| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: The Clean Energy Project - Phase 2 Forum
Thread: Windows 7 PC records Computation Error for all Clean Energy Project Phase 2 Jobs |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 24
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Are you running one clean energy project workunit at a time or multiple ones?
Run windows defender which is a part of windows 7 and see if you got any malware or third party program running in the background that installed without your knowing it. I had a program like that caused bluescreens to occur when clean energy project was running. Had to run windows defender twice to get rid of the issue. |
||
|
|
Michael2901
Veteran Cruncher Joined: Feb 6, 2009 Post Count: 586 Status: Offline Project Badges: 
|
I have tried running windows defender but found no malware.
I have been running them one at a time on my Winows 7 PC. My Vista PC's had no problem running up to 4 Clean Energy jobs simultaneously, one for each core. Incidentally it looks as if the Clean Energy Project phase 2 has also temporarily run of work as none of my PC's have been able to get any additional Clean Energy work units for about 1 week now. |
||
|
|
Carlos R. Moreira
Cruncher Brazil Joined: Sep 30, 2011 Post Count: 31 Status: Offline Project Badges: 
|
About not getting any work in your other pc's, The Clean Energy Project - Phase 2 got it's priority reduced to medium if i'm not mistaken, so, if you are running multiple projects could be the priority that's keeping your from getting CEP2 WU's.
----------------------------------------Also, in your Profile Device Manager options, you have the "Project Specific Settings" to CEP2, the "Number of workunits per host" option was resetted couple days ago with one of the updates on the projects, you might check that also... About the errors you have been experiencing, i've noticed that when processing CEP2, u have high use of the hard drive, i have a WD with 64mb cache here, no problems so far..., try to replace the hard drive of your machine and run some tests. Another thing i remembered, on Boinc manager, in your projects tab, select WCG and "Reset Project", but beware that if you do that with current jobs, those will be cancelled. Resetting the project will eliminate any problem that u might have related to corrupt files or anything related to the projects. Best Regards, Carlos R. Moreira  [Edit 1 times, last edit by Carlos R. Moreira at Dec 1, 2011 10:33:23 AM] |
||
|
|
krakatuk
Advanced Cruncher Germany Joined: Oct 3, 2008 Post Count: 141 Status: Offline Project Badges: 
|
Since CEP2 is known to have a massive memory usage, I would check the memory first.
----------------------------------------Try to remove all memory rigs except one and make a test run. If it doesn't help - replace the remaining rig by another one and try again.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi
I am no expert with computers but I had issues as well W7 64 bit. Like the suggestions above it was the memory...........nothing else caused issue but running CEP2 did. Just my pennies worth |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
About not getting any work in your other pc's, The Clean Energy Project - Phase 2 got it's priority reduced to medium if i'm not mistaken, so, if you are running multiple projects could be the priority that's keeping your from getting CEP2 WU's. Also, in your Profile Device Manager options, you have the "Project Specific Settings" to CEP2, the "Number of workunits per host" option was resetted couple days ago with one of the updates on the projects, you might check that also... About the errors you have been experiencing, i've noticed that when processing CEP2, u have high use of the hard drive, i have a WD with 64mb cache here, no problems so far..., try to replace the hard drive of your machine and run some tests. Another thing i remembered, on Boinc manager, in your projects tab, select WCG and "Reset Project", but beware that if you do that with current jobs, those will be cancelled. Resetting the project will eliminate any problem that u might have related to corrupt files or anything related to the projects. Best Regards, Carlos R. Moreira Actually, you are mistaken on your first paragraph. CEP2 is normal feeder prorioty, and to interpret some offline words by knreed, kind of on super normal priority. Set it to X, report a task and on the next work fetch [after the report clears] it should be near certain not to get CEP2 first. In fact, if you have client 6.13.12 alpha [I'm testing], you get a new CEP2 job in the same scheduler cycle as the report. More so, the fetch is executed even before the acknowledge is printed in the message log [if you've got the log flag on]. This approach is to help increase to rather miserable share of 4% in 10 active sciences. Anyone who runs unlimited, and I've seem comments to that effect, and has other sciences also selected finds that hardly anything else is getting through, i.e. If you still want some other sciences on the side, you'd have to play the 2-16 selection and cache size tuning. Mine is presently set to 4 on 1 day cache, so it get's about 30% of the daily time on my quad. @Micheal2901, work is plenty full for whoever wants it. There is zero indication in the daily production numbers that there is supply stagnation i.e. it will be something in your settings. Message log would print things like "insufficient memory for xyz... " if there were a problem or "No work available for". The test is always to set a project to exclusive and see if you can get a science you're not getting. If you do report too many fails the quota rule kicks in. You'd not get them either, but that would be printed in the message log. On memory, running OCCT, memtest86 or similar is likely to catch bad memory strips (not rigs ;-). Tearing strips out I'd not start doing until it's sure there's an issue. --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Dear krakatuk, paulbrett, and others,
Just to be clear - the memory use of CEP2 is not really THAT excessive. Each wu is limited to 512MB and if you use LAIM that adds a bit, but even 4 simultaneous wus shouldn't overwhelm a modern PC. If you are concerned about the health of your RAM - there are readily available test programs which you can run over night. Best wishes from Your Harvard CEP team |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Do not take my comment as one of issue, it was meant to help.
I have 8gb and only 4 cores but its over clocked...........it was fully stable before and after once I had slowed it down a little. All I was saying was it did something that none of the stress testing applications managed. Maybe it can be marketed as the 'ulitmate' memory stress tester |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
report a task and on the next work fetch [after the report clears] it should be near certain not to get CEP2 first. Yes, noticed this welcome improvement. When I chose to run 3 CEP2 tasks and a low cache, that's exactly what I want to run, 3, at all times. Good work  On memory usage. Well balanced computer systems should have an amount of RAM proportionate to the number of cores/threads, the speed of the hard drive, the GPU, PSU and so on. Bespoke systems tend however to vary from this norm, so some might be GPU heavy, CPU or RAM light (meida), or the opposite (servers). With ever increasing memory overheads from operating systems, programs and tools (updates) lesser and aging systems are getting exposed. Take for example a basic dual core desktop, 3 or 4 years old; the standard 2GB RAM use to be plenty, but now with the additional operating system updates and patches, suite of desktop programs and AV package, almost 2GB of memory is already in use. Doesn't leave much for CEP2 tasks! Using the 'While processor use is less than %-age' often helps prevent failures, as does keeping a very low cache of tasks, and running a known safe number of CEP2 tasks. On some systems I schedule Boinc to not run during periods that I run automatic AV scans, and other maintenance programs. |
||
|
|
Jake1402
Senior Cruncher USA Joined: Dec 30, 2005 Post Count: 181 Status: Offline Project Badges: 
|
I had the same problem on a Vista laptop. Every CEP2 wu errored out. Had to stop running CEP2 on that laptop.
----------------------------------------
Join the Chicago-IL-USA team!
2 AMD FX 8320/AMD R9 270X/Win 10 2 AMD FX 8320/AMD RX 560/Linux Mint 20.3 (both computers DOA) Intel Pentium G240/Win 10 |
||
|
|
|