At about 3:30 EST all of the electronic cash registers at the Wal-Mart on Glenwood Avenue in Raleigh ceased functioning. The system was down for about 30 minutes. I have no idea why the registers crashed except I heard a manager say that the crash was caused by a transaction at one of the registers. I am aware of no evidence of a y2k connection. I've never seen the registers crash at a Wal-Mart before.

You will recall my earlier post about the 860 megawatt Shearon Harris nuclear plant 20 miles south of Raleigh experiencing an automatic emergency shut down a week or so ago reportedly due to worker error.

I'm only reporting these items for any value they may have in eventually helping others see any patterns which may develop. I'm not trying to insinuate that they are y2k related.

-- Puddintame (dit@dot.com), January 24, 1999

Answers

Your posting brings to light an issue that I have been pondering about.

I have been using a credit card with an expiry date of 'OO'. So far wherever I have used it I have had no problems. Does this indicate that all related systems are dealing with Y2K and are compliant.

Also, at the beginning of this year we started receiving medical bills submitted last month and rejected by our insurance carrier as identifying us as not being covered on those dates. We have never had this problem before. We have had consistence coverage with the same company for the past few years. My husband has given the notices to the department that handles all the medical and dental insurance nonsense where he works. They couldn't enlighten him as to why we should have received such notices. They had been receiving the same. We have since discovered that the carrier is not Y2K compliant!! Could it be Y2K related we are wondering.

That leads me to what I have been pondering about. Maybe it's a good thing for a lot of things like this to happen through the year. Businesses can discover where their vunerabilities lie and hopefully deal with them now and be prepared for next year. There again is this too simple and is there a bigger picture that I should be looking at.

-- Carol (usa-uk@email.msn.com), January 24, 1999.

I too hope that a series of "gradual" errors will ramp up public response - if only so people will remain skeptical of reassurring but vague promises from the government. The more who are prepared if only for 3-4 days of trouble, the fewer who will panic at the first 30 minutes of a blackout.

-- Robert A. Cook, PE (Kennesaw, GA) (cook.r@csaatl.com), January 24, 1999.

"I have been using a credit card with an expiry date of 'OO'. So far wherever I have used it I have had no problems. Does this indicate that all related systems are dealing with Y2K and are compliant. "

Not really. It means the immediate expiration date problem has been addressed. This has nothing to do with due dates, paid-on dates, late dates, transaction dates, interest calculations over the transition, and many others.

However, Arnold Trembley has those who read his St. Louis Time Machine reports pretty convinced that Mastercard, at least, has all of this completely tested for all known possible future spike dates.

Then again, the credit card companies are service bureaus for the issuing banks. Mastercard has invited those banks to use their time machine for testing, and the response rate from banks has been dismal. Either they've done all their own testing (hah!) or they're not ready for that stage yet. Stay tuned...

-- Flint (flintc@mindspring.com), January 24, 1999.

The credit card machines at the local CVS Pharmacy were down for a couple of days. I don't know if they are working yet.(Bethesda MD)

-- Mike Lang (webflier@erols.com), January 24, 1999.

Could this be.... We've had our medical insurance with Blue Cross Blue Shield for many years. My youngest son who is 8yrs old born in 1990, has been sick lately. All his medical bill claims have all of a sudden been having porblems. Blue Cross has been saying he's not covered. We've called sent letters, called while at the doctors office to confirm to them he is indeed covered. Finally someone at Blue Cross told us the problem. His birthdate was coming up the year 1919. She said they changed something in their computers a few months ago. Could this be y2k related?....He sure looks good for an old guy.

pain

-- Melin (jonah2101@aol.com), January 24, 1999.

