failover

Windows 3.1 is dead - so what?

Today, John Sheesley amused me with his attempt to use Windows 3.1 (actually 3.11 Windows for Workgroups) as a workstation of 2008. Those of us who remembers WFW, may wonder why on earth he would do that for, but I leave it to that.

One should think that using a software that was designed only 15 - 20 years ago should be quite possible today. After all, a PC is still a PC, right?

Not so. A PC of today (2008) is based upon the same principles as back in the early 1990's, but the technology has reach a maturity point where backwards compatibility hardly exists.

Yes, you may get WfW to actually work, if you like many others keep an old 486, or even a P1/P2 in a closet nearby. The challenge will be to render it useful, as John realises:

"The latest and greatest Web browser I could find for Windows 3.x was Internet Explorer 5.01. This was suprisingly modern — I thought IE 4.01 was the last version for Windows 3.x. Alas, it’s not modern enough. After installing this browser, it quickly became apparent that you can go almost nowhere with IE 5.01."

And IE 5 is not 15 years ago - it is more like 5.

Now, why should you, a security minded reader, care about the fact that John failed this project?

Several reasons comes to mind:
  • From a continuency point of view - if you rely on some old hard-/software, make sure you have the tools and systems available in case of a failure.
It is not enough to dust off the old software boxes, you need to set up a system in parallel and see that you have all the bits it takes. I even suggest you make this a routine thing - once every year, month or week - depending on the criticallity. I also suggest you start plan to change the old system for something a bit more modern.
  • From a data storage point of view - if a software can become totally obsolete in less than 5 years, you need to make very sure that your backup systems; long-time storage; and other data you may require access to in the future; uses technology you are able to access.
If you have a bundle of old tapes from a streamer that got replaced some time back - will you be able to access that data? Also consider the long-time effect light, magnetizm and dust has on equipment. CDs and DVDs are not safe for more than 10 years storing - but even that is no guarrantee. (Opposed to the advertisments in the midle of the 1990's, claiming CDs to be the best long-term datastorage available...could last for a houndred years, they claimed...)
  • From a compliance point of view - if you are obligued to store information for a periode of time - 3, 5, 7 or 10 years - you are also obligued to be able to access the same data in the future. It is your responsibility, and it is usually a good idea to plan the technology at the same time you plan what and where to store your data.
I find one of the comments to the article particularly nice:

"What will happen in the next 5 years? Is XP, 2000 going to be out of date also? Will we be able to keep up with the changes that are coming? How can we keep up to speed? "


Yes, XP, 2000 and many others will be out of date. 2000 actually did this summer, when MS pulled the support for it. Anyone remember OS/2? DOS I guess most still remember, but only hardcore, old-guys still uses it to some extent. The world (and the technology with it) moves on.

To keep up with changes, you need to keep track on what is going on, as well as on your own requirements. You are required to update once in a while, but you do not need to jump on the latest versions of everything - unless you have special requirements. It is simply a matter of balancing your needs.



NSB halts all trains due to fire in signaling system

NSB - the Norwegian State Railway company - somehow managed to run their signaling central without redundancy.

So - why care?

The signaling central is where all track signaling and train signaling is collected, analyzed and controlled. I.e. where the red and green lights are managed. You know, the lights that tell the train driver if he can move or halt his train. In order to avoid crashing.

So you would think that the signaling central is a key element to keep the trains running securely. You might even think that the signaling central is business critical. And I would agree fully.

But NSB do not agree. Or their security people lack fantasy enough to imagine the signaling central to fail. Or - the money people decided that the costs of setting up a redundant solution is way too high compared to the risk.

We will certainly know during the next few weeks.

Because last night, the main communication cable systems caught fire, and the signaling central lost communication. At the current moment, we do not know if this was an accident, or on purpose as in France. But there is no strikes going on here in Norway, so it might be accidentally.

The fire have caused all trains to a full stop, creating a chaos in the Oslo region. Still, the NSB representative do not think they need a redundant solution.

I am surprised. I would expect that a business critical system like this would have its backups, redundancy and contingency systems.  That it would not be possible for one small fire to shut down all operations of the company.

I mean, that is simply plain stupidity. 

What you can learn from this?

Accidents DO happen. Secure your business critical systems adequately.  

Syndicate content

Recent comments