FAA Crash Apparently on Windows Network
nes...@wigner.berkeley.edu
the "National Airspace Data Interchange Network":
<Quote>
According to an internal FAA document, the system, called the National
Airspace Data Interchange Network, crashed on Thursday and caused in
134 departure delays. The Salt Lake City system also took over but had
problems with the high queue level, the document said. The system also
failed in June 2007.
</Quote>
http://www.washingtonpost.com/wp-dyn/content/article/2008/08/26/AR2008082602203.html?hpid=sec-nation
Network apparently runs on Windows. From "FAA Modernizing National
Airspace Data Interchange Network with Stratus ftServer systems":
<Quote>
The FAA is implementing the Stratus servers, which use Intel Xeon 2.8
MHz large cache MP processors and support the Microsoft Windows
operating system, at control centers in Atlanta and Salt Lake City.
</Quote>
http://www.stratustechnologies.com/news/2005/20050314a.htm
Research by PJ on groklaw.
Tuesday's failure originated in Atlanta. I wonder DFS had anything to
do with it. Maybe it crashed while he was busy posting Linux horror
stories.
Roy Schestowitz
Hash: SHA1
____/ nes...@wigner.berkeley.edu on Wednesday 27 August 2008 20:17 : \____
> In addition to Tuesday's disaster, there have been earlier crashes in
> the "National Airspace Data Interchange Network":
>
> <Quote>
> According to an internal FAA document, the system, called the National
> Airspace Data Interchange Network, crashed on Thursday and caused in
> 134 departure delays. The Salt Lake City system also took over but had
> problems with the high queue level, the document said. The system also
> failed in June 2007.
> </Quote>
>
>
http://www.washingtonpost.com/wp-dyn/content/article/2008/08/26/AR2008082602203.html?hpid=sec-nation
>
> Network apparently runs on Windows. From "FAA Modernizing National
> Airspace Data Interchange Network with Stratus ftServer systems":
>
> <Quote>
> The FAA is implementing the Stratus servers, which use Intel Xeon 2.8
> MHz large cache MP processors and support the Microsoft Windows
> operating system, at control centers in Atlanta and Salt Lake City.
> </Quote>
>
> http://www.stratustechnologies.com/news/2005/20050314a.htm
>
> Research by PJ on groklaw.
That was obvious from previous reports about last year's incidence. See reply
to Richard.
> Tuesday's failure originated in Atlanta. I wonder DFS had anything to
> do with it. Maybe it crashed while he was busy posting Linux horror
> stories.
Or maybe he spilled mustard on his <strike>shirt</strike> boxers. He works in a
garage.
- --
"There's a lot of Linux out there -- much more than Microsoft generally signals
publicly -- and their customers are using it..." --Paul DeGroot, a Directions
On Microsoft analyst.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEARECAAYFAki10XYACgkQU4xAY3RXLo4+BgCeMLGt2FIukPe3lGpn+zEUZpPA
mgsAn2PIPnHKJ4RAXkRDwXTpgB72IvXH
=yIPP
-----END PGP SIGNATURE-----
Sinister Midget
> In addition to Tuesday's disaster, there have been earlier crashes in
> the "National Airspace Data Interchange Network":
> http://www.washingtonpost.com/wp-dyn/content/article/2008/08/26/AR2008082602203.html?hpid=sec-nation
>
> Network apparently runs on Windows. From "FAA Modernizing National
> Airspace Data Interchange Network with Stratus ftServer systems":
> Tuesday's failure originated in Atlanta. I wonder DFS had anything to
> do with it. Maybe it crashed while he was busy posting Linux horror
> stories.
He probably uploaded one of his high-demand Excel spreadsheets that he
gets paid a lot of money to work on.
--
VISTA: Very Idiotic Software Trashing Applications
---- Posted via Pronews.com - Premium Corporate Usenet News Provider ----
http://www.pronews.com offers corporate packages that have access to 100,000+ newsgroups
Ramon F Herrera
<ness...@wigner.berkeley.edu> wrote:
> In addition to Tuesday's disaster, there have been earlier crashes in
> the "National Airspace Data Interchange Network":
>
> <Quote>
> According to an internal FAA document, the system, called the National
> Airspace Data Interchange Network, crashed on Thursday and caused in
> 134 departure delays. The Salt Lake City system also took over but had
> problems with the high queue level, the document said. The system also
> failed in June 2007.
> </Quote>
>
>
> Network apparently runs on Windows. From "FAA Modernizing National
> Airspace Data Interchange Network with Stratus ftServer systems":
>
> <Quote>
> The FAA is implementing the Stratus servers, which use Intel Xeon 2.8
> MHz large cache MP processors and support the Microsoft Windows
> operating system, at control centers in Atlanta and Salt Lake City.
