| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: GO Fight Against Malaria
Thread: x2x23's are duff (it's my machine - sorry  ) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges: 
|
you've taken me out...again
----------------------------------------that's another day on a quad core lost lost as I have reached my quota This project is turning out to be less than reliable  Could the scheduler be modified so that a different batch could be given out if you've received one back as a computation error ?  [Edit 2 times, last edit by David Autumns at Feb 15, 2012 6:18:32 PM] |
|||
|
||||
|
mikey
Veteran Cruncher Joined: May 10, 2009 Post Count: 824 Status: Offline Project Badges:
|
you've taken me out...again that's another day on a quad core lost lost as I have reached my quota This project is turning out to be less than reliable Could the scheduler be modified so that a different batch could be given out if you've received one back as a computation error ? There are MANY reason for a unit to be not crunched properly mostly it is due to our own pc's though, not the workunit itself. I KNOW you are not a rookie but have you tried rebooting the pc? Do you crunch for any other Boinc Projects? Are they working properly? What version of Boinc are you using? As far as your 'quota' goes it should improve radically as soon as you figure out what the problem is, Boinc, at the Server level, does an auto reduction in the amount of units we users can get as we return invalid units. This returns to normal very quickly as we again start returning valid units.   |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi David Autumns,
GOFAM uses VINA to crunch the work units. VINA seems to be pretty sensitive. For example, I think that AMD Socket-A cpus (Family 6) give floating point results that do not validate against other cpus. What sort of system is giving you errors? It might run other projects all right if you exclude VINA projects. Lawrence |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7846 Status: Offline Project Badges: 
|
I have done over a thousand of these and the errors are pretty rare, although I just had one the other day.
----------------------------------------GFAM_ x2x23_ TB_ ENR_ 0005363_ 0082_ 1-- 611 Error 2/7/12 16:32:42 2/9/12 01:30:03 11.79 139.8 / 0.0 </stderr_txt> <message> <file_xfer_error> <file_name>GFAM_x2x23_TB_ENR_0005363_0082_1_0</file_name> <error_code>-161</error_code> </file_xfer_error> This was on an AMD 9150e quad. Unless it is a recurring error I would not worry about it. Cheers
Sgt. Joe
----------------------------------------*Minnesota Crunchers* [Edit 2 times, last edit by Sgt.Joe at Feb 15, 2012 4:19:01 PM] |
||
|
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges:
|
Hi Lawrence
----------------------------------------Apologies it was a 4:30 in the morning diagnosis based on my previous experience with this project. It's my duff machine instead What about my reasonable suggestion on the scheduler? If you are running a single project then with "Computation Failure" after failure caused by the same batch of Work Units you are out for the count for up to 24 hours, pretty quickly - you and everyone else in the same boat. If the scheduler - on spotting multiple computation failures - jumped in and allocated WU from a different batch - you are still in the game, not just gathering idle time. Dear all sorry for any confusion Dave |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could the scheduler be modified so that a different batch could be given out if you've received one back as a computation error ? The implementation details involved may be another matter, but I see the conceptual basis of the idea as desirable.-- snip from David Autumns [Feb 15, 2012 4:27:39 AM] post P.S: As WCG evolves, I'm sure there will be some kind of a 'feedback-interaction' mechanism between the nature-and-specifics-of-the-work on one hand, and on the other hand the nature-and-specifics-of-the-machine that will do the work -- so that all adjustments/corrections, as needed, will be done autonomously and in the background. ; |
|||
|
||||
|
mgl_ALPerryman
FightAIDS@Home, GO Fight Against Malaria and OpenZika Scientist USA Joined: Aug 25, 2007 Post Count: 283 Status: Offline Project Badges: 
|
Hi David,
Your apology is accepted. To quote the Breakfast Club, "the world is an imperfect place. Screws fall out all the time." ;) I'm glad that this problem was caused by your computer, instead of being a problem with the project as a whole. Thank you very much for your interest and your support, Alex L. Perryman, Ph.D. |
|||
|
||||
|
|