| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Beta Test Clean Energy Type A / B version 6.29 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 170
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Another 'Error: 2' failure here to add to the others reported.
Result Log <core_client_version>6.2.18</core_client_version> <![CDATA[ <stderr_txt> Calling initGraphics() INFO: No state to restore. Start from the beginning. [ERROR] Failed to open either source or destination files while copying wcgrestart.rst to ../../projects/www.worldcommunitygrid.org/BETA_E000338_275A_00251600d_4_3. Error: 2 called boinc_finish </stderr_txt> ]]> Workunit Status: BETA_ E000338_ 275A_ 00251600d_ 5-- In Progress 15/03/09 05:14:16 16/03/09 19:33:45 0.00 0.0 / 0.0 BETA_ E000338_ 275A_ 00251600d_ 4-- Error 14/03/09 22:11:33 15/03/09 05:13:48 6.93 107.0 / 0.0 <-------- MINE BETA_ E000338_ 275A_ 00251600d_ 3-- Error 14/03/09 06:24:51 14/03/09 22:11:17 10.94 125.0 / 0.0 BETA_ E000338_ 275A_ 00251600d_ 2-- Error 12/03/09 22:20:35 14/03/09 06:24:29 7.67 127.5 / 0.0 BETA_ E000338_ 275A_ 00251600d_ 1-- Inconclusive 10/03/09 16:58:06 12/03/09 11:52:44 15.37 181.1 / 0.0 BETA_ E000338_ 275A_ 00251600d_ 0-- Error 10/03/09 16:58:03 12/03/09 22:20:11 14.90 127.6 / 0.0 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Another 'Error: 2' failure here to add to the others reported. Some of these are weird. A WU I've done successfully has failed for every other machine that's done it. I wonder if another one will ever complete it? And what happens to my result if not? Perpetually "inconclusive"? |
||
|
|
David_L6
Senior Cruncher USA Joined: Aug 24, 2006 Post Count: 296 Status: Offline Project Badges: 
|
Another 'Error: 2' failure here to add to the others reported. Some of these are weird. A WU I've done successfully has failed for every other machine that's done it. I wonder if another one will ever complete it? And what happens to my result if not? Perpetually "inconclusive"? I've had that happen and the results ended up as "Too Late".  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Why do tasks fail? The whole point of beta testing is to find out how and why tasks fail. Errors are just as valuable to the techs. Yes, this means seeing how (and if) the same task fails in the same way on all computers. They want the errors! Your aborted tasks, on the other hand, help no-one. It counts as an error, but provides no information. Now, if you don't like having errors, we respect that. Simply opt out. I see it this way: The intention of a normal project are the results of valid workunits. These are awarded with credits/points. The intention of Beta Tests should be to find errors. Therefor the time spent for crunching these WUs should be awarded regardless of error or valid. But so far I see no evidence that it happens that way. My Beta units are listed as error on the result status page (I did approx. 2 days runtime in Error WUs) and disappear from there after some time. No indication that my contribution counted in credits/points to my overall status or in runtime toward my Beta badge. Yes, I could opt out. But I would like to contribute (and get my badge  ). But it is a bit unfair that the contribution that helps this "Beta test project" most (the intended outcome: errors) is not honored in a way as it is done in the other projects (the intended outcome: valid calculations). Greetings Thorsten |
|||
|
||||
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3716 Status: Offline Project Badges: 
|
Thorsten,
----------------------------------------Your point of view is making sense. I have asked the staff to see your post and to give their position about it. Cheers. Jean. |
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
If at all credit for errors, not more than would the task have completed normally. Beta program enrollment does expect an amount of hands on. Evil me exaggerating, it's not like 72 hours later finding a task that's stuck in a loop that should have taken 3 hours. In past the techs have been very lenient granting credit manually where credit is thought appropriate, as mentioned sometimes not passing via the RS page. Making it a set rule... not convinced.
----------------------------------------As for the several times seen comment "how do I otherwise know". Well, while that question is ok, I'll trust WCG to follow true with adding something to the MG page. Those who track their detail results will know anyway. Else, not going to get into elevated discussions. Been here too long to know that pragmatism is required on this, from me inclusive.
WCG
Please help to make the Forums an enjoyable experience for All! |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thorsten, Your point of view is making sense. I have asked the staff to see your post and to give their position about it. The difference is whether the WUs end with an Error or with Invalid status. We can't really know what causes each. Since the Invalid still must yield something of value to the Beat test, it can be awarded points. If the Error WU gives back something of value, then it could be awarded points, but if it yields nothing to the Techs, then it has no value. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
As a middle way one could also consider not to grant points for error/invalid beta units, but to add the time spend on such a unit to the runtime, so they at least count towards the Beta badge (afterall, the time was spend on the Beta)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thorsten, Your point of view is making sense. I have asked the staff to see your post and to give their position about it. The difference is whether the WUs end with an Error or with Invalid status. We can't really know what causes each. Since the Invalid still must yield something of value to the Beat test, it can be awarded points. If the Error WU gives back something of value, then it could be awarded points, but if it yields nothing to the Techs, then it has no value. This is true for normal science/project work. The purpose for the Beta Test is not to produce valid calculations. Instead it should help to produce an error free science application, which later in the project will produce valid calculations and (ideally) no errors. From this point of view is every error, invalid and valid result in Beta Test helpful to the techs. It is their responsibility to stop or modify the Beta Test when they have got enough data to fix an error or find the cause for massive occurances of invalids. Another thing that caught my eye: Errors in Beta Testing should also not render a client as unreliable. When I filled my buffer with the new released HFCC WUs I noticed that all 15 were of quorum 2. Before my beta testing errors my machine was flagged as reliable. I know because I got several FAAH and DDDT with quorum 1 right before beta testing. It seems that every project which produces a high error rate (for whatever reason) will have a negative impact to efficiency of the zero redundancy projects. Greetings Thorsten Edited for grammar [Edit 1 times, last edit by Former Member at Mar 16, 2009 6:59:54 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Some of these are weird. A WU I've done successfully has failed for every other machine that's done it. I wonder if another one will ever complete it? And what happens to my result if not? Perpetually "inconclusive"? I've had that happen and the results ended up as "Too Late". I was lucky. Finally, a machine came along which made it too, even though it failed to make its deadline by some hours. BETA_ E000338_ 293A_ 00251600v_ 6-- In Progress 16/03/09 16:20:26 18/03/09 00:01:14 0.00 0.0 / 0.0 BETA_ E000338_ 293A_ 00251600v_ 5-- Valid 15/03/09 08:39:29 17/03/09 03:49:29 15.55 213.8 / 173.1 BETA_ E000338_ 293A_ 00251600v_ 4-- Error 14/03/09 23:39:23 15/03/09 08:38:55 2.87 85.2 / 0.0 BETA_ E000338_ 293A_ 00251600v_ 3-- Error 14/03/09 14:46:48 14/03/09 23:38:46 8.81 42.6 / 0.0 BETA_ E000338_ 293A_ 00251600v_ 2-- Error 14/03/09 06:09:37 14/03/09 14:46:07 8.58 46.5 / 0.0 BETA_ E000338_ 293A_ 00251600v_ 1-- Error 10/03/09 17:00:58 10/03/09 22:59:47 4.06 63.4 / 0.0 BETA_ E000338_ 293A_ 00251600v_ 0-- Valid 10/03/09 17:00:48 14/03/09 06:08:37 28.87 130.4 / 173.1 -- mine |
||
|
|
|