| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Sorted by 'Sent' time, first copy went out on November 27, today me client finally met with a wingman who rushed through copy number 7 on December 15, within 10 hours from receipt [mine took 12 hours to be returned]. Over 17 days to get this one complete, and you wonder, what happened. Maybe the techs would want to look on reversing the by and large diluted 'reliability' rules. Just plain too long for my taste.
----------------------------------------FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 6-- 706 Valid 12/14/13 18:30:38 12/15/13 04:13:16 1.47 42.1 / 41.3 FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 5-- - No Reply 12/11/13 06:30:36 12/14/13 18:30:36 0.00 0.0 / 0.0 FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 4-- - No Reply 12/7/13 18:30:24 12/11/13 06:30:24 0.00 0.0 / 0.0 FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 3-- 706 Valid 12/7/13 12:49:48 12/8/13 00:23:12 1.25 40.6 / 41.3 < moi FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 2-- - No Reply 11/27/13 18:30:15 12/7/13 18:30:15 0.00 0.0 / 0.0 FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 1-- - Detached 11/27/13 12:49:33 11/27/13 18:29:44 0.00 0.0 / 0.0 FAHV_ x3AVLbINfbB_ 0335590_ 0323_ 0-- - No Reply 11/27/13 12:49:23 12/7/13 12:49:23 0.00 0.0 / 0.0 edit: The ? in the title ;O [Edit 1 times, last edit by Former Member at Dec 15, 2013 11:21:55 AM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
Something goofy is going on and I haven't had a chance to look at it yet. Since so much of this is tied up in the server code that I'm currently working on updating, I don't want to spend a lot of time investigating until it is done.
|
||
|
|
Movieman
Veteran Cruncher Joined: Sep 9, 2006 Post Count: 1042 Status: Offline |
Couldn't this just well be withing the realm of possibility of you getting a WU, it being sent to someone else, them shutting down for whatever reason, then it goes to someone else, they also have an issue, bad drive or..then on and on..I agree not probable but very possible..I've had WU that didn't validate for a month plus and always assumed that is what happenned.
----------------------------------------![]() |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Kevin,
Re Is http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,36060 that 'something goofy going on' thing? Would they [those who for instance rent EC2 time in interrupted blocks] maybe have need of an immediate reporting enabling? [YKWIM ;] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
Yes - in fact if they were to look at http://boinc.berkeley.edu/wiki/Client_configuration
They would find it useful to enable fetch_minimal_work report_results_immediately |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
If you set fetch_minimal_work, you limit a device to just 1 job, so not a good idea me thinketh [tested this and discussed with Dr.A on alpha mail list]. If WCG were to be set default with 0.00 Min buffer setting, any task completed would be reported almost immediately, making the second option superfluous.
----------------------------------------From Wiki: <fetch_minimal_work>0|1</fetch_minimal_work> Read a bit odd since sometimes 'device' is inferred as being 1 processor [a thread] Client 7 has the good habit of fetching a new task just a minute or so before the previous finishes, so technically never an idle moment [except now of course with the 'assigned to other platforms' abundance. [Edit 1 times, last edit by Former Member at Dec 18, 2013 10:35:33 PM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
You are right - I misread. Thought it meant one per core. It has been awhile since I've used/tested that setting.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
BYW, I like this one more:
<fetch_on_update>0|1</fetch_on_update> When updating a project, request work even if not highest priority project. List-add.pngNew in 7.0.54 Essentially if a wasted connect is initiated, to clear 'ready to reports' [always a topic of discussion], it will regardless of a 'don't need' state request a minimum amount [more] work... If you do this often enough you can have quite a pile up [done it], which then postpones the next workfetch and thus makes for more build-up of Ready to Report and so on... a vicious circle of stimulating members to think things are wrong [we had one report again today]. And so on. |
||
|
|
|