> </Quote>
>
> http://www.stratustechnologies.com/news/2005/20050314a.htm
>
I wonder whether it is possible to the Freedom of Information Act, in
order to get the tech skinny.
As soon as I saw the news about the computer crash, I immediately
reacted instinctively: "Aha! I bet they were running Windows!".
Then I said to myself: "Nah! That's impossible, they cannot be that
irresponsible, after all there are lives at stake".
-RFH
Psyc Geek (TAB)
:: The FAA said the source of the computer software malfunction was
:: a "packet switch" that "failed due to a database mismatch."
This could have happened on ANY SYSTEM.
Does anybody in COLA actually read the stories?
Richard Rasker
Any day now we'll hear that his Linux tales of woe were routed through that
particular server, and simply overloaded the OS.
Richard Rasker
--
http://www.linetec.nl
Tim Smith
<f4a9d760-ad33-43e7...@y21g2000hsf.googlegroups.com>,
Ramon F Herrera <ra...@conexus.net> wrote:
> As soon as I saw the news about the computer crash, I immediately
> reacted instinctively: "Aha! I bet they were running Windows!".
Since a large majority of everything outside of internet infrastructure,
embedded devices, and business enterprise back end systems run Windows,
that's a pretty safe bet.
> Then I said to myself: "Nah! That's impossible, they cannot be that
> irresponsible, after all there are lives at stake".
Windows is very widely used for things where lives would be at stake if
it failed. Everywhere from the structural design of new building, to
controlling industrial robots, to running the navigation systems on
fishing and cargo vessels, you'll find Windows firmly entrenched.
--
--Tim Smith
Jerry McBride
Funny it happened on a HIGHLY tested and CERTIFIED OS and Hardware mix...
And whats up with "having to be rebooted" all the time?
--
Jerry McBride (jmcb...@mail-on.us)
Rex Ballard
> Psyc Geek (TAB) wrote:
> > :: The FAA said the source of the computer software malfunction was
> > :: a "packet switch" that "failed due to a database mismatch."
> > This could have happened on ANY SYSTEM.
> Funny it happened on a HIGHLY tested and CERTIFIED OS and Hardware mix...
> And whats up with "having to be rebooted" all the time?
I remember working for a competitor to Stratus, and later being a
customer. When you
used the Stratus UNIX system you turned it on and it took an act of
god to bring it down.
Many of their systems had 99.99% uptime PER SERVER. In addition,
their cut-over was
very fast and reliable, so it pretty easy to get 99.999% availablity -
which translates to 5 minutes
of down-time PER YEAR, and that didn't include exclusions for
"maintenance" or for "backups" or
for "routine reboots".
Windows was down for several HOURS for TWO YEARS IN A ROW!! Add to
that an hour a week for "routine mantenance" and "routine reboots".
That drops the availability down to about 98%.
> Jerry McBride (jmcbr...@mail-on.us)
Moshe Goldfarb.
FAA systems get re-booted all the time and usually at least once a week.
It's called an IPL....
All systems are mirrored and they do a switch over to test.
One thing the article mentioned that is true is that they have been cutting
the performance and testing phases of upcoming hardware/software upgrades
in order to get this stuff out to the field.
Why?
Because the current system is at least 10 years, yes 10 years behind what
is available now.
And that is a conservative estimate.
But of course the OSS mouthpiece Groklaw says something so it has to be
true.
No wonder the credibility of that site is tanking lately.
I have to laugh watching you armchair losers trying your best to blame this
on Microsoft, and it might very well turn out that way but the interesting
part is how little you really know about this situation yet how rabid and
zealot like you act.
It just confirms my suspeician that most of you have no clue how a glass
house operates and how systems are used in the real world.
Your P3 running Linux doesn't count, although it might be a step up for the
FAA in some cases.
BTW here is a link I just found which confirms what I was telling you.
BTW they are WAY behind in the upgrade.
Take note of the machine type and the 370 instruction set, which I
mentioned in my OP.
Read it and weep, or run and slink like you usually do.
http://www.gcn.com/print/vol19_no8/1711-1.html#
--
Moshe Goldfarb
Collector of soaps from around the globe.
Please visit The Hall of Linux Idiots:
http://linuxidiots.blogspot.com/
Moshe Goldfarb.
The people in this group have no idea what they are talking about.
I often wonder how many of them actually earn their living in the
information technology field.
See my other post to back up my claims...
BTW I make a game of looking at monitors that may be running in the
background on various science channel programs and 9 times out of 10 they
are running either Windows or OSX.
