnold.trembley/ "Y2K? Because Centuries Happen!"