LHC@home

I posted a problem I had in the Questions/Problems section a few minutes ago, but since I'm sure more people will see it here, I'm posting a link to it.

http://lhcathome.cern.ch/forum_thread.php?id=362

Chris

I replied.

But it's true that the questions/problems boards tend to have a lot of unanswered questions.
______________________________________________________________
Did your tech wear a static strap? No? Well, there ya go! :p

Hi,

I have got two WU so far that have errored...

The error occurs wheen the WU is 100% of processing.

I get a popup box labeled Fortran that have an Information message saying "jwe0019i-u The program has terminated abnormally with Exception Code ECEPTION_ACCESS_VIOLATION" and has an OK button

Pressingf OK gives another Fortran Inforamtion Box with "Error summary (Fortran) error number jwe0019i u 1 total error count = 1" and has an OK button.

Pressing ok give a Computation error on the WU.

WU was v64lhc1000pro3s14_16544.46_1_sixvf_15479_1_0 (
http://lhcathome.cern.ch/result.php?resultid=151928)

Stderr.txt says "2004-09-22 19:53:45 [LHC@home] Unrecoverable error for result v64lhc1000pro3s14_16544.46_1_sixvf_15479_1 (The session was canceled. (0xf0) - exit code 240 (0xf0))"

Had this error yesterday on my notebook.
Everything ran fine rill it reached 100% and than

bam error

here u see>> http://lhcathome.cern.ch/result.php?resultid=155159

I saw this too with CLI installed as a service. The service would start but not run until I had stopped it and run the GUI. Then I saw the Fortran errors, and the GUI kicked off with the next unit. Thereafter the service would run again.

Giskard - the first telepathic robot.

Ok, a totally new bug species. Nice.


Markku Degerholm
LHC@home Admin

Sorry Markku :)

If it makes you feel any better it's been pretty tame since then (as far as I can tell). But my computer apparently has unexpectedly restarted at least a couple of times in the past few days so I can't be 100% sure. I suspect that the restarting (due to a "bugcheck" ) is from overclocking, but again, I'm not positive. I did check another computer I have, an Intel P3 1.0GHz with 512RDRAM and WinXP SP2 (which has not restarted) and it shows no errors.

Chris

> Ok, a totally new bug species. Nice.
>
>
> Markku Degerholm
> LHC@home Admin
>

I had the error code for the first time.

The message of the error is an exclamation message with content:
"CreateFile,errno=80,unit=18".

Boinc create following message:

LHC@home - 2004-09-25 11:54:05 - Unrecoverable error for result v64lhc1000pro22s4_6517.73_1_sixvf_45152_2 (De toegangscode is ongeldig. (0xc) - exit code 12 (0xc))
LHC@home - 2004-09-25 11:54:05 - Deferring communication with project for 1 minutes and 0 seconds
LHC@home - 2004-09-25 11:54:05 - Computation for result v64lhc1000pro22s4_6517.73_1_sixvf_45152 finished
SETI@home - 2004-09-25 11:54:05 - Pausing result 06my04aa.15945.3825.978394.2_0 (left in memory)
LHC@home - 2004-09-25 11:54:06 - Starting result v64lhc1000pro37s4_6581.9_1_sixvf_60850_1 using sixtrack version 4.46

De toegangscode is ongeldig is Dutch and means
The access code is invalid.

No graphics neither screensaver was used.

Using Boinc v4.09 and sixtrack v4.46, Win XP Pro SP1, BoincLogX v1.11.


Greetings from Belgium

I got a new Fortran error but no message in the message tab of Boinc. It occured right at this moment:

LHC@home - 2004-10-02 15:48:02 - Computation for result v64lhc1000pro55s6_8558.59_1_sixvf_138528 finished
SETI@home - 2004-10-02 15:48:02 - Resuming result 25ap04aa.23553.17201.473568.29_6 using setiathome version 4.05
LHC@home - 2004-10-02 15:48:02 - Pausing result v64lhc1000pro53s6_8548.24_1_sixvf_136415_3 (left in memory)
SETI@home - 2004-10-02 15:48:02 - Starting result 01mr04ab.8181.546.934666.164_5 using setiathome version 4.05
LHC@home - 2004-10-02 15:48:02 - Started upload of v64lhc1000pro55s6_8558.59_1_sixvf_138528_3_0
LHC@home - 2004-10-02 15:48:09 - Finished upload of v64lhc1000pro55s6_8558.59_1_sixvf_138528_3_0

Unfortunately, the windows of the Fortran error could not be viewed. The taskbar that is normally autohided could not be hided anymore and the window of the Fortran error message could not be closed as the content of it could not be viewed.
Right clicking on the Fortran icon in the taskbar was working but using "close" was not working.

