| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Human Proteome Folding - Phase 2
Thread: A few unusual HPF2 work units |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 129
|
|
| Author |
|
|
teletran
Senior Cruncher Joined: Jul 27, 2005 Post Count: 378 Status: Offline |
Teletran u should be  ....all 3 give points long as the HPF2 is in the bug fix period......whats questionable is where six (6), the max turn in on an 'error only' quorum result, all running dead on the same point in the WU.....grants no CPU time or canonical credit. At least i'd expect the CPU time used. As for points recognition the lowest of quorum .....oh well, when i smell out 4 with 2 open 'in progress' i just hit the abort button. Why would 5 and 6 succeed.....a statistical non-event on the many of these already seen....the 2 i saw mentioned above did exactly as per the RickH predictions.Sekerob, I'm not too concerned with points, just want all the work I do to be valuable (and not just in terms of fixing the bugs, though I realize that is necessary first). Just hoping things get worked out soon, since this is my main project of interest here. In the meantime, I'm running cancer units without a hitch. I'm glad people are keeping this thread going and helping with the problem. I'll keep checking it and as soon as HPF2 is running smooth, I'm back :) |
|||
|
||||
|
BobCat13
Senior Cruncher Joined: Oct 29, 2005 Post Count: 295 Status: Offline Project Badges: 
|
BOINC User 114748, Host ID 40048
WinXP SP2 za093_00189_10 WU checkpoints to 65.591% then no further progress. CPU running at 100%. Checkpoints were between 5-10 minutes up to 65.591%, but no checkpoint for last 8+ hours. |
||
|
|
boulmontjj
Senior Cruncher France Joined: Nov 17, 2004 Post Count: 317 Status: Offline Project Badges:
|
I have some problems like that actually.
----------------------------------------I cancelled 2 WU under boinc because the CPU was 100% used but no progression and the dead line was passed. I still have 2 WU with WCG agent that does the same thing but for those, i don't know how to cancel them. So i think i will reinstall WCG agent to restart with a new WU and forget those 2 that just make are worming my CPU for nothing. The name of devices concerned by that WCG agent problem are Jean-Jacques and jj-at-work if this can help you to find out the WU that cause problem. The 2 WU that i cancelled under Boinc are za092_ 00450 on Papa unit and za075_ 00421 on Remy unit. For that last one, stop to send it please to members because there are 11 that have not cancelled it yet and are running there PC for nuts and 4 members that have already cancelled the unit. ---------------------------------------- [Edit 1 times, last edit by boulmontjj at Jul 21, 2006 12:01:43 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello boulmontjj,
To terminate a HPF2 unit running under the UD client, right click at the bottom of the screen and select Task Manager. Then select wcg_hpf2_rosetta. Then click on End Task. You will then get a new work unit. Lawrence |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
|
||
|
|
boulmontjj
Senior Cruncher France Joined: Nov 17, 2004 Post Count: 317 Status: Offline Project Badges:
|
Ok for the methode to cancel a WU under WCG agent lawrencehardin.
----------------------------------------I will do it on monday at work because it runs for nearly 2 weeks for nothing. And i will do it today on my personnal computer at home. Have a good week-end. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
mr hardin,
my machine specs. 2 opteron 244's 1.5 gb of ram win xp pro boinc version 5.4.9, not optimized hpf2 version 5.07 boinc account # 239921 the workunits in question za075_00421_14 105+ hours stuck at 42.105%, past it's deadline of 7/21 za120_01243_2 43+ hours at 0%, deadline of 7/27 these are the only 2 units i have ever have go bad on this machine, even when it was running rosetta @ home and we were having a lot of the 1% bugs. i'm going to terminate them both |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Good deal, vavega!
I have heard some more info about the debugging. It seems that changing the program (by adding a debug section, for example) changes the behavior. Work units that end in errors suddenly start working while other work units that all ran well suddenly start failing. So we may start issuing a special debug version of Rosetta and run some work units on the grid to locate the units that run into the bug. But first, we'll bang away on the staff computers for a while, to see if we can avoid using the grid for debugging. So terminate any HPF2 work units that have not been able to process. Lawrence |
||
|
|
DanNorthDE
Cruncher Joined: Dec 7, 2005 Post Count: 4 Status: Offline Project Badges: 
|
My HPF2 work unit progress bar will not go beyond 17.3 percent and when it restarts with a new log on, it starts over at 15 % and the same amout of hours listed when it was at 15 %. So this unit will never complete unless maybe it ends in one very long session lasting days.
The Agent version is 3.0 (2844), the Device name is North, the device ID is 252251. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi DanNorthDE,
The HPF2 project has been temporarily put on hold. You should terminate the WU in progress on your computer. To terminate an HPF2 unit running under the UD client, right click at the bottom of the screen and select Task Manager. Then select wcg_hpf2_rosetta. Then click on End Task. If you are not signed up to receive work from other WCG projects such as FAAH and HDC then you will not receive a new work unit at this time. |
||
|
|
|