STL Time Machine Report #39 - Monday 15 February 1999 (1999-02-15) We had another overnight problem, file not transmitted due to a network computer down, and the wrong beeper number is still on the list. Since we're supporting the Time Machine just like production, the problem gets escalated after 30 minutes. This problem was environmental, but it would be nice to get that beeper list updated. The users seem to be getting a bit nervous with the upcoming freeze on non-Y2k application upgrades. It seems like there's still plenty of non-Y2k work, but it may get pushed into the year 2000. UPGRADES It looks like the upgrade from MVS/ESA 5.2.2 to OS/390 2.5 will happen in Late March, for the primary data center. The backup data center is already on OS/390. The third-party database should be upgraded to an LE/MVS compliant release before then. The CICS regions should be on LE/MVS by late March or shortly thereafter. And various Unix Y2k hardware and OS upgrades are scheduled through July, 1999. After July, we will be working on post-1999 projects. SIGNS? I'm hearing stories of date failures. None of them are serious, yet. They're about like the recent story in the St. Louis Post-Dispatch "Test for Y2K problem causes errors on monthly sewer bills" (Saturday, 1999-02-13, page 6, front section). In some cases the problems are noticeable by the general public, but nothing that can't be handled by "fix on failure". And don't kid yourselves, every business computer system runs in F-O-F mode, all the time. I haven't seen anything in the local news about any problems with Missouri Unemployment systems. That doesn't mean there aren't any, but it's not the end of the world yet. So what's going to happen? What if the power stays on, the phones mostly work, the government mostly works (if anyone is paying attention to it), and the economy mostly holds together? Can we declare victory? Bradley Sherman's predictions page stops at 1999-12-31, last I looked. So what's over there on the other side? We haven't talked enough about trailing event failures, the first monthly processing after rollover, the first quarterly, the first yearly job. If usenet stays up, I expect c.s.y2k to be debating the real effects well into 2000, even after skipping the 30th of February. IT'S A DILBERT WORLD What do you do when you see your boss interviewed in a trade publication, and he explains his IT strategy is to downsize the entire staff? Bring the contractors in when you need a project done, and show them out the back door when it's finished. The mainframe requires a permanent staff, so get rid of the mainframe. Those LAN's can run themselves. We don't need applications programmers when we have Oracle, Peoplesoft, and SAP. We don't need a technical support staff if we have Unix, NT, Winframe, and Client/Server. STUPID COBOL TRICKS I missed my chance to give an example of weird code. Here's one that might look familiar (are you listening, DD?): IF 540-DATE-DDD < 365 ADD 1 TO 540-DATE-YYDDD ELSE IF 540-DATE-DDD > 365 ADD 635 TO 540-DATE-YYDDD ELSE IF 88-NORMAL-YEAR ADD 636 TO 540-DATE-YYDDD ELSE ADD 1 TO 540-DATE-YYDDD END-IF END-IF. Now, if I didn't screw it up in the anonymizing process, this code works and is Y2k-compliant, providing the leap year rules are correct. But isn't this why we wrote standardized date routines in the first place? As of 1999-02-02, my countdown now reads: 53 days until 1999-04-09 (IBM Julian date 99-099) 206 days until 1999-09-09 (9/9/99) 320 days until 2000-01-01 (Rollover) Previous Year 2000 Time Machine Reports are available at: http://home.att.net/~arnold.trembley/tmr.htm I'll be adding some more links to my web pages soon. 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!"