| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Server Software Upgrade issues, Mar-6-2012 ~14:30 UTC - 21:00 UTC |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 177
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
....yop, it will take sometime to settle down....
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks. That would be consequential to switching off the ''repair/reliability" rating until there is enough statistics. Till then, results go to any device without considering their current history of return time... all was clean slate after the upgrade, except your My Grid statistics and the assigned results ;>)
--//-- |
||
|
|
sk..
Master Cruncher http://s17.rimg.info/ccb5d62bd3e856cc0d1df9b0ee2f7f6a.gif Joined: Mar 22, 2007 Post Count: 2324 Status: Offline Project Badges: 
|
Slate left untouched; didn't see any mid day stats update. Graphs also look the same as 12h+ ago.
"Statistics Last Updated: 08/03/12 12:06:02 (UTC) [2 hour(s) ago]" Had tasks validate this morning, but no update to any My Grid stats. |
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
Mine look fine. All updated. May help to flush your browser cache.
---------------------------------------- |
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges: 
|
I wasn't referring to probation before being allowed to do zero redundancy tasks alone but host_app_version.scale_probation which relates to when credit scaling is allowed for a host. The per-application scaling of the run-time estimates happens after 10 validated tasks, and if not mistaken they're using the same limit for the credit-system. In case of multiple plan-classes like example 32-bit and 64-bit the test for which application is fastest is also done after 10 validated tasks for each application-plan-class.  "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
kashie
Cruncher Joined: Mar 7, 2007 Post Count: 46 Status: Offline Project Badges: 
|
Thanks for the clarification. I no longer think the reduction in claimed credit was due to probation. The speed and amount of the reduction misled me to look for alternative reasons.
|
||
|
|
Ingleside
Veteran Cruncher Norway Joined: Nov 19, 2005 Post Count: 974 Status: Offline Project Badges:
|
Was surprised didn't get any CEP2-work at all today, so changed to only CEP2, and got this message:
----------------------------------------08.03.2012 20:32:29 | World Community Grid | Message from server: The Clean Energy Project - Phase 2 requires 63.48 KB/sec download bandwidth. Your computer has been measured at 61.38 KB/sec. This with a profile set to get 4 CEP2 at once. Just in case it was a problem with wrong preferences being used, changed to "Unlimited", and got these messages: 08.03.2012 20:51:15 | World Community Grid | Message from server: The Clean Energy Project - Phase 2 requires 63.48 KB/sec download bandwidth. Your computer has been measured at 61.38 KB/sec. 08.03.2012 20:51:15 | World Community Grid | General prefs: from World Community Grid (last modified 08-Mar-2012 20:50:41) So, this is 2 bugs: 1: CEP2 doesn't by-pass the bandwidth-limits as it should then selected to have more than 1 CEP2-task at once. 2: The project-specific CEP2-preference is incorrectly treated as a global BOINC-preference. "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
A couple of updates here on what we are working on:
1) We still have a script running that is boosting the number of results_per_day manually for everyone. This is problematic as it is allowing computers that only return errors to run unchecked. This is the next problem we will work on. 2) There is a continuing large backlog of results that have a hr class assigned, this is making distribution of work somewhat problematic. This may be due to #1 above. 3) The distributions stats for credit computations do not appear to always be updating properly (these are the stats that are used in the credit computations) 4) The load on the database remains very high compared to before the change. We have made some changes to improve this but more work is needed. In addition, the items reported in this thread (such as Ingleside above) and elsewhere in the forums are also being added to the queue to resolve where possible. Please continue to post observations and comments. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Would it ease things to reduce the project count temporarily or only send out resends for some projects?
Might allow you to move away from temp fix/script 1). Then you could add the projects back one at a time. |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
The last two results downloaded for HCC began with estimated run times of ~25 days. The finished in just over one hour. I noticed this on two different machines. It doesn't negatively affect me but I can see it playing havoc with some machines.
---------------------------------------- Distributed computing volunteer since September 27, 2000 |
||
|
|
|