Post #137,278
1/23/04 6:27:48 AM
|
Spirit responds, sort of
[link|http://marsrovers.jpl.nasa.gov/newsroom/pressreleases/20040122a.html|http://marsrovers.jp...es/20040122a.html]
In typical manner, the report does not explain in detail what the real problem is - everything it seems must be subjected to spin control, even a damn robot.
-drl
|
Post #137,286
1/23/04 8:31:32 AM
|
What are you talking about?
They don't KNOW what the real problem is, so how can you expect them to give a detailed explanation?
The aren't getting any telemetry. There's no way to diagnose on anything other than a gross level.
Regards,
-scott anderson
"Welcome to Rivendell, Mr. Anderson..."
|
Post #137,287
1/23/04 8:48:09 AM
|
Re: What are you talking about?
1. Is the hardware busted? Evidence?
2. Is the software busted? Evidence?
3. Is the power supply busted? Evidence?
4. Is something unknown busted? Lack of evidence?
These NASA reports are famously empty of content, across the board. Everything out of their press agent's mouth is spun into the best possible form, as if planets and spacecraft were running for re-election.
-drl
|
Post #137,293
1/23/04 9:22:57 AM
|
The engineers can't come up
with a single problem that would explain all symptoms. That's waht I heard on NPR today.
(multiple simultaneous failures? I hope they considered that, NPR was not too clear)
--
Select [link|http://www.glumbert.com/pictures/Default.asp?index=30|here].
|
Post #137,296
1/23/04 9:30:01 AM
|
BS
They must have several working scenarios. We deserve to know what it is. There is no point in hiding information.
Look how they spun the Columbia disaster in the beginning. The management at that place is a continuing disaster.
-drl
|
Post #137,319
1/23/04 10:25:40 AM
|
They've given us those.
Power supply problem or possible system crash. Until they get telemetry, there's no way of knowing more.
Regards,
-scott anderson
"Welcome to Rivendell, Mr. Anderson..."
|
Post #137,321
1/23/04 10:28:56 AM
|
Re: They've given us those.
Why not just assume bad software and upload a new start? If that doens't work it's dead anyway.
I was listening to one of the team leaders describe his local tests involving an impossible scenario. He was testing an impossible scenario, just in case it magically became possible on Mars. Another example I linked involved shock testing a flight instrument when the test would prove nothing other than, yes, you can break this instrument if you shock it.
I just get the feeling again and again that NASA has some of the sharpest balloons in the world.
-drl
|
Post #137,326
1/23/04 10:51:29 AM
|
Why not just reboot
a multimillion dollar piece of equipement that cannot be reached for service for next 100 years?
Microsoft mentality in action.
--
Select [link|http://www.glumbert.com/pictures/Default.asp?index=30|here].
|
Post #137,328
1/23/04 10:56:57 AM
|
Because it may not survive the reboot.
Especially if it's a power issue.
On the other hand, if they can get telemetry first before rebooting (which they can always do as a last resort), they can do some actual debugging instead of acting like NT admins.
Now that Spirit has responded with 20 minutes of data, I'd say that the NASA guys are right on with what they're trying to do. Your solution is a last resort only.
Regards,
-scott anderson
"Welcome to Rivendell, Mr. Anderson..."
|
Post #137,295
1/23/04 9:29:39 AM
|
#4
They don't have any frigging evidence. That's exactly what they're saying.
How the hell do you diagnose white noise or a tone? You don't.
Regards,
-scott anderson
"Welcome to Rivendell, Mr. Anderson..."
|
Post #137,382
1/23/04 7:08:04 PM
|
New info
It's spontaneously attempting to reboot again and again. There is only one OS I know of that does this - NT. I would not be surprised if this fucker is running some version of embedded NT.
I did find out it has a VME bus - perhaps our embedded engineer (Scott2? ChrisR? Burns?) could tell us something.
-drl
|
Post #137,400
1/23/04 7:52:04 PM
|
Nope. VxWorks
[link|http://www.windriver.com/news/press/20040105.html|http://www.windriver...ess/20040105.html]
What exactly do you expect it to do, halt and catch fire?
--
Chris Altmann
|
Post #137,403
1/23/04 7:56:44 PM
1/23/04 8:01:13 PM
|
LRPD!
No, Mr. Bond, I expect it to dump!
What exactly is VxWorks based on architecture-wise?
PS good find.
I notice that Windows is an allowed development platform. Maybe this thing is klezzed.
"We've been getting odd messages - Last Internet Update, Bug Advice, MS Network Security Division...."
-drl
Edited by deSitter
Jan. 23, 2004, 08:01:13 PM EST
|
Post #137,356
1/23/04 3:12:24 PM
|
A Martian must have picked it up and shook it.
bcnu,
Mikem
I don't do third world languages. So no, I don't do Java.
|
Post #137,566
1/24/04 8:44:30 PM
|
Back from the Brink -
Seems they latched onto a diagnosis of 'bad flash memory'; were able to restore basic functioning with orders to eschew flash -for now-. Then - it was time to go beddy-bye.
So it seems that communications are restored.. also it appears: their Fire-Ax fallback, "break glass and use This" planning wasn't too shabby.
Guess: once control is orderly - bet they can test, reformat flash, possibly mapping around heavy-ion blasted? bad elements. If not successful, then - RAM will be limited to perhaps less sophisticated command storage for next events. I kinda doubt there's a HD within 50M miles.
Rates a Whew!
|
Post #137,612
1/25/04 12:19:27 AM
|
Re: Back from the Brink -
Still, this was a basic design screwup in the software - a computer should never continuously reboot like that. Why am I in doubt about the skills of younger programmers?
And why do I suspect C++ as the culprit?
-drl
|
Post #137,663
1/25/04 12:17:22 PM
|
Continuous rebooting
If there was hardware damage, then the software is outside it's spec range, and anything can happen, including continuous rebooting. If the flash is damaged (say some bits were flipped by a stray cosmic ray), it might very well be that the third instruction on boot is a branch to the first instruction... which is a go gently into that good night scenario.
--\n-------------------------------------------------------------------\n* Jack Troughton jake at consultron.ca *\n* [link|http://consultron.ca|http://consultron.ca] [link|irc://irc.ecomstation.ca|irc://irc.ecomstation.ca] *\n* Kingston Ontario Canada [link|news://news.consultron.ca|news://news.consultron.ca] *\n-------------------------------------------------------------------
|
Post #137,666
1/25/04 12:58:02 PM
|
Re: Continuous rebooting
C'mon Jake, even an IBM PS/2 would run a POST and halt if hardware problems were found. This sounds like it was loading something from corrupted flash RAM - it should have never got that far.
Did an impatient engineer insist on a quick POST without full checks of the hardware? :)
Anecdote: The first lunar landing was nearly aborted because debugging code was not properly backed out of the production software (actually firmware, magnetic core ROM :) This piece of code was meant to simulate a broken rendezvous radar on an IBM 360 LM hardware simulator - the break was to get the computer to try to calculate an angle with both a sine and a cosine of 0 - impossible. The code would not execute on landing because the rendezvous radar was supposed to be turned off - however an error in procedures caused the radar to be left on during the landing, and the bad code started stealing cycles and eventually overloaded the computer several times during the actual landing! The computer was designed to restart with the last "state vector" of the LM's position and velocity, so the landing went on as planned.
-drl
|