Rex Ballard
wrote:
> > Network apparently runs on Windows. From "FAA Modernizing National
> > Airspace Data Interchange Network with Stratus ftServer systems":
> > <Quote>
> > The FAA is implementing the Stratus servers, which use Intel Xeon 2.8
> > MHz large cache MP processors and support the Microsoft Windows
> > operating system, at control centers in Atlanta and Salt Lake City.
> > </Quote>
Stratus made some amazing computers, many of which were extremely
trusted. But in those days, those computers ran UNIX, not Windows.
After (how many?) crashes since the deployment of the Windows versions
I wonder if the FAA is thinking about switching back to the UNIX
versions. We know of the ones in 2008 and June 8 2007, but how many
other Windows related delays have there been since the system was
deployed in 2005, and in both of the two public failures, the recovery
time was several hours.
Hadron
> On Wed, 27 Aug 2008 17:01:10 -0700, Tim Smith wrote:
>
>> In article
>> <f4a9d760-ad33-43e7...@y21g2000hsf.googlegroups.com>,
>> Ramon F Herrera <ra...@conexus.net> wrote:
>>> As soon as I saw the news about the computer crash, I immediately
>>> reacted instinctively: "Aha! I bet they were running Windows!".
>>
>> Since a large majority of everything outside of internet infrastructure,
>> embedded devices, and business enterprise back end systems run Windows,
>> that's a pretty safe bet.
>>
>>> Then I said to myself: "Nah! That's impossible, they cannot be that
>>> irresponsible, after all there are lives at stake".
>>
>> Windows is very widely used for things where lives would be at stake if
>> it failed. Everywhere from the structural design of new building, to
>> controlling industrial robots, to running the navigation systems on
>> fishing and cargo vessels, you'll find Windows firmly entrenched.
>
> The people in this group have no idea what they are talking about.
> I often wonder how many of them actually earn their living in the
> information technology field.
Lets see:
Tattoo Vampire is an estate agent
High Plains Hypocrite uses windows at work
chrisv runs an off license or something
spamowitz lives in a University cell
etc etc
Laughable isn't it?
Hadron
> On 2008-08-27, nes...@wigner.berkeley.edu <nes...@wigner.berkeley.edu> claimed:
>> In addition to Tuesday's disaster, there have been earlier crashes in
>> the "National Airspace Data Interchange Network":
>
>> http://www.washingtonpost.com/wp-dyn/content/article/2008/08/26/AR2008082602203.html?hpid=sec-nation
>>
>> Network apparently runs on Windows. From "FAA Modernizing National
>> Airspace Data Interchange Network with Stratus ftServer systems":
>
>> Tuesday's failure originated in Atlanta. I wonder DFS had anything to
>> do with it. Maybe it crashed while he was busy posting Linux horror
>> stories.
>
> He probably uploaded one of his high-demand Excel spreadsheets that he
> gets paid a lot of money to work on.
Erm, people get paid a LOT of money to design good spreadsheets with
interfaces to a back end DBMS.
What is your point?
The Ghost In The Machine
<hadro...@gmail.com>
wrote
on Thu, 28 Aug 2008 06:15:49 +0200
<g958q4$qbi$2...@registered.motzarella.org>:
Not to mention that that back end DBMS could be anything
from MS SQL Server to a high-powered SAS affair, though
many would probably just go with Oracle.
All of these are drivable from Linux Java and Windows ODBC.
The failure originated somewhere else, and undetectably.
The issue apparently was a combination of Hampton crashing
and the backup in Salt Lake City overloading and bogging
down, presumably because nobody tested it in such a
real-world failure scenario.
Basically, nobody could file flight plans, and hundreds of
flights had to be cancelled.
http://www.foxnews.com/story/0,2933,411381,00.html
They're still looking for the actual cause, but it looks
to be distributed.
--
#191, ewi...@earthlink.net
If your CPU can't stand the heat, get another fan.
** Posted from http://www.teranews.com **
Linonut
> BTW here is a link I just found which confirms what I was telling you.
> BTW they are WAY behind in the upgrade.
>
> Take note of the machine type and the 370 instruction set, which I
> mentioned in my OP.
>
> Read it and weep, or run and slink like you usually do.
>
> http://www.gcn.com/print/vol19_no8/1711-1.html
Why would we weep or run/slink about it?
Besides, that article is 8 years old. What actually was done /today/?
--
First as to speech. That privilege rests upon the premise that
there is no proposition so uniformly acknowledged that it may not be
lawfully challenged, questioned, and debated....
Linonut
> Basically, nobody could file flight plans, and hundreds of
> flights had to be cancelled.
