* Posts by Bruce Grunewald

11 publicly visible posts • joined 30 Aug 2008

'Software delivered to Boeing' now blamed for 737 Max warning fiasco

Bruce Grunewald

Re: Proper Certification

Airbus have had their share of problems with rogue software fighting with pilots for control of the aircraft. And those stupid little fighter jet sticks caused an Air France crash because the F/O was in full pitch up but the Captain couldn't tell and they stalled.

I think the linked control columns in Boeing aircraft are much safer because both pilots can readily tell what the controls are doing. Boeing was also more conservative in introducing software based "assists" (like MCAS) than AirBus. I like a lot of things about the cabin design in AirBus planes, but I don't think they are quite as safe as Boeing. It might only be a .1% difference, but if that gets you killed you are still dead.

Boeing boss denies reports 737 Max safety systems weren't active

Bruce Grunewald

AOA is not unused or new

AOA is one of the inputs the "stick shaker" stall warning, so they have been around for a long time.

Windows 3.1 rebooted: Microsoft's DOS destroyer turns 20

Bruce Grunewald
Mushroom

I think you all have nostalgamnesia

I remember WFW 3.1 and 3.11 with not the same fondness as many of y'all. I had to support a bunch of machines running "dual stack" NetWare plus TCP/IP (to run TN3270 primarily) and I spent hours researching BSOD issues (where the B meant Black). This was typically cause by some errant program branching to memory location 0. Oh, how I cursed NetWare, assuming that it must be the problem.

So at the new company we get Chameleon TCP/IP because it has NFS support, and we use an old Sun box as our (only) server. PCs ran great, but the file locking support didn't work properly in Chameleon (I think), so having shared directories was dangerous.

Enter NT Server 3.5 running on a 486DX2/66 Compaq server to do file and print serving. So now we need to run NetBIOS and Chameleon as a "dual stack" to get access to everything, and lo and behold the dreaded BSOD is back. We never did resolve the problem on Win 3.x - either both dual stack setups had a buggy driver or it was inherent to Windows. I was quite happy to upgrade to Windows 95 which had built in TCP/IP support and finally cured the problem.

Feds unlock suspect's encrypted drive, avoid Constitution meltdown

Bruce Grunewald
Thumb Down

The bigger picture

I don't have a problem with the police trying to prove mortgage fraud against these two small fish, but could we go after some of the people that stole billions from the American people?

Eric Holder? Barack (I don't think they did anything illegal) Obama?

If the American sheeple don't wake up, there won't be a Bill of Rights anymore. GW Bush started using it to wipe his @ss after they passed the "Patriot Act", possibly the most ironically named bill ever.

I strongly disagree with the judge's decision to demand the encryption key. If it goes to the US Supreme Court I am pretty confident that it would be reversed, but that won't prevent you from rotting in jail and spending all your money on lawyers for the years that it will take.

iPhone 4/4S 'self combusts' in airliner inferno

Bruce Grunewald
Joke

Google just wants.....

Android users to get their Just Desserts.

RIM: 'Faulty switch took out faulty-switch-proof network'

Bruce Grunewald
FAIL

Fourth installment of Transformers?

RIM's explanations so far have as many plot holes as a Transformers movie.

We need a Deep Berry to give us the real skinny.

Bruce Grunewald
Pint

Yes, I noticed that too

So obviously they were designed to fail a different way and this failure caught them off guard.

BlackBerry services splutter back into action, again

Bruce Grunewald
Alert

Young Jedi, much to learn you have

It is a common mistake to attribute an outage to nefarious hacking when simple human stupidity is a more probable cause.

RIM somewhere attributed the problem to a software upgrade gone bad. Having seen similar problems first hand in large (carrier) networks, the long outage doesn't surprise me. It ends up being like a self-inflicted DOS attack, and the more you try to fix it the more the problem cascades due to the law of unintended consequences.

But hey, it might be wily hackers after all.

RIM stands, staggers, falls again

Bruce Grunewald
Boffin

RIM calls their data/routing centers a NOC

I work for a managed services provider, and to us and most others in IT, the NOC monitors the health of the network, but RIM are Canadians.

I was surprised by initial reports that BES was unaffected, because every BES server must communicate with a NOC to exchange mail with the handsets. It is possible that RIM provides more back end services to BIS than BES, and that is why BIS is affected more. My recollection is that there were 2 or 3 NOCs that handle the entire world. It seems like a crazy architecture to me, being so centralized, but I assumed they had some really good backup/failover capability. I would have thought that RIM would have the ability to fail over to a backup NOC, but their architecture may make this difficult. Keep in mind that every handset is sending all data communication (web, email, BBM, IM, etc) through a NOC, The only exceptions are SMS (done by the voice network) and Internet access if you are using WiFi instead of the cellular network.