Greetings from Belgium

oups !!!

those ended up with an -1.#j% in the process box tranfer tab

LHC@home - 2004-10-06 22:57:09 - Started upload of v64lhc1000prothree15s16_18537.35_1_sixvf_28387_2_0
LHC@home - 2004-10-06 22:57:15 - Finished upload of v64lhc1000prothree15s16_18537.35_1_sixvf_28387_2_0
LHC@home - 2004-10-06 22:57:15 - No data transferred

LHC@home - 2004-10-06 23:21:23 - Started upload of v64lhc1000prothree15s16_18545.27_1_sixvf_28475_1_0
LHC@home - 2004-10-06 23:21:27 - Finished upload of v64lhc1000prothree15s16_18545.27_1_sixvf_28475_1_0
LHC@home - 2004-10-06 23:21:27 - No data transferred

the second one was obtained by closing boinc_gui and starting it about 1 sec later it simply terminates the wu short and upload's but with -1.#j%
( win 98 ) No fortan popup message.

2004-10-09 18:05:32 [SETI@home] Pausing result 14mr04aa.17223.14977.186078.34_2 (removed from memory)
2004-10-09 18:05:32 [LHC@home] Restarting result v64lhc1000profour18s14_16543.47_1_sixvf_15468_3 using sixtrack version 4.46

fortran error message at 18:05:32 computer frooze had to reboot .

message after reboot:


SETI@home - 2004-10-09 23:45:23 - Deferring computation for result 14mr04aa.17223.14977.186078.34_2
LHC@home - 2004-10-09 23:45:23 - Resuming computation for result v64lhc1000profour18s14_16543.47_1_sixvf_15468_3 using sixtrack version 4.46
--- - 2004-10-09 23:45:38 - May run out of work in 0.75 days; requesting more
LHC@home - 2004-10-09 23:45:38 - Requesting 64992 seconds of work
LHC@home - 2004-10-09 23:45:39 - Sending request to scheduler: http://lhcathome-sched1.cern.ch/scheduler/cgi
LHC@home - 2004-10-09 23:45:39 - Started upload of v64lhc1000profour18s14_16569.39_1_sixvf_15756_2_0
LHC@home - 2004-10-09 23:45:46 - Scheduler RPC to http://lhcathome-sched1.cern.ch/scheduler/cgi succeeded
LHC@home - 2004-10-09 23:45:46 - Message from server: No work available
LHC@home - 2004-10-09 23:45:46 - No work from project
LHC@home - 2004-10-09 23:45:46 - Deferring communication with project for 1 hours, 0 minutes, and 0 seconds
LHC@home - 2004-10-09 23:45:56 - Finished upload of v64lhc1000profour18s14_16569.39_1_sixvf_15756_2_0
LHC@home - 2004-10-09 23:45:56 - Throughput 4119 bytes/sec
LHC@home - 2004-10-09 23:48:08 - Unrecoverable error for result v64lhc1000profour18s14_16543.47_1_sixvf_15468_3 (The session was cancelled. (0xf0) - exit code 240 (0xf0))
LHC@home - 2004-10-09 23:48:08 - Deferring communication with project for 57 minutes and 38 seconds
LHC@home - 2004-10-09 23:48:08 - Computation for result v64lhc1000profour18s14_16543.47_1_sixvf_15468 finished
SETI@home - 2004-10-09 23:48:09 - Restarting result 14mr04aa.17223.14977.186078.34_2 using setiathome version 4.05
LHC@home - 2004-10-09 23:48:09 - Started upload of v64lhc1000profour18s14_16543.47_1_sixvf_15468_3_0
LHC@home - 2004-10-09 23:48:14 - Finished upload of v64lhc1000profour18s14_16543.47_1_sixvf_15468_3_0
LHC@home - 2004-10-09 23:48:14 - No data transferred

> 2004-10-09 18:05:32 [SETI@home] Pausing result
> 14mr04aa.17223.14977.186078.34_2 (removed from memory)
> 2004-10-09 18:05:32 [LHC@home] Restarting result
> v64lhc1000profour18s14_16543.47_1_sixvf_15468_3 using sixtrack version 4.46
>
> fortran error message at 18:05:32 computer frooze had to reboot .
>

Sorry about that. Did you have graphics enabled (normal or screensaver) ? Which graphics card you have?

