| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
arcarter
Cruncher Joined: Jun 10, 2014 Post Count: 4 Status: Offline Project Badges:
|
I have not received any new tasks since August 11. I have checked everything i can think of but cannot see any reason why they stopped being delivered.
----------------------------------------looking at the log I get this message 2020-08-22 7:59:54 PM | World Community Grid | Sending scheduler request: Requested by user. 2020-08-22 7:59:54 PM | World Community Grid | Not requesting tasks: don't need (CPU: not highest priority project; AMD/ATI GPU: ) Any help would be appreciated. Thanks Anthony [Edit 1 times, last edit by arcarter at Aug 23, 2020 12:03:31 AM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7846 Status: Offline Project Badges:
|
Are you running any other projects besides WCG ?
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
arcarter
Cruncher Joined: Jun 10, 2014 Post Count: 4 Status: Offline Project Badges:
|
Yes, two other projects Rosetta@home and Einstein@home. Resource share is set to 100.
Everything was working until Aug. 11th and then the WCG tasks just stopped, the others are working fine. |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7846 Status: Offline Project Badges:
|
Try suspending the other two projects and when WCG kicks back in, gradually adjust the resource share on the other two projects until you get the mix you desire.
----------------------------------------Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
arcarter
Cruncher Joined: Jun 10, 2014 Post Count: 4 Status: Offline Project Badges:
|
That worked.
Thanks Don't understand why it stopped sending work, no changes were made and it had been working with project set to 100 share for years. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Did you add Rosetta/Einstein round about August 11? It's when BOINC starts recalculating RAC/REC and will give the added projects a catch up. Eventually WCG should have returned to the mix.
Note that BOINC does block fetches, not a little bit of this, a little bit of that. If work is fetched, it asks the project with the momentary highest priority on the fetch count and backfills the buffer and will keep doing that until it's the turn of the next. |
||
|
|
arcarter
Cruncher Joined: Jun 10, 2014 Post Count: 4 Status: Offline Project Badges:
|
No they have all been running together for a long time, I was also running the seti project until they shut it down but I have that suspended.
----------------------------------------[Edit 1 times, last edit by arcarter at Aug 23, 2020 11:59:41 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Strange, but anytime we manually intervene in the work fetch scheduling such as now suspending other projects to get work for WCG will just put the underlying balancing calculations out of whack and turn the situation from bad to worse as decribed here https://boinc.berkeley.edu/wiki/REC-based_scheduler
There's a file client_state.xml where these control <rec> and <rec_time> values are held. There used to be a command to reset all these counters, one by one, but it has not worked since the new REC (Recent estimated credit) method got implemented. This leaves only exiting(unloading) the client, open the file in a ascii text editor and setting them to zero. How long it then takes to re-balance I don't know. Could be weeks which also, to include, depends on the size of the work buffer. |
||
|
|
|