| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Work buffer - non existent |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 21
|
|
| Author |
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges: 
|
Hey guys,
recently, I noticed a massive drop in points. Checking my computers, I have found that my computers are storing very little work at the moment. 8 thread machine, buffer of 4.5+1 days is running 8 + 2 tasks. 80 thread machine, buffer of 2.5 + 1 days is running 64 tasks 80 thread machine, buffer of atleast 1 day is running 64 tasks 48 thread machine, buffer of 1 + 1 days is running 48 + 2 tasks 24 thread machine, buffer of 1.5+ 1 days is running 24+2 tasks etc etc Basically, these machines have run 24/7 without issues for months, if not years. There has never been an issue with buffers or anything like that. It seems to have changed to a buffer of 2 tasks, no matter what the local settings or the profile settings. And for the 80 thread machines, limited to 64 tasks at a time Any ideas? Is there a thread that I missed? |
||
|
|
BladeD
Ace Cruncher USA Joined: Nov 17, 2004 Post Count: 28976 Status: Offline Project Badges: 
|
No, but my first thought what projects are selected?
---------------------------------------- |
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges:
|
My 'Default' profile is Zika + TB.
I did recently switch exclusively to MIP, to get a few years and get my 10yr badge. But quickly switched away from it as the points drop was horrific. |
||
|
|
BladeD
Ace Cruncher USA Joined: Nov 17, 2004 Post Count: 28976 Status: Offline Project Badges:
|
On your Device Profiles page, what are your Project Limits?
----------------------------------------You may have to add other projects.  ---------------------------------------- [Edit 1 times, last edit by BladeD at Mar 5, 2019 4:41:23 AM] |
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges:
|
I checked those and all of them are unlimited, except MIP which is set at 3 units.
Workunit cache for that profile is 2.0 days, but the local settings are 4.5+1.0 days. I am truly miffed. |
||
|
|
ca05065
Senior Cruncher Joined: Dec 4, 2007 Post Count: 325 Status: Offline Project Badges: 
|
cc_config has an option ncpus :
<ncpus>N</ncpus> Act as if there were N CPUs; e.g. to simulate 2 CPUs on a machine that has only 1. To use the number of available CPUs, set the value to -1 (was 0 which in newer clients really means zero). A few years ago I seem to remember the default maximum was 64 cpus. Larger machines had to explicitly state the number available. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
You can go the boinc working directory and look in the sched_request_www.worldcommunitygrid.org.xml file and find <work_buf_min_days> and <work_buf_additional_days> to see if what is being sent to the server is what you expect it to be. If not, then it is picking up information from a place you are not considering like a local preferences file.
|
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges:
|
ca05065 - For my windows machines, they all seem to running max threads and a buffer of 2 tasks.
My 80 thread machines are on linux mint and I never had to do the <ncpus> trick as far as I remember. It's been a while since I set them up. I have done it on windows before. Doneske - I found the buffers as you said. All but the 'school' had a minimum buffer of 2 days. Local prefs were 1.2 days. I went and changed school in my device profiles from 0 to 0.5 to see if that helps. Lets see how that goes. |
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges:
|
On one of my 8 thread machines, it ran dry and I took the opportunity to update the client to the latest version.
Still no luck. Looking through the logs, I seem to have reached some kind of limit. 08/03/2019 08:47:44 | World Community Grid | Sending scheduler request: To fetch work. 08/03/2019 08:47:44 | World Community Grid | Requesting new tasks for CPU and Intel GPU 08/03/2019 08:47:45 | | Suspending GPU computation - computer is in use 08/03/2019 08:47:47 | | Resuming GPU computation 08/03/2019 08:47:47 | World Community Grid | Scheduler request completed: got 0 new tasks 08/03/2019 08:47:47 | World Community Grid | No tasks sent 08/03/2019 08:47:47 | World Community Grid | No tasks are available for OpenZika 08/03/2019 08:47:47 | World Community Grid | No tasks are available for Help Stop TB 08/03/2019 08:47:47 | World Community Grid | This computer has reached a limit on tasks in progress 08/03/2019 08:47:47 | World Community Grid | Project has no tasks available |
||
|
|
Byteball_730a2960
Senior Cruncher Joined: Oct 29, 2010 Post Count: 318 Status: Offline Project Badges:
|
Hey there guys,
So...in 4 months, I still don't have a buffer of more than 2 WUs on any machine and my 80 thread machines are only running 64 threads. I have double checked my local preferences and my global preferences for all 4 profiles and can find nothing that would indicate why I have been suddenly restricted. Is there anyway of getting hold of an admin to see if something has changed at a deeper level? |
||
|
|
|