>
> http://www.foxnews.com/story/0,2933,411381,00.html
>
> They're still looking for the actual cause, but it looks
> to be distributed.
That article mentions NADIN. Here's a description of it:
http://www.stratustechnologies.com/news/2005/20050314a.htm
The FAA is implementing the Stratus servers, which use Intel Xeon 2.8
MHz large cache MP processors and support the Microsoft Windows
operating system, at control centers in Atlanta and Salt Lake City.
Uninterrupted availability of the NADIN 1 is important to all aspects
of the aviation industry, as well as the nation's economy and,
increasingly, as a tool to help protect national security. After
thorough technology and product evaluations, Stratus was chosen as
best able to provide an open platform with 99.999 percent uptime
reliability -- which is mandatory for running an application as
important as NADIN 1 -- together with the required caliber of
maintenance, logistical support, and long service life.
So it was Windows. In 2005.
Never trust a COLA poster (even myself, but especially flatfish) --
follow it up yourself.
So how does the Wall Street Journal characterize the Windows-based
NADIN system?
http://online.wsj.com/article/SB121988583932078345.html?mod=googlenews_wsj
FAA Rushes to Fix Aging
System That Caused Delays
Aging? It's only three freakin' years old, and it's already decrepit?
Apparently:
The Federal Aviation Administration said it is overhauling an
error-prone computer system that caused hundreds of delayed flights
Tuesday.
Error-prone? Is that Stratus's direct fault (bad code), or indirect
fault (bad choice of problematic platform)?
However, NADIN is apparently not just Windows:
...said the agency would accelerate its upgrade of the fraying Nadin
system, which dates back to the 1980s. An improved version with
vastly higher memory will be installed by the end of the year, he
said, with improvements noticeable by the end of September.
The article also talks about a Harris telcom system.
There is obviously a /lot/ going on in the background in this story.
--
Ten of the meanest cons in the state pen met in the corner of the yard to
shoot some craps. The stakes were enormous, the tension palpable.
When his turn came to shoot, Dutsky nervously plunked down his
entire wad, shook the dice and rolled. A smile crossed his face as a
seven showed up, but it quickly changed to horror as third die slipped out
of his sleeve and fell to the ground with the two others. No one said a
word. Finally, Killer Lucci picked up the third die, put it in his pocket
and handed the others to Dutsky.
"Roll 'em," Lucci said. "Your point is thirteen."
Linonut
Or Irix.
I got a good chuckle out of seeing "xterm" in a window caption one time.
--
Boss: You forgot to assign the result of your map!
Hacker: Dang, I'm always forgetting my assignations...
Boss: And what's that "goto" doing there?!?
Hacker: Er, I guess my finger slipped when I was typing "getservbyport"...
Boss: Ah well, accidents will happen. Maybe we should have picked APL.
-- Larry Wall in <1997103117...@wall.org>
nes...@wigner.berkeley.edu
>
> FAA Rushes to Fix Aging
> System That Caused Delays
>
> Aging? It's only three freakin' years old, and it's already decrepit?
> Apparently:
Yeah, and at 99.999% up time, it can only have 1 hour down every 11
years. So far it's been quite a lot more than that in 3 years.
DFS
> On 2008-08-27, nes...@wigner.berkeley.edu
> <nes...@wigner.berkeley.edu> claimed:
>> In addition to Tuesday's disaster, there have been earlier crashes in
>> the "National Airspace Data Interchange Network":
>
>> http://www.washingtonpost.com/wp-dyn/content/article/2008/08/26/AR2008082602203.html?hpid=sec-nation
>>
>> Network apparently runs on Windows. From "FAA Modernizing National
>> Airspace Data Interchange Network with Stratus ftServer systems":
>
>> Tuesday's failure originated in Atlanta. I wonder DFS had anything
>> to do with it. Maybe it crashed while he was busy posting Linux
>> horror stories.
>
> He probably uploaded one of his high-demand Excel spreadsheets that he
> gets paid a lot of money to work on.
Poor jealous Gidget.
