| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: An 18-day work unit? HOLY COW! |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I just checked my agent, and discovered that I'm only 5% finished with job and my computer has already spent 21.5 hours on it. My device is rated for this job at '129', so it's not as if it is just running this very slowly, but at this rate it is going to take almost 18 days to complete. I've never had one to take so long. It is for the Proteome_Folding_2 project.
Cheers. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Bill,
----------------------------------------would u be so kind to visit other threads on same... we've had a few cases in the last week on HPF2. Is it still clocking CPU time (see taskmanager) and how much? Post device ID and estimated time of retrieval, which is likely equal to the last time reported in your My Device Stats.... then give the cow the boot by killing the wcg_xxxxxxx.exe process. Should get u a fresh job. EDIT: Seeing Uplinger's post below, here is a message advising the version numbers that were valid as of Nov.20 http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=9985 cheers PS.... a holy cow..... is that when believe 'rocked'?
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Nov 22, 2006 4:23:03 PM] |
||
|
|
uplinger
Former World Community Grid Tech Joined: May 23, 2005 Post Count: 3952 Status: Offline Project Badges: 
|
Bill,
Can you please check the version number of the application you have first before aborting it? It should be in the top right corner of the graphics. -Uplinger |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Bill, Version 5.1.1.0.Can you please check the version number of the application you have first before aborting it? It should be in the top right corner of the graphics. -Uplinger I had to restart my computer since my original post, and IIRC, when the UD agent came back up, it indicated that it had zero execution time and zero percent completed, so I don't know if it has now retrieved a new work unit, or if I checked before it had fully started and not yet loaded whatever it had saved. Anyway, right now it says: 45 hours and 19 minutes and still only 5% completed. But I think perhaps I know what the problem is. I am running the BOINC agent simultaneously with the UD agent. The reason is that I was having trouble with BOINC. It would complete a job and then it was not communicating with WCG to exchange work, and so a couple of times I found that my computer was not crunching at all. So I figured that I'd run both of them, letting them split their work in whatever way, and if BOINC stalled, then at least I'd have UD running full speed at that point. However, it does appear that BOINC is now running okay, so I plan to abort the UD agent and not restart it, but wanted to provide this info and complete any discussion before doing so. If the UD agent is being given second priority and will take forever to complete a work unit, then I imagine I'm just wasting that little bit of power because I'll probably miss the report deadlines. Any thoughts? |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Bill, would u be so kind to visit other threads on same... we've had a few cases in the last week on HPF2. Is it still clocking CPU time (see taskmanager) and how much? Post device ID and estimated time of retrieval, which is likely equal to the last time reported in your My Device Stats.... then give the cow the boot by killing the wcg_xxxxxxx.exe process. Should get u a fresh job. EDIT: Seeing Uplinger's post below, here is a message advising the version numbers that were valid as of Nov.20 http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=9985 cheers PS.... a holy cow..... is that when believe 'rocked'?  Anyway, I'll check the other threads in a minute, but to respond right now, I checked your link and I'm running the correct version. I'm not worrying about getting a fresh job if I am properly running a legitimate task, no matter how long it takes. On the other hand, see my reply to Uplinger's post that I thought might explain this ... but now I'm not sure because I have just checked taskmanager per your suggestion, and there definitely seems to be something wrong.First, I have both BOINC (Ver. 5.4.11) -- which is displaying that it is still current accumulating CPU time and making progress toward completion -- running simultaneously with the UD agent which is SUPPOSED to be running Proteome_Folding_2 -- version 5.1.1.0 -- BUT my taskmanager shows that BOINC is using 0 percent of my cycles, and that "wcg_faah_autodock_5.28_windows_intelx86" is using the bulk of my cycles -- up to 95%. I would presume that the "...faah..." means "Fights AIDS At Home", but my UD manager says I'm doing Proteome_Folding_2. You asked for my Device ID -- it is 413566 -- but I don't know how to check time of retrieval but as a rough estimate, my computer hasn't been off and it's been running this for almost 46 hours, so I'd say that it started this at approximately 1:15 p.m. CST on Wednesday, Nov.22. I am planning to terminate UD and just run BOINC, however I will wait awhile to see if anyone can explain this or needs me to check anything. Cheers. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Glad u posted that.... was hesitant, but i was drawing back from responding that HPF2 is excessively overly courteous to absolutely anything even to processes running at same priority. Experimented in past where absolutely no add-ons or tweak or throttles would plainly give 50/50..... not anymore.
----------------------------------------BOiNC.exe, UD.exe, UD_9xxxx.exe of the BOiNCmgr.exe should not use CPU time or just a little if the graphics are on screen. It's the wcg_xxxxx processes (u should have 2 with 2 agents at minimum on a single core machine, one with hpf2 in it and one with faah). Those should get the idle time..... 90+ percent normally and split equally without throttle features active. It's sheer by speculation and deduction that i came to this view.... bang me on the head with some d-oh solution...... anyone?
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks for the prompt reply, Sekerob. I have checked and you are correct that BOiNC.exe, UD.exe, UD_9xxxx.exe, and BOiNCmgr.exe are not using any cycles. You are also correct that I have two wcg_xxxxx processes with one with faah and the other with hpf2, but neither are currently using any cycles because BONIC is now running a SETI program. Don't worry; I'm not giving that any priority, but just joined so that I can help spread word about my 2plus2is4 project.
Cheers. |
||
|
|
|