| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: FightAIDS@Home
Thread: can't get any more FAAH work units |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 19
|
|
| Author |
|
|
logaritse
Advanced Cruncher Indonesia Joined: Apr 23, 2008 Post Count: 104 Status: Offline Project Badges: 
|
Wooho i'm getting FAAH Wu :) *lol* i think its matter of time *may be FAAH Wu need to be loaded properly* :D
----------------------------------------
Simplicity meet classic
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Actually, there was a note in the back-room that someone's larger shared file system got corrupted after which there was a mountain of bad results returned for a longer period. There was a manual intervention when this was discovered so all those clients were set to 1 WU per day until good work is returned. The knock-on of the repair jobs pile up was that the distributor, by design, only sent these to 'fast/reliable' clients... hence the hiccup and regular, but not high speed returners seeing a temp 'no work...'
----------------------------------------Happens once in a blue moon, that's why the "if there is not work for my fav project... give me something else"' option remains to be a recommended field to select in the My Projects/Device Profile. Thus, if you crunch, make sure to keep that buffer relatively short to remain in the 'fast/reliable' pool. Max 2-3 days cache would keep you there. Sorry for the inconvenience.
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Oct 3, 2009 5:20:10 PM] |
||
|
|
dvlrnr
Advanced Cruncher Estonia Joined: Nov 23, 2008 Post Count: 64 Status: Offline Project Badges: 
|
Thanks for clearing that up, Sekerob. I've also had a lot of Rice and HPF WUs coming through as optionals when FAAH hasn't been available. I've set a shorter buffer now, keeping the "other projects" box ticked, of course.
----------------------------------------
I love the smell of well spent clock cycles in the morning..
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hey there
"Thus, if you crunch, make sure to keep that buffer relatively short to remain in the 'fast/reliable' pool. Max 2-3 days cache would keep you there. " Can you tell me how exactly I do this? Thanks. |
||
|
|
dvlrnr
Advanced Cruncher Estonia Joined: Nov 23, 2008 Post Count: 64 Status: Offline Project Badges:
|
Hey there "Thus, if you crunch, make sure to keep that buffer relatively short to remain in the 'fast/reliable' pool. Max 2-3 days cache would keep you there. " Can you tell me how exactly I do this? Thanks. In your profile (My Grid), go through Device Manager -> Selected Profile (Default) -> Custom Profile -> Workunit Cache Settings -> Cache N.N extra days of work. Set N.N to a number of your choosing, with 2.0 - 3.0 having been recommended above. As for myself, I've been running a cache of 1.0 days lately. However, I seem to again be experience problems filling the buffer with a FAAH/HFCC mix. While returning WUs every few hours, I haven't picked up a single WU over the last 24 hours.
I love the smell of well spent clock cycles in the morning..
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Could be you had a few long ones that inflated the Duration Correction Factor which is slow to adjust down.
----------------------------------------I've got a duo client exclusively on HFCC and pulled a new job just minutes ago. The message log is free of any 'no work for...'. The jobs' been varying between 6 hours and 11 hours... yes really... so sometimes have with 1.0 days buffer setting 7-8 in the queue, sometimes just 5. If you have 8 in buffer when it was tuned to 6 hours and suddenly they count as 11 hours, the buffer in my case goes from 48 hours to 80 hours size. To check, visit the client_state.xml file (read only please), and find timestats and the world community grid section "duration_correction_factor". If for instance 2.0 you'll have the cause quite surely. With the up and coming 6.10 client you'll be able to check these values from within the manager.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
YEAH ME TOO WHATS UP?
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
With all that's been said in this thread so far, nothing is up much. Production is fully normal, the number or results validated totally normal. Got some message log with some diagnostic flags activated so we can do more than a wild 'what's up' stab?
----------------------------------------thanks All capital posts are btw in nettison language considered shouting... I'm very sensitive to that ;D
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I'm running FAAH only for a few days, I was on 'No new tasks' since I had a kinda of long queue recently, 8 days, and I was about to finish my WU queue so allowed new tasks to fill up my 5-day queue just a minute ago and it downloaded 3 new WUs just fine.
Good luck. Cheers. |
||
|
|
|