| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Invalid Tasks |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 18
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello nivrip,
----------------------------------------Although there is a fairly new AMD family that has problems when matching against other CPUs, the older AND chips work well. The only thing I can think of is a hardware problem of some sort. If Update does not fix it, then try memory test utilities and as a last resort see if there are any projects that it produces valid results for. Lawrence Added: I am wrong. Please ignore this. [Edit 1 times, last edit by Former Member at Jun 10, 2013 7:20:15 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
05/06/2013 23:51:21 Processor: 1 AuthenticAMD AMD Athlon(tm) XP 2000+ [Family 6 Model 6 Stepping 2]
----------------------------------------PLEASE LOOK: The Newer (Than this one) have the same problem BUT there are enough of them in use that the techs have set them up as a seperate class and when they get work their wingman will be the same type cpu and will therefore validate. this class does not have enough in use to use this method becuase it will clog the feeder waiting for a wingman to come along. it will not validate if the wingman has a newer AMD or an Intel cpu. The key to identify these is the [Family 6 Model 6 Stepping 2], they are older SINGLE core, P4 class cpus. They will validate ANY project except VINA based. NO amount of cleaning, memtest, cooling etc will fix this problem. Edit: See this thread: https://secure.worldcommunitygrid.org/forums/...ead,32481_offset,0#361498 [Edit 1 times, last edit by Former Member at Jun 10, 2013 2:30:50 AM] |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7844 Status: Offline Project Badges: 
|
Looks like fredski has hit the nail on the head.
----------------------------------------
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Been there/done that ;) still have a few, they have their own profile :))
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
What fredski said, "That would be a logical conclusion" (per Spock too). Noticed on occasions when my wingman on Zero redundant tasks checks all got returned with same timestamp i.e. luck of the draw all going to the same wingman, both having called for a larger set of work at same time. [If] it worked like that to get to the 'do it alone' state, nothing can be done, but if it ran with valid for a long time, hundreds of results, then that would have been too much of the good fortune and I'd consider a hardware/OS issue. What strategy to apply to find out is a long shot. If not limited to 1 device, apply profile selectiveness, and exclude VINA based jobs from the suspect host. Of those:
GFAM - Done DSFL - Done soon SN2S - Still long time to go. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Please read the whole thread I quoted above, if ID as family 6, it is dead (aka invalid) if you need a wingman. the odds of getting a family 6 wing with 0.12% contrib is not worth the fight. Any\every thing else works. it is not a hardwaere issue, it is not a oc/cooling issue, it is (in my book) a validator issue. This is not going to get fixed, there is no fix we crunchers can do, it is a validator problem, but as only 0.12% of the users have this problem...
----------------------------------------[Edit 1 times, last edit by Former Member at Jun 10, 2013 8:14:44 AM] |
||
|
|
nivrip
Senior Cruncher North Yorkshire Joined: Sep 13, 2007 Post Count: 285 Status: Offline Project Badges: 
|
Thanks for all the replies, fellas.
----------------------------------------At least this ancient PC (it's now 11 years old, and is only used for crunching) can continue with the non-VINA units, so I am happy with that. Having read some of the other threads on this matter I still can't work out how the PC managed to crunch all and every task received until two weeks ago. Maybe it's on its last legs. 
ЮРКШИР КРУНЧЕР
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It's a complete riddle how that came to be, 1 in 1000 chance of hitting on a matching CPU wingman.
PPC and not far after, XP may be the next platforms to be dropped. Newest clients are I think to have read, no longer tested on there, at least not a showstopper. What else to expect when MS is no longer going to do security patches for it after April 8, 2014. http://www.bbc.co.uk/news/science-environment-22728014 |
||
|
|
|