Here's how I double your earnings:
Public Function foundDSN(dsnName As String, odbcINIfile As String) As
Boolean
foundDSN = False
If Dir(odbcINIfile) <> "" Then
Open odbcINIfile For Input Access Read As #1
Do Until EOF(1)
Line Input #1, sTxt
If Left(sTxt, Len(dsnName)) = dsnName Then
foundDSN = True
Close #1
Exit Function
End If
Loop
Close #1
End If
End Function
Here's how I triple your earnings:
Public Function nextKey(tblName As String) As Long
Call DeleteQuery("Q_NextKey")
Set qItem = db.CreateQueryDef("Q_NextKey")
qItem.ReturnsRecords = True
qItem.Connect = "ODBC;DSN=dsn;UID=id;PWD=pwd"
qItem.ODBCTimeout = 15
qItem.SQL = "EXEC owner.sp_nextkey """ & tblName & """"
qItem.Close
db.QueryDefs.Refresh
Set rs = db.OpenRecordset("Q_NextKey")
nextKey = rs(0)
rs.Close
Set rs = Nothing
Call DeleteQuery("Q_NextKey")
End Function
And here's how I quadruple your earnings:
Dim objPutStream As Object
Set objPutStream = objPutDoc.GetPropertyValue("%CONTENT")
Dim bData() As Byte, readBytes As Long
ReDim bData(FileLen(docPath))
readBytes = FileLen(docPath)
Open docPath For Binary Access Read Lock Read As #1
Get #1, , bData
objPutStream.Write bData, readBytes
objPutStream.SetComplete
Close #1
Set objPutStream = Nothing
Moshe Goldfarb.
> * Moshe Goldfarb. peremptorily fired off this memo:
>
>> BTW here is a link I just found which confirms what I was telling you.
>> BTW they are WAY behind in the upgrade.
>>
>> Take note of the machine type and the 370 instruction set, which I
>> mentioned in my OP.
>>
>> Read it and weep, or run and slink like you usually do.
>>
>> http://www.gcn.com/print/vol19_no8/1711-1.html
>
> Why would we weep or run/slink about it?
>
> Besides, that article is 8 years old. What actually was done /today/?
As of yesterday, most of it is STILL running in the FAA centers.....
They are way behind.
As for your other article, it is flawed but I am not at liberty to speak
about it. Suffice to say when they are talking "aging systems" while they
IMPLY recent Windows systems in the article, they are really speaking of
the entire infrastructure which if you saw it you would probably never fly
again and not because of Windows.
Additionally the new platform has been under test for a couple of years and
IS being deployed slowly, very slowly so the OVERALL SYSTEM, is a complex
hybrid.
The IBM 9672 GEN 1 mainframes running 370 code ARE STILL IN OPERATION with
2 of them at each site. They along with the AIX systems the controllers at
the centers use are DIRECTLY RESPONSIBLE for tracking the planes.
IOW the little triangles you see on the screens.
Like I said, you guys have no clue.....
Tim Smith
"Moshe Goldfarb." <brick_...@gmail.com> wrote:
> BTW I make a game of looking at monitors that may be running in the
> background on various science channel programs and 9 times out of 10 they
> are running either Windows or OSX.
Same on History channel. A lot of episodes of "Modern Marvels" go to
interesting factories or labs, and you see Windows all over those places.
Some examples of the kind of places I've spotted Windows on History,
Discovery, National Geographic, etc.:
* A show about the "Water Cube" building in Beijing (a remarkable
building). Looks like the engineering work was pretty much all done on
Windows, including design, structural analysis, analysis of earthquake
response, analysis of fire spread and evacuation speed, etc. I didn't
see any signs of non-Windows computers.
* On "America's Port" (a show that chronicles activity at the Port of
Los Angeles), the computers handling the navigation and steering
controls on nearly all the giant cargo ships seems to be Windows.
Various computers in port operations all seemed to be Windows.
* The fishing fleets in the Bering Sea and the North Atlantic seem to be
heavy on Windows in the wheel house of the boats.
And you know what? Windows works fine for all these things. So would
DOS. So would OS/2. So would Linux. So would OS X. These various
industrial and scientific systems aren't like desktops--you don't have
people installing random third-party software on them. In a controlled
environment, pretty much any OS is OK. In an uncontrolled environment,
they all have problems.
--
--Tim Smith
Linonut
> In article <wlpvvyi027lr$.1pjlp3xt...@40tude.net>,
> "Moshe Goldfarb." <brick_...@gmail.com> wrote:
>> BTW I make a game of looking at monitors that may be running in the
>> background on various science channel programs and 9 times out of 10 they
>> are running either Windows or OSX.
>
> Same on History channel. A lot of episodes of "Modern Marvels" go to
> interesting factories or labs, and you see Windows all over those places.
Wow! Lots of people use Windows!
That's some hard news there!
> And you know what? Windows works fine for all these things. So would
> DOS. So would OS/2. So would Linux. So would OS X. These various
> industrial and scientific systems aren't like desktops--you don't have
> people installing random third-party software on them. In a controlled
> environment, pretty much any OS is OK. In an uncontrolled environment,
> they all have problems.
I heartily agree (except that I find Windows cumbersome).
--
Using TSO is like kicking a dead whale down the beach.
-- S. C. Johnson