| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Completed Research
Forum: Computing for Sustainable Water Forum
Thread: Computing for Sustainable Water Problems Thread |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 254
|
|
| Author |
|
|
deltavee
Ace Cruncher Texas Hill Country Joined: Nov 17, 2004 Post Count: 4842 Status: Offline Project Badges: 
|
All three repair jobs I got after the validator problems turned invalid. As a result my computer became unreliable and all single redundancy WUs I returned afterwards turned inconclusive. All of those inconclusives turned invalid as soon as the second task was returned. Because of the new unreliability I received no further WUs with single redundancy, so there were no more inconclusives as well. And without single redundancy the validation seemed to work better - there were no more invalids. Don't know what would happen if the machine became reliable again because I deselected the project... The same thing happened to me. But I haven't gotten any invalids for two days and all my machines are reliable again. The new WUs are all returning as valid.  |
||
|
|
Hypernova
Master Cruncher Audaces Fortuna Juvat ! Vaud - Switzerland Joined: Dec 16, 2008 Post Count: 1908 Status: Offline Project Badges: 
|
I have gotten a good number of Invalids but they are granted points. About 50% of what is claimed.
----------------------------------------At the moment my machines are set so as to receive only CFSW WU's. But it seems they are not enough available as I receive a majority of WU's from a mix of other projects. Is there an issue with WU's production or maybe with the capability to manage the result.   |
||
|
|
petehardy
Senior Cruncher USA Joined: May 4, 2007 Post Count: 318 Status: Offline Project Badges: 
|
The fault seems to be in the "Quorum 1" mechanism.
----------------------------------------I don't remember that being tested during the betas.  "Patience is a virtue", I can't wait to learn it! |
||
|
|
petehardy
Senior Cruncher USA Joined: May 4, 2007 Post Count: 318 Status: Offline Project Badges:
|
The experienced by now should know that it's folly to jump all cores on a new science release. "Fools(and Badge Hunters) rush in where angels fear to tread" But we can rush out again! "Patience is a virtue", I can't wait to learn it! |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges:
|
I found a correlation; as soon as I stopped crunching SWater I stopped getting invalids, inconclusivies, errors and single quorum PV's.
----------------------------------------   |
||
|
|
Jack007
Master Cruncher CANADA Joined: Feb 25, 2005 Post Count: 1604 Status: Offline Project Badges: 
|
Well, I'll leave it on,
----------------------------------------they will learn from our mistakes, and that's the whole point of it isn't it. Besides, it's the only blue badge I don't have   |
||
|
|
mikey
Veteran Cruncher Joined: May 10, 2009 Post Count: 821 Status: Offline Project Badges:
|
I found a correlation; as soon as I stopped crunching SWater I stopped getting invalids, inconclusivies, errors and single quorum PV's. I have had NO Inconclusive's since yesterday, knocking on wood! Invalids either!!!! Valids though are coming in by the BUNCHES!!! I am sharing time with Poem for now, but will then go 'whole hog' for CFSW for a little bit.   |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
I'm trying to look into this issue but here is what I see for results returned in the past 24 hours:
nanoprobe: 16 Pending Validation Dataman: 39 Pending Validation Overall: 18587 Pending Validation (26.9%) |
||
|
|
Dataman
Ace Cruncher Joined: Nov 16, 2004 Post Count: 4865 Status: Offline Project Badges:
|
Dataman: 39 Pending Validation 82 Valid That would be because I turned SWater off over 24 hours ago. The PV's are eventually validating. The inconclusives/invalids are constant at about 60 each. The rest of what you see is Leish and Schisto. Not trying to start a controversy; just not going to pay for invalids.  [Edit 1 times, last edit by Dataman at Apr 23, 2012 3:06:13 PM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
We tracked down the issue. Here is what was happening:
----------------------------------------The quorum and number of copies sent for this project is set to 1. When that first copy is assigned to device, the device is checked to see if it has proven itself for that project. If it has not, then the quorum and the number of copies sent is set to 2 for the workunit and another copy is sent out. This case worked correctly. If a device had proven itself for a project (and the quorum and number of copies remained at 1), then when the result is returned and validation is run for the workunit two things are checked. 1) If the device has become 'unproven' for the project and 2) A random chance of being double checked. If either of these were true, then the result is set to inconclusive and number of copies is set to 2 (however, the quorum is left at 1). The case where a device remains proven and the result is not randomly selected for double checking worked correctly. However, the case where a device either becomes unproven or is randomly selected for double checking did not work correctly. The error resulted in valid results being marked invalid. This error was corrected with the changes that were released yesterday (mid-afternoon US time so around 20:00 UTC). I apologize for not recognizing that it was in some cases marking results invalid. The validator itself was exiting when it attempted to process many of workunits in this case so I focused on that. However, I missed that in some cases it was marking valid results invalid as well. The good news is that this is fixed. [Edit 3 times, last edit by knreed at Apr 23, 2012 3:06:15 PM] |
||
|
|
|