I did work for a credit card authorization provider and they ran everything on Tandem or Stratus systems (built with lots of internal redundancy), and if they lost an entire machine they could bring up another one very quickly. They had redundant X.25 links from multiple providers, and an X.25 switch (my part, sigh) with a lot of built in redundancy. Those guys were serious about eliminating single points of failure. If they are still around, perhaps they can get some consulting work with RIM.

Also, don't assume that when RIM says a "core switch" failed, they mean a Cisco 6509 or similar. They don't really tell partners anything about the internals of the NOC, but it wouldn't surprise me if a "core switch" is what we would call a server, or perhaps even a mainframe. Remember they are doing message switching, which is usually a store and forward setup like SMTP.mail servers.

Boxing boomers bounced building in Seoul

Bruce Grunewald
Boffin

Narrows bridge was not felled by resonance.

This has been a common misstatement, even in engineering text books, as to the reason the bridge failed.

According to the Washington State DOT, it was caused by aerodynamics:

http://www.wsdot.wa.gov/tnbhistory/Machine/machine3.htm

(1) The principal cause of the 1940 Narrows Bridge's failure was its "excessive flexibility;"

(2) the solid plate girder and deck acted like an aerofoil, creating "drag" and "lift;"

(3) aerodynamic forces were little understood, and engineers needed to test suspension bridge designs using models in a wind tunnel.

The primary explanation of Galloping Gertie's failure is described as "torsional flutter." It will help to break this complicated series of events into several stages.

Here is a summary of the key points in the explanation.

1. In general, the 1940 Narrows Bridge had relatively little resistance to torsional (twisting) forces. That was because it had such a large depth-to-width ratio, 1 to 72. Gertie's long, narrow, and shallow stiffening girder made the structure extremely flexible.

2. On the morning of November 7, 1940 shortly after 10 a.m., a critical event occurred. The cable band at mid-span on the north cable slipped. This allowed the cable to separate into two unequal segments. That contributed to the change from vertical (up-and-down) to torsional (twisting) movement of the bridge deck.

3. Also contributing to the torsional motion of the bridge deck was "vortex shedding." In brief, vortex shedding occurred in the Narrows Bridge as follows:

(1) Wind separated as it struck the side of Galloping Gertie's deck, the 8-foot solid plate girder. A small amount twisting occurred in the bridge deck, because even steel is elastic and changes form under high stress.

(2) The twisting bridge deck caused the wind flow separation to increase. This formed a vortex, or swirling wind force, which further lifted and twisted the deck.

(3) The deck structure resisted this lifting and twisting. It had a natural tendency to return to its previous position. As it returned, its speed and direction matched the lifting force. In other words, it moved " in phase" with the vortex. Then, the wind reinforced that motion. This produced a "lock-on" event.

4. But, the external force of the wind alone was not sufficient to cause the severe twisting that led the Narrows Bridge to fail.

5. Now the deck movement went into "torsional flutter."

"Torsional flutter" is a complex mechanism. "Flutter" is a self-induced harmonic vibration pattern. This instability can grow to very large vibrations.

When the bridge movement changed from vertical to torsional oscillation, the structure absorbed more wind energy. The bridge deck's twisting motion began to control the wind vortex so the two were synchronized. The structure's twisting movements became self-generating. In other words, the forces acting on the bridge were no longer caused by wind. The bridge deck's own motion produced the forces. Engineers call this "self-excited" motion.

It was critical that the two types of instability, vortex shedding and torsional flutter, both occurred at relatively low wind speeds. Usually, vortex shedding occurs at relatively low wind speeds, like 25 to 35 mph, and torsional flutter at high wind speeds, like 100 mph. Because of Gertie's design, and relatively weak resistance to torsional forces, from the vortex shedding instability the bridge went right into "torsional flutter."

Now the bridge was beyond its natural ability to "damp out" the motion. Once the twisting movements began, they controlled the vortex forces. The torsional motion began small and built upon its own self-induced energy.

In other words, Galloping Gertie's twisting induced more twisting, then greater and greater twisting.

This increased beyond the bridge structure's strength to resist. Failure resulted.

A third of Vista PCs downgraded to XP

Bruce Grunewald
Flame

XP sux less compared to Vista

Remember XP product activation? Windows Genuine Advantage?

XP just seems better now because Vista is such a load. M$ peaked with Windows 2000. No activation, no WGA, no DRM. XP is just 2000 with a cartoon GUI.

Remember all the advanced features Vista was supposed to have? M$ had to strip them all out to deliver anything. Now it's just XP with Aero and extra suckage.

WTF does Vista do that you can't do with XP or Linux? And don't say "oh, it's so pretty", pretty you can add on to the OS you have. Vista is a solution looking for a problem to solve. Haven't all the real OS advancements already been made? And no, the GUI is not the OS, despite what M$ would like you to think. And what do you really need from a GUI? Fast, intuitive, and consistent? You can't get that by changing the damn thing all around with every new OS. I always run XP in "Classic" mode because that XP start menu is pure crap. And don't get me started about Cisco's web site that they feel compelled to eff up every year or so.........