| 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 |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
My 6.10.60 / 64-bit client for Win7 has downloaded 32-bit jobs for C4CW 6.40. Strange indeed, my 64bit rigs have already got 6.41_64 on them now they just got this with new tasks. Wed 07 Mar 2012 16:32:35 EST World Community Grid Started download of wcg_c4cw_lmps_6.40_i686-pc-linux-gnu |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I'm getting double points on my Linux machine as well. 11.10 boinc version 6.12.33 64-bit. ...looking at mine in PV, the claimed points are twice what was given before the change. Is this wrong, or, are we now seeing a general increase in credit given to bring us more into line with the rest of BOINC?C4CW, CEP2, and SN2S all affected. All in PV. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
....some are querying on how to set for GPU work. On 'other' BOINC projects you can set just for GPU and no CPU. You do NOT have to have to accept CPU work for the project just to get GPU. The GPUuses (on mine,) some 0.05% of a CPU for the GPU work. So on my 4 cores I get all of them running on WCG with some 0.05% used to support the GPU work.
|
||
|
|
pcwr
Ace Cruncher England Joined: Sep 17, 2005 Post Count: 10903 Status: Offline Project Badges: 
|
Work is going out. There were some surprises with transition cases that we didn't observe in our testing so we had to some quick fixes. Unfortunately, a few early results returned will have to be rerun  So I don't waste another 5+ days of urgent WUs, I am going to have to increase my cache by over a day on all my computers for the next downtime. Didn't help that my internet connection also went down 3 hours prior to WGC. Also, the validator doesn't seem to be working. Patrick  [Edit 1 times, last edit by pcwr at Mar 7, 2012 7:37:28 AM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges: 
|
I'm getting double points on my Linux machine as well. 11.10 boinc version 6.12.33 64-bit. ...looking at mine in PV, the claimed points are twice what was given before the change. Is this wrong, or, are we now seeing a general increase in credit given to bring us more into line with the rest of BOINC?C4CW, CEP2, and SN2S all affected. All in PV. We are allowing the new credit code from Berkeley determine how much credit is awarded after this update. However, we won't see the full impact for a week due to the way it works. It is interesting to note though that according to: http://boincstats.com/stats/project_cpcs.php we are awarding half of what Seti grants, so a double of credit would bring us inline (I use Seti as a reference because they are most closely connected with David and BOINC).* * Well - we mostly are doing this. The logic in the validators is the same as pre-update. We are hoping to simplify that logic once the new credit system has proved itself. [Edit 1 times, last edit by knreed at Mar 7, 2012 8:25:14 AM] |
||
|
|
knreed
Former World Community Grid Tech Joined: Nov 8, 2004 Post Count: 4504 Status: Offline Project Badges:
|
Also, the validator doesn't seem to be working. Yup - still working through things. There were some serious performance issues when we first brought the system up after the update. I've got the biggest ones fixed so now I'm working on bringing the rest of the system up. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Funny, when doubling the credit had also the Cobblestones need doubled to get a TeraFLOP of contribution/performance, at the time the WCG machine getting slashed in half in power with BOINCStats. Coincidence? (Something is hurting right between the ears)
--//-- |
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges: 
|
Also, the validator doesn't seem to be working. Yup - still working through things. There were some serious performance issues when we first brought the system up after the update. I've got the biggest ones fixed so now I'm working on bringing the rest of the system up. Kevin, I suspect that this thing I've noticed with the Try Validation link on my ready-to-validate jobs is something you've done as part of making sure the validation aspect of the upgrade is okay. That effectively "suspends" validation on those wu's until it gets switched back. The validators can be brought back up, checked out and then be given a handful of wu's to validate and, if that's okay, more until things look good enough to open them up for normal processing. BTW, I think it's reasonable to presume it's been quite a day for you and your cohorts. Might it be prudent to push back the filesystem upgrade a week until things settle down a tad? |
||
|
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges: 
|
I thought I would abort my sent before the update tasks, so I don't run into any more transition issues, and boost up my queue ready for the 8th
----------------------------------------But I have have now reached my daily quota of 1 task on my many cored machine(s) This is normally increased after a good work unit has been return and further work flows However... Not today Thought I should let you know You are doing a grand job Knreed. I know how difficult it is to keep smiling after the effort and hours put into a cutover. Take care Dave  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
And so the client logs after many a year of printing:
18-Feb-2012 12:43:29 [World Community Grid] [sched_op] Server version 601 18-Feb-2012 12:43:29 [World Community Grid] Project requested delay of 11 seconds is now printing 07-Mar-2012 10:10:56 [World Community Grid] [sched_op] Server version 700 07-Mar-2012 10:10:56 [World Community Grid] Project requested delay of 11 seconds A day for the records. --//-- |
||
|
|
|