5th Computing error for S5R2 on one host

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282,700
RAC: 0

RE: Assembler takes that

Message 63024 in response to message 63023

Quote:

Assembler takes that from the sublime to the ridiculous. :-)

add injury,insult

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282,700
RAC: 0

RE: With great power comes

Message 63025 in response to message 63023

Quote:
With great power comes great responsibility

Oh and, OK "Uncle Ben"...

Annika
Annika
Joined: 8 Aug 06
Posts: 720
Credit: 494,410
RAC: 0

*rofl* Really nice comments,

*rofl* Really nice comments, you two. I've understood about 3/4 of what all the assembler guys are trying to do here in the message boards, not really fit in assembler (yet) but I think it's great that you share some ideas here, maybe they'll help. And even better if you do it in such a funny way :-D

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282,700
RAC: 0

RE: *rofl* Really nice

Message 63027 in response to message 63026

Quote:
*rofl* Really nice comments, you two. I've understood about 3/4 of what all the assembler guys are trying to do here in the message boards, not really fit in assembler (yet) but I think it's great that you share some ideas here, maybe they'll help. And even better if you do it in such a funny way :-D

I honestly have no clue about assembler code... I was just trying to make a funny... Kinda like this while loop

while (horse == dead)
{
beat = true;
}

Voyager
Voyager
Joined: 8 May 05
Posts: 6
Credit: 147,290,875
RAC: 281,309

Just now found out that

Just now found out that another 4 WU's failed while I was sleeping.
And again with the same text :

===== WARNING: XLALComputeFaFb() should not be used with upsampled-SFTs!
XLAL Error - XLALComputeFaFb (ComputeFstat.c:390): Invalid argument
Level 0: $Id: HierarchicalSearch.c,v 1.164 2007/04/25 14:31:47 bema Exp $
Function call `COMPUTEFSTATFREQBAND ( &status, fstatVector.data + k, &thisPoint, stackMultiSFT.data[k], stackMultiNoiseWeights.data[k], stackMultiDetStates.data[k], &CFparams)' failed.
file HierarchicalSearch.c, line 1010
2007-05-07 08:30:33.1562 [normal]:
Level 1: $Id: ComputeFstat.c,v 1.67 2007/02/26 14:15:43 reinhard Exp $
2007-05-07 08:30:33.1562 [normal]: Status code -1: Recursive error
2007-05-07 08:30:33.1562 [normal]: function ComputeFStatFreqBand, file ComputeFstat.c, line 151
2007-05-07 08:30:33.1562 [normal]:
Level 2: $Id: ComputeFstat.c,v 1.67 2007/02/26 14:15:43 reinhard Exp $
2007-05-07 08:30:33.1562 [normal]: Status code 5: XLAL function call failed
2007-05-07 08:30:33.1562 [normal]: function ComputeFStat, file ComputeFstat.c, line 286
2007-05-07 08:30:33.1562 [CRITICAL]: BOINC_LAL_ErrHand(): now calling boinc_finish()

]]>

What the ......... is wrong with this run, still only failures on the win/intel box, and a lot of wasted time/energy.
As there are no problems with any thing other running on this box the faults are not caused by the computer, it seems like a software fault.
Maybe I'll just turn EAH off for now and wait 'till they sort it out.

Rob.

Odysseus
Odysseus
Joined: 17 Dec 05
Posts: 372
Credit: 19,454,387
RAC: 3,046

RE: No, I haven't noticed a

Message 63029 in response to message 63021

Quote:
No, I haven't noticed a WU crashing on a Mac, either.


I had one—see Compute error, exit status 193 (0xc1) in the Problems forum—but the host that got it has since returned several successful results, so it may have been a fluke, unrelated to the more widespread problems.

Majdic
Majdic
Joined: 1 Mar 06
Posts: 1
Credit: 607,538
RAC: 0

Too many errors for WUs which

Too many errors for WUs which took extremely long to compute. I'll take a break from Einstein for a while until such problems are solved.

Dave Burbank
Dave Burbank
Joined: 30 Jan 06
Posts: 275
Credit: 1,548,376
RAC: 0

Linux users look to this

Linux users look to this thread for a beta app that addresses this issue.

There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman

Voyager
Voyager
Joined: 8 May 05
Posts: 6
Credit: 147,290,875
RAC: 281,309

It happened again, another 2

It happened again, another 2 WU's with the same fault on my win/intel box.
I've aborted the last unit in my cache and I'm breaking away from EAH for the moment, with the win/intel box that is.
This box has wasted enough cpu time on these units, I've posted the problem in the bug report msg board but there's no response.
I'll come back when the problems are solved.

Rob.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.