Could this be.... We've had our medical insurance with Blue Cross Blue Shield for many years. My youngest son who is 8yrs old born in 1990, has been sick lately. All his medical bill claims have all of a sudden been having porblems. Blue Cross has been saying he's not covered. We've called sent letters, called while at the doctors office to confirm to them he is indeed covered. Finally someone at Blue Cross told us the problem. His birthdate was coming up the year 1919. She said they changed something in their computers a few months ago. Could this be y2k related?....He sure looks good for an old guy.

pain

-- Melin (jonah2101@aol.com), January 24, 1999.

"About 30 minutes" strongly suggests a routine problem, because it's rather too short for someone to fix a bug on a production system even in the unlikely event that there was a hot-shot programmer on the supermarket's site.

More probebly, the computer went down because of a "random" event, maybe a power glitch that got throught the UPS (if any), and either (1) there was no hot backup system or (2) the hot backup failed. The problem was resolved once the computer system was cold-restarted.

Happened a while ago at a supermarket here, for rather longer. They invented a customer-relations procedure to cope; new customers arriving weren't let in, customers already in the store were allowed to check out by paying what they thought they owed (presumably having to re-shelf everything in their trolleys would cost a lot more).

-- Nigel Arnot (nra@maxwell.ph.kcl.ac.uk), January 25, 1999.

Hi,

I appreciated your neutral and potentially helpful comment about the crashes you have identified. Maybe a pattern is out there to be seen, and I've been interested in what has already happened as I was of the belief that the 1998-1999 roll over may have brought with it problems. Yet, I have heard nothing that ties into the Y2K issue. But I do have an experience to recount. I would be interested to learn of others.

I found myself victim of a Y2K bug when testing a sample of data- logging system that we built (our own design: both hardware and operating system as well as it's application code) for networked Gamma radiation monitoring. We were aware of the Y2K issues way back in 1989. Complacent as we were, we experienced this issue in December 1997. I was about to reply to a local environmental health authority, here in the UK, following an enquiry of our logger's Y2K compliance. I decided to do a last minute test before putting a statement in writing. The data logger entered failure mode when it was running from the day before the last day of the month into the last day just as it was updating its event scheduler to schedule the making of an out going phone call on the following day ie: in the 1st day of 2000. It actually got into an endless loop when it was winding a virtual software clock forward that we had devised in its code to setup future events immediately after current events had executed. The offending event that it was trying to setup was its routine daily event to phone in with its latest days data to transfer it to its host! A silly error in the code; an experienced Unix/C programmer since the seventies had used the year not year+century convention and Bingo! - A millenium bug. I never sent the letter of compliance! - I spent the next year reconvening the programming team to figure a solution which took on board the fact that this network (small: with only 40+ remote gamma monitors) had hardware in remote places and no one was going to pay much the get it compliant. I think we were lucky, we give these remote logger there time each night when they phone the host. Thus we can shift them back 56 years (56 keep Day of week and leap year in sync.) an just add an offset to the time of the data they return of +56 years. With a few other things to consider it looks like we will get on top of this problem. I cannot help asking myself: "who has the privilege of having designed everything in their systems, have control over the internal time clocks of their systems because they phone teir hosteach night?" - We clearly are not the only ones who screwed up. We are also in a very good position as we are not directly commercial therefor we don't need to hide our mistakes in order to survive in a "dog eat dog" environment. It's certainly kept me busy sorting it out! Graham Denman Project coordinator of the Argus Environmental monitoring trust.

-- graham denman (gd@2000ad.org.uk), January 27, 1999.

Graham: any multiple of 28 years works as well, back to 1600. I suspect a great deal of what you report is going on, the geekvine is full of similar stories. I *hope* this means we're weeding out a lot of these bugs while we can still cope with them, without any real external impact (as in your example).

In general, the consensus is that we will be in trouble, but that we are handling it so far. The latest survey statistics I've seen have been variable (depends on who you survey, how you survey, what you ask, how your question is interpreted, don't get me started...) but show about an 80% incidence of these problems respondents are aware of having at least one of (forgive the syntax). Still no external impact. I don't know if this is good news or bad.

-- Flint (flintc@mindspring.com), January 27, 1999.