| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: New Beta Test - Nov 27, 2014 [ Issues thread ] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 253
|
|
| Author |
|
|
yoro42
Ace Cruncher United States Joined: Feb 19, 2011 Post Count: 8979 Status: Offline Project Badges: 
|
Four WU In Progress (Completion % shown): 22.9% - BETA_ OET1_ 0000065_ xSDGP_ 0094_ 0-- android_232dc43a In Progress 12/4/14 21:56:55 12/8/14 21:56:55 0.00 / 0.00 0.0 / 0.0 77.3% - BETA_ OET1_ 0000065_ xSDGP_ 0239_ 0-- android_232dc43a In Progress 12/4/14 21:51:17 12/8/14 21:51:17 0.00 / 0.00 0.0 / 0.0 FINI (see below) 87.8% - BETA_ OET1_ 0000065_ xSDGP_ 0145_ 0-- android_232dc43a In Progress 12/4/14 19:47:39 12/8/14 19:47:39 0.00 / 0.00 0.0 / 0.0 68.5% - BETA_ OET1_ 0000065_ xSDGP_ 0130_ 0-- android_232dc43a In Progress 12/4/14 19:47:38 12/8/14 19:47:38 0.00 / 0.00 0.0 / 0.0 BETA_ OET1_ 0000065_ xSDGP_ 0145_ 0-- android_232dc43a Pending Validation 12/4/14 19:47:39 12/5/14 19:06:35 20.10 / 20.38 86.7 / 0.0  |
||
|
|
Yarensc
Advanced Cruncher USA Joined: Sep 24, 2011 Post Count: 136 Status: Offline Project Badges: 
|
I have one of the android units from batch 065 that errored out at around 70% (13 hours) on my HTC One running KitKat. It was checkpointing fine (between 5-30 minutes each) and everything was progressing smoothly (unlike the Windows betas) until the computational error hit.
I'm not positive, but it might have happened when it was force paused due to the battery level dropping too low (it was set to run on battery). This task, along with two others that only ran for an hour, was listed as a 'segmentation violation' in the log, however this one added these lines the two short ones didn't have #00 pc 0001f0bc <unknown> (???) [0xb43878] #01 pc 00102688 /data/data/edu.berkeley.boinc/client/projects/www.worldcommunitygrid.org/wcgrid_beta20_v (_ZL21worker_signal_handleri+0x14) [0xebfbff91] |
||
|
|
KWSN-A Shrubbery
Senior Cruncher Joined: Jan 8, 2006 Post Count: 476 Status: Offline Project Badges: 
|
All my Android beta downloads get checksum or signature verification error. All the wingmen have similar errors. KWSN, are you still seeing this with version 7.09 which ran against batch 65? Batch 69 had this issue which was running 7.08. Thanks, -Uplinger Ok, so it turns out that this phone is still downloading the same errored tasks over and over and giving the same error. It's preventing me from downloading any tasks at all. I reset the project. Guess we'll see if it makes a difference. Edit: OK, able to download FAAH work so the reset appears to have worked.  [Edit 1 times, last edit by KWSN-A Shrubbery at Dec 6, 2014 2:10:50 AM] |
||
|
|
Rob_B
Cruncher The Netherlands Joined: Jun 2, 2007 Post Count: 22 Status: Offline Project Badges: 
|
I got one beta wu on one of my linux crunchers today and suspended it at around 45%. Now about an half hour after resuming the wu it is at 10% and it seems like it is stuck there.
----------------------------------------I will let it run for a while and see what happens. Computer: i7-3770-linux  [Edit 1 times, last edit by Rob_B at Dec 6, 2014 9:16:11 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I got one beta wu on one of my linux crunchers today and suspended it at around 45%. Now about an half hour after resuming the wu it is at 10% and it seems like it is stuck there. I will let it run for a while and see what happens. Computer: i7-3770-linux I have the same behavior on my Mac. |
||
|
|
Rob_B
Cruncher The Netherlands Joined: Jun 2, 2007 Post Count: 22 Status: Offline Project Badges:
|
I see it isn't stuck anymore. But it makes 10% steps. And each step takes more than one hour.
----------------------------------------This wu is a resend already. The wu status shows one wingman as "no reply" and the other took over 15 hours to complete. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
These lengthy Beta v7.07 units should be worth attempting if you have a fast CPU, so that they have a better chance of finishing before a deadline is hit. Keith Uplinger has mentioned the value of doing this to help provide the techs with data about them.
|
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1403 Status: Offline Project Badges: 
|
These lengthy Beta v7.07 units should be worth attempting if you have a fast CPU, so that they have a better chance of finishing before a deadline is hit. Keith Uplinger has mentioned the value of doing this to help provide the techs with data about them. These tasks are almost worthless for the researchers. The clients are not able to upload the results, because BOINC aborts the task. It's not the deadline causing troubles, but the allowed maximum runtime. Fast machines get less runtime for these tasks than slow machines, so fast machines also suffer from the "Maximum elapsed time exceeded". That maximum time is calculated in seconds from the division of 294697.70Gfpops/Gfpops_of_your_machine. That 294697.70Gfpops came from the WCG-staff and was way to low for these heavy tasks. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The clients are not able to upload the results, because BOINC aborts the task. My point is that some (faster) clients do finish and upload results for some of these lengthy Beta v7.07 units, so it is worth attempting them on a faster machine. Yes, it's inefficient, but not worthless. Here's one recent example (mine was _5):BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 5-- 707 Valid 05/12/14 17:43:39 06/12/14 11:47:37 17.89 17.1 / 17.1 BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 4-- - No Reply 04/12/14 17:35:01 06/12/14 03:11:00 0.00 0.0 / 0.0 BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 3-- 707 Valid 04/12/14 08:07:37 05/12/14 19:27:25 23.59 17.1 / 17.1 BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 2-- 707 Error 02/12/14 17:34:26 04/12/14 16:20:36 32.69 17.1 / 0.0 BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 1-- 707 User Aborted 02/12/14 07:37:52 02/12/14 17:34:22 0.09 3.1 / 0.0 BETA_ OET1_ 0000307_ xEBGP-OM_ rig_ 0323_ 0-- 707 Error 02/12/14 07:37:25 04/12/14 08:05:09 22.60 17.1 / 0.0 |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
These tasks are almost worthless for the researchers. The clients are not able to upload the results, because BOINC aborts the task. It's not the deadline causing troubles, but the allowed maximum runtime. Fast machines get less runtime for these tasks than slow machines, so fast machines also suffer from the "Maximum elapsed time exceeded". That maximum time is calculated in seconds from the division of 294697.70Gfpops/Gfpops_of_your_machine. That 294697.70Gfpops came from the WCG-staff and was way to low for these heavy tasks. Agree. I have fast machines but have been aborting any tasks that have been sent out more than 3 times after checking my results and finding none that validate. How many times do the techs need to see the same error? Complete waste of computing time IMHO.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
|