| 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 |
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges: 
|
Thanks Nanoprobe. Knew I saw it somewhere but couldn't find it again about the points. With the increase in points, badges will come quicker. Hope WCG considers more badges 5, 10, etc years.
----------------------------------------
Crunching for humanity since 2007!
 |
||
|
|
Bugg
Senior Cruncher USA Joined: Nov 19, 2006 Post Count: 271 Status: Offline Project Badges: 
|
Thanks Nanoprobe. Knew I saw it somewhere but couldn't find it again about the points. With the increase in points, badges will come quicker. Hope WCG considers more badges 5, 10, etc years. How would an increase in points give faster badges? Badges are based on time spent crunching, not points earned for that time, right? Or am I missing something?  i5-12600K (3.7GHz), 32GB DDR5, Win11 64bit Home |
||
|
|
JmBoullier
Former Community Advisor Normandy - France Joined: Jan 26, 2007 Post Count: 3715 Status: Offline Project Badges:
|
What you will be missing out on too is the high credits currently granted... in three days time they'll bound to have gravitated to a lower average. Rob, It's not happening exactly the way you say it.  High credits have lasted only a few hours and the new claims are much lower than what they were before the upgrade.  See below (all results within the same benchmarking period): X0960059581455200510201243_ 0-- JmB-6600-Ub64 Valid 05/03/12 19:38 06/03/12 00:17 0,87 30,0 / 30,0 |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Thanks Nanoprobe. Knew I saw it somewhere but couldn't find it again about the points. With the increase in points, badges will come quicker. Hope WCG considers more badges 5, 10, etc years. How would an increase in points give faster badges? Badges are based on time spent crunching, not points earned for that time, right? Or am I missing something? No you're not missing something. I had to comment on another poster who was overclocking to get badges faster. It really makes me wonder how people can misunderstand things so badly.  Distributed computing volunteer since September 27, 2000 |
||
|
|
Bugg
Senior Cruncher USA Joined: Nov 19, 2006 Post Count: 271 Status: Offline Project Badges:
|
Thanks Nanoprobe. Knew I saw it somewhere but couldn't find it again about the points. With the increase in points, badges will come quicker. Hope WCG considers more badges 5, 10, etc years. How would an increase in points give faster badges? Badges are based on time spent crunching, not points earned for that time, right? Or am I missing something? No you're not missing something. I had to comment on another poster who was overclocking to get badges faster. It really makes me wonder how people can misunderstand things so badly. It should be fairly obvious that if someone runs at 2.0 GHz for 24 hours and another person runs at 4.5 GHz for 24 hours, they're both still doing 24 hours of crunching. The only REAL difference would be number of results returned overall as the person running faster would get more. Even the points might not be very different, as that is also based on time spent computing. It might be slightly different, but not by much at all. It reminds me of a grade school word problem for math. If David spent 24 hours walking at 2 miles per hour, and John spent 24 hours walking at 4 miles per hour, which one spent more time walking? i5-12600K (3.7GHz), 32GB DDR5, Win11 64bit Home |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Well to me it does. If there are no reliable hosts, all goes quorum 2, either immediate or in the second instance. At present we have a mix of work in the cycle. But then 1. Work assigned prior to the upgrade, which was circulated using the reliability rating. Returned after the upgrade requires a 2nd wingman. Not aware that at present time tasks are allowed to go it alone. a) I would rather expect to see those WUs sitting in PV jail, not as "inconclusive" b) at the time I mentioned this, only three out of 30 hosts showed up on the inconclusive list, I would have expected that many more hosts would be in that list, beside the occasional ones. And it seemed to be only CEP2 and C4CW WUs only, none of the SN2S ones... 2. Loss of some post service resume *early* returned results due some bad rule in the transitioner that had not shown it's face during testing. Not seen any go AWOL since the batch I had last night... that's now 18 hours ago. Since all return fine, with a little delay on validations at times. After the system came back up last evening, PV jail filled up with about 12 pages (=180) of WUs until I went to bed. Would have expected going by "normal production" that there would have been roughly 240-250WUs in PV jail in the morning unless the validator got turned back on (which wasn't the case until I went to bed).In the morning then, there where 8 pages of PV (=120 WUs), with roughly 50-60 validated over night and about 30 showing inconclusive. That means that somehow 40-50 WUs went MIA... Ralf -For inconclusive look at the distribution timestamps of your return and time send of your wingman. -Inconclusive generally applies to the single distribution sciences [CEP2/C4CW/FAAH/HFCC where *afterwards* a second copy is send out to check. -Inconclusive occasionally occurs when 2 wingman dont agree, then a 3rd copy acts as tiebreaker. For details of statuses and distribution see the Start Here FAQ's. Happened to some that as said connected early in the service resume. The validator was switch on at about 11:00 AM my time https://secure.worldcommunitygrid.org/forums/wcg/printpost_post,367570 which is about 4 Central time in the USA. Edit: And with assistance of an insider view, the mystery is resolved, for me. Look in the message log where the Resend due lost is logged, 143 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_1076_1 144 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_0022_1 145 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_0174_0 146 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0725_1 147 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0770_0 148 