| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: More than one WU per core |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 17
|
|
| Author |
|
|
fufu
Cruncher Joined: Mar 20, 2006 Post Count: 46 Status: Offline Project Badges: 
|
Hi Paul,
thank you. "Connect to network about every" has the value "0.0" days "Cache" has "10.0" extra days of work |
||
|
|
fufu
Cruncher Joined: Mar 20, 2006 Post Count: 46 Status: Offline Project Badges:
|
I think it has probably to do with my WCG profile settings. I run a clean installation on a raspberry pi, too. There I have the same behaviour. I changed the local settings to get 10 days of work, but nothing happens. I have 4 active WUs and 2 additional WUs ready to crunch on the pi.
|
||
|
|
AgrFan
Senior Cruncher USA Joined: Apr 17, 2008 Post Count: 396 Status: Offline Project Badges: 
|
What are the values for these settings?
----------------------------------------Profile - Workunit Cache Settings Connect to network about every xxx days? Cache xxx extra days of work Local - Options, Computing Preferences, Computing Tab, Other section Store at least xxx days of work Store up to an additional xxx days of work Defaults for both settings are 0.2 and 0.3 days. Try clicking 'Use Web Prefs' to see if the problem still exists. It may help to know the configuation of the machine (CPU, RAM, available disk space, etc). I'm guessing you are caching new work before the upcoming outage. I would suggest only caching up to 7 days of new work to minimize the risk of missing deadlines.
[Edit 8 times, last edit by AgrFan at Feb 13, 2022 7:25:12 PM] |
||
|
|
fufu
Cruncher Joined: Mar 20, 2006 Post Count: 46 Status: Offline Project Badges:
|
My computer:
AMD Ryzen 5950X with 32GB RAM and 295GB of free disc space. I tried clicking "use web prefs" but it didn't change a thing, unfortunately. The web profile is configured like this: - every project allowed - ARP limited to 16 WUs, all other projects are set to unlimited - project weight 100.0 - connect to network all 0.0 days - cache 10 days of work My local profile is configured like this: - 100% of CPUs and 100% of CPU time - save work for at least for 10 days - save work for additional 10 days In the control settings I said it should always crunch on CPU and GPU and it should always connect to the network. |
||
|
|
PMH_UK
Veteran Cruncher UK Joined: Apr 26, 2007 Post Count: 786 Status: Offline Project Badges: 
|
Suggest "connect to network" 7 days and "cache" 3 days.
----------------------------------------As deadlines are max 7 days you won't get more (unless exposed to the app_config.xml bug that keeps fetching). Paul.
Paul.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
What is the message in the log when the scheduler request runs?
|
||
|
|
AgrFan
Senior Cruncher USA Joined: Apr 17, 2008 Post Count: 396 Status: Offline Project Badges:
|
My profile settings are 'Connect to Network' 0.2 days (default) and 'Cache' 7 days. I'm only running MCM. I have no problems getting 7 days of work.
----------------------------------------You may need to reset your web profile to the Standard settings, set Cache to 7 days and refresh your local settings by clicking the 'Use Web Prefs' button. Do you see any messages in the Event Log when downloading new work?
[Edit 9 times, last edit by AgrFan at Feb 14, 2022 2:02:59 AM] |
||
|
|
|