It seems that after sixtrack crashes, the OpenGL session (with key bindings and fullscreen gfx) is not always cleared by the operating system. So I think this is partly windows/GFX driver problem. Of course that fortran error is problem of ours.

Markku Degerholm
LHC@home Admin

I also had a crash / fortran error. This happened after I clicked "OK" on that error message:

LHC@home - 2004-10-10 15:37:19 - Started upload of v64lhc1000profour20s10_12563_1_sixvf_25675_0_0
LHC@home - 2004-10-10 15:37:22 - Finished upload of v64lhc1000profour20s10_12563_1_sixvf_25675_0_0
LHC@home - 2004-10-10 15:37:22 - No data transferred

I wonder why so many scientists keep sticked to such an old and inflexible programming language... :-/

http://www.ibiblio.org/pub/languages/fortran/unfp.html

> http://www.ibiblio.org/pub/languages/fortran/unfp.html

Thanks for that link, Guido, but I know about Fortran and the 2 revisions since Fortran 77...
But it still misses features like OOP, for example. IMHO mandatory things that have been included in C++ many years ago.

gute wissenschaftler in ihren Fachgebieten,aber bei den boinc projekten merkt man das programmierung nicht von grund auf gelernt wurde,was sich jetzt bitter r�cht egal welches boinc projekt man nimmt !
http://www.fs.fed.us/gpnf/volcanocams/msh/images/mshvolcanocam.jpg

> gute wissenschaftler in ihren Fachgebieten,aber bei den boinc projekten merkt
> man das programmierung nicht von grund auf gelernt wurde,was sich jetzt bitter
> r�cht egal welches boinc projekt man nimmt !

Can you please translate?

Markku Degerholm
LHC@home Admin

>
> I wonder why so many scientists keep sticked to such an old and inflexible
> programming language... :-/

Fortran is pretty effective for calculations, because the code can be optimized by the compiler pretty well.

Of course, one thing is that there are so much reusable existing scientific code done with fortran that it is often easiest to continue using fortran.


Markku Degerholm
LHC@home Admin

gute wissenschaftler in ihren Fachgebieten,aber bei den boinc projekten merkt
> man das programmierung nicht von grund auf gelernt wurde,was sich jetzt bitter
> r�cht egal welches boinc projekt man nimmt !


trial of translation

..good researchers/scientists in their domains,
in point of view boinc projects,
it looks like they haven't learned programming from the basics,
now the revenge is comming back, independently which project is taken/focused.


please don't beat me, as mentioned, only a try of translation..

_________________________________________________________


Hm.. in the banking sector, a lot of cobol and fortran applications are still running and running,

in the industry sector, robot automation (Automobile-), LOGO (1960+) is used as programming language very frequenctly.

In the Flugabfertigung (Airport checkin/booking) ofthen seen, 3270 mainframe (old old old)

In all those cases, the solution is in foreground, not the programming language

A programming interface (Programmiersprache) is only a way to give a set of instructions.

The brain-ware, this is the most important factor, in my eyes.

sure, every compiler has it's limitaions.


We can "create" a good documentation, using word or amipro or TexAS or open Office or what else, not the tool, the result is in foreground.

The need, the demand and HOW it's solved, this is important.

> Fortran is pretty effective for calculations, because the code can be
> optimized by the compiler pretty well.
>
> Of course, one thing is that there are so much reusable existing scientific
> code done with fortran that it is often easiest to continue using fortran.

OK, I agree that it is for sure a benefit to have lots of already existing code, but I strongly doubt that the Fortran compilers are better in code optimizing than for example the GNU C++ compiler...

Have you thought about adding some kind of "debug log" to the sixtrack code? Maybe this could make bug tracing a bit easier...