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0767_1 149 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0084_0 150 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0123_1 151 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0601_1 152 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0664_0 153 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000018_0068_1 154 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000026_0360_0 Then look in the Tasks view. They're back, for a second time crunch... they're presently recomputed on my computer. --//-- [Edit 2 times, last edit by Former Member at Mar 8, 2012 7:18:44 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks Nanoprobe. Knew I saw it somewhere but couldn't find it again about the points. With the increase in points, badges will come quicker. Hope WCG considers more badges 5, 10, etc years. How would an increase in points give faster badges? Badges are based on time spent crunching, not points earned for that time, right? Or am I missing something? No you're not missing something. I had to comment on another poster who was overclocking to get badges faster. It really makes me wonder how people can misunderstand things so badly. It should be fairly obvious that if someone runs at 2.0 GHz for 24 hours and another person runs at 4.5 GHz for 24 hours, they're both still doing 24 hours of crunching. The only REAL difference would be number of results returned overall as the person running faster would get more. Even the points might not be very different, as that is also based on time spent computing. It might be slightly different, but not by much at all. It reminds me of a grade school word problem for math. If David spent 24 hours walking at 2 miles per hour, and John spent 24 hours walking at 4 miles per hour, which one spent more time walking? Hyperthreading is already doubling recorded computing time, so really one could argue that 4GHZ was recorded as 2x2GHZ ;P (I'll hide for a while) --//-- |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Strange things afoot with tasks being sent out or not it seems.
Minimum Quorum: 2 Replication: 2 Result Name App Version Number Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 2-- - In Progress 3/7/12 05:11:37 3/19/12 05:11:37 0.00 0.0 / 0.0 E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 1-- 640 Pending Validation 3/7/12 04:24:45 3/8/12 05:21:10 7.65 134.2 / 0.0 E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 0-- 640 Pending Validation 2/26/12 04:14:55 3/7/12 05:48:06 10.06 504.4 / 0.0 Two already pending another sent out why, and i have another new CEP task that mine is the only one sent so will the other one be sent out after i return my result.  I thought that both copies would be sent at the same time , yes.  |
||
|
|
TPCBF
Master Cruncher USA Joined: Jan 2, 2011 Post Count: 1951 Status: Recently Active Project Badges: 
|
Happened to some that as said connected early in the service resume. The validator was switch on at about 11:00 AM my time https://secure.worldcommunitygrid.org/forums/wcg/printpost_post,367570 which is about 4 Central time in the USA. Not sure if that refers to my suspicion of WUs are MIA. I just checked only cursory and found one of the CEP2 WUs (as they are hard to come by anyway) and the task E206530_209_C.25.C21H11NS3.01671669.0.set1d06 still shows as "in progress (since 3/5/12 15:15:54) though it was already uploaded at 3/6/12 3:03 Edit: And with assistance of an insider view, the mystery is resolved, for me. Look in the message log where the Resend due lost is logged, 143 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_1076_1 144 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_0022_1 145 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000011_0174_0 146 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0725_1 147 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0770_0 148 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0767_1 149 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0084_0 150 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0123_1 151 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0601_1 152 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000015_0664_0 153 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000018_0068_1 154 World Community Grid 7-3-2012 0:16:13 Resent lost task SN2S_AAB88508_0000026_0360_0 Then look in the Tasks view. They're back, for a second time crunch... they're presently recomputed on my computer. --//-- 3/6/2012 3:03:42 AM World Community Grid Computation for task E206530_209_C.25.C21H11NS3.01671669.0.set1d06_0 finishedI am pretty sure that I saw that WU in the PV list (so it had already be reported) before the validator had been turned on), but it still shows in the "in progress list" There is not a single "resent" message in the message log of that host... Ralf  |
||
|
|
KWSN - A Shrubbery
Master Cruncher Joined: Jan 8, 2006 Post Count: 1585 Status: Offline |
Strange things afoot with tasks being sent out or not it seems. Minimum Quorum: 2 Replication: 2 Result Name App Version Number Status Sent Time Time Due / Return Time CPU Time (hours) Claimed/ Granted BOINC Credit E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 2-- - In Progress 3/7/12 05:11:37 3/19/12 05:11:37 0.00 0.0 / 0.0 E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 1-- 640 Pending Validation 3/7/12 04:24:45 3/8/12 05:21:10 7.65 134.2 / 0.0 E206313_ 386_ C.24.C18H12N2OSSi2.00822641.3.set1d06_ 0-- 640 Pending Validation 2/26/12 04:14:55 3/7/12 05:48:06 10.06 504.4 / 0.0 Two already pending another sent out why, and i have another new CEP task that mine is the only one sent so will the other one be sent out after i return my result. I thought that both copies would be sent at the same time , yes. I bolded the appropriate sections to unravel this mystery. The bottom result went past the deadline so a third result was sent. After that, the older result was returned. Now, if the first result gets server aborted or not will be something to follow. Distributed computing volunteer since September 27, 2000 |
||
|
|
|