| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: Website Support
Thread: no task available |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 10
|
|
| Author |
|
|
IT022906
Cruncher Joined: Feb 4, 2005 Post Count: 16 Status: Offline Project Badges: 
|
it is more than 24 h. that no task available.....why?
 |
|||
|
||||
|
BobbyB
Veteran Cruncher Canada Joined: Apr 25, 2020 Post Count: 603 Status: Offline Project Badges: 
|
I just put in a refreshed unit and getting none either but my other machines are getting their fill. Try manual update using BoincMgr. Try it often enough.
----------------------------------------You could try another projects like Rosetta. Their WUs are about 7 hours and they also go dry at times.as does Denis. Here is what I did. All other projects are at 0 resource share and I left WCG at 100 (project weight). Now, when the backups pull in work, they only get as many as the cores you have. WCG should still ping for work periodically, and because they have precedence, work should come it it's there. Luck of the draw?? This may not work for you depending on your situation: power of the machine, 24/7, etc, etc. I have a few more tweaks for when WCG is dry or drying. Edited: Just checked. It seems I was wrong. I have not been receiving work but just eating my cache. [Edit 2 times, last edit by BobbyB at Mar 30, 2024 4:29:30 PM] |
||
|
|
Mike.Gibson
Ace Cruncher England Joined: Aug 23, 2007 Post Count: 12146 Status: Offline Project Badges: 
|
Last units received Thursday.
|
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7581 Status: Recently Active Project Badges: 
|
Last unit received 2024-03-29 15:37:52 UTC
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 873 Status: Offline Project Badges: 
|
I started getting new MCM1 work again around 06:45 (UTC) on 2024-03-31, and I haven't noticed another stall in supply since then.
I seem to recall mention of something they run to try to regulate the work supply, and I'm wondering if it gets confused if there are unusually high numbers of tasks out there when some systems have accumulated so much work that they can't possibly return it before deadline; I have been getting quite a lot of retries because of missed deadlines, and (if I can believe the API wingman feed) almost all the retries were because of a fairly small group of systems which usually seem to return stuff after 4 days or longer when not msising deadlines... If that has been happening, it might also explain why the tech team thought there was plenty of work available :-) Cheers - Al. |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7581 Status: Recently Active Project Badges:
|
Consider yourself lucky. The last work unit I got was 2024-03-29 15:37:30 UTC. Been dry,dry dry ever since.
----------------------------------------Cheers Edit; Work has resumed at 2024-04-01 00:42:08 UTC Cheers
Sgt. Joe
----------------------------------------*Minnesota Crunchers* [Edit 1 times, last edit by Sgt.Joe at Apr 1, 2024 1:26:22 AM] |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 873 Status: Offline Project Badges:
|
Sgt. Joe,
Glad you eventually started to get work again :-) I suspect my "luck" may be because of a combination of profile limits, restricted cores per project and small buffers! (If I can't be sure of returning work within a day, I don't want to download it!) Even the largest of my systems is unlikely to need many tasks to fulfil a request... I can't prove that it's a request size issue, of course :-) -- and I also can't explain why my Intel systems got a lot more retries during the slack period than my Ryzens did (though I think that might be down to something in the code-path for matching retries to existing host types.) Too many mysteries for my liking, but without research access to any WCG-specific code, mysteries they'll remain... :-) Cheers - Al. |
||
|
|
Grumpy Swede
Master Cruncher Svíþjóð Joined: Apr 10, 2020 Post Count: 2092 Status: Offline Project Badges: 
|
My turnaround time is only hours, at most 12 hours with MCM, but when it comes to OPNG tasks, the turnaround time even with a full 50 OPNG cache, is just a few hours.
A very small cache may be the reason why I usually get tasks, even when others don't get anything. If there are only resends available, I often get lots of those. |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1932 Status: Offline Project Badges:
|
My turnaround time is only hours, at most 12 hours with MCM, but when it comes to OPNG tasks, the turnaround time even with a full 50 OPNG cache, is just a few hours. Also running just standard cache, and while there seems to be a serious slowdown early Saturday, last night both MCM1 and OPNG came in plenty. Now let's see if this holds until all by backup WUs are flushed out, I hate aborting any project...A very small cache may be the reason why I usually get tasks, even when others don't get anything. If there are only resends available, I often get lots of those. Ralf  |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7581 Status: Recently Active Project Badges:
|
I suspect my "luck" may be because of a combination of profile limits, restricted cores per project and small buffers! My queues are small enough where I turn around tasks within 24 hours. There is some difference in how quickly the Windows systems gets tasks versus the Linux systems when the supply is restricted. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
|