| 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 - July 21, 2017 [ Issues Thread ] |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 171
|
|
| Author |
|
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: 
|
Got 2 resends on my XP-x64 SP2 device (3770K CPU in Z68 mb) and they both errored out immediately.
----------------------------------------Wingman 0 was No Reply, and wingman 2 is In Progress in both cases. Error logs on WCG website are the same for both: .. Result Name: BETA_ beta26_ 00000041_ 1913_ 1-- .. <core_client_version>7.6.22</core_client_version> .. <![CDATA[ .. <message> .. couldn't start app: CreateProcess() failed - (unknown error) .. </message> .. ]]> The BOINC client messages log also has the same content for both: .. 22983 World Community Grid 3/08/2017 12:25:04 PM [error] Process creation failed: (unknown error) - error code 193 (0xc1) .. 22984 World Community Grid 3/08/2017 12:25:04 PM [error] Process creation failed: (unknown error) - error code 193 (0xc1) .. 22985 World Community Grid 3/08/2017 12:25:05 PM [error] Process creation failed: (unknown error) - error code 193 (0xc1) .. 22986 World Community Grid 3/08/2017 12:25:06 PM [error] Process creation failed: (unknown error) - error code 193 (0xc1) .. 22987 World Community Grid 3/08/2017 12:25:06 PM [error] Process creation failed: (unknown error) - error code 193 (0xc1) .. 22988 World Community Grid 3/08/2017 12:25:08 PM Computation for task BETA_beta26_00000041_1934_1 finished .. 22989 World Community Grid 3/08/2017 12:25:08 PM Output file BETA_beta26_00000041_1934_1_r1242449635_0 for task BETA_beta26_00000041_1934_1 absent The other WU that errored in the same way was BETA_ beta26_ 00000041_ 1913_ 1 [Update]:'Both wingmen who received #2 copies of these WUs retrned Valid results, but they were both crunching under Linux, not XP-x64. I wondered whether the process creations failed due to lack of available memory, so I checked. Running 2 x FAHV WUs, there's about 1.8GB physical, and 1.6GB extra in pagefile available. Plenty.[/Update] [OT] OK, experts on Rosetta/protein folding software: Does Rosetta perform simulations and/or output results that can't be run with existing GPU-based algorithms such as those used by (BOINC-based) GPUGRID ot (non-BOINC-based) Folding@Home ? While AFAIK those are currently implemented only in Nvidia's CUDA GPGPU language, I think it would be a waste of CPU resources and electrical energy to use Rosetta for this new project without a GPU alternative, if the latter could be used. [/OT] [Edit 1 times, last edit by Rickjb at Aug 4, 2017 5:50:29 AM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Got 1 WU on linux x64, running for 10 hours now at 76%. Uses ~420MB of virtual memory.
----------------------------------------[OT Reply] As far as I know, being a frequent Rosetta cruncher: Rosetta doesn't directly simulate protein folding; it uses prediction algorithms to try and achieve a similar result using less computational power. It is a predictive folder, not a computationally exact one. As such, Rosetta's programming is inherently serial and would not benefit from GPU acceleration. [Edit 1 times, last edit by Former Member at Aug 3, 2017 4:16:22 AM] |
||
|
|
PMH_UK
Veteran Cruncher UK Joined: Apr 26, 2007 Post Count: 761 Status: Offline Project Badges: 
|
I got the same error on my XP machines.
----------------------------------------Result Name: BETA_ beta26_ 00000006_ 1843_ 0-- <core_client_version>7.2.47</core_client_version> <![CDATA[ <message> couldn't start app: CreateProcess() failed - (unknown error) </message> ]]> Paul.
Paul.
|
||
|
|
BobCat13
Senior Cruncher Joined: Oct 29, 2005 Post Count: 295 Status: Offline Project Badges: 
|
I got the same error on my XP machines. Result Name: BETA_ beta26_ 00000006_ 1843_ 0-- <core_client_version>7.2.47</core_client_version> <![CDATA[ <message> couldn't start app: CreateProcess() failed - (unknown error) </message> ]]> Paul. Just like HSTB, when this beta executable was compiled for Windows the MajorSubsystemVersion was set to 6 which means Vista is the oldest Windows OS that will work. I changed the setting to 5 to see if WinXP would work and both tasks completed and validated, so XP will work if the executable is set to allow computing on WinXP. |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Just sending the fox into the hen house: Members expect WCG to provide absolutely safe / virus-checked, stable software and tasks. Is WCG expected to continue build application for a no longer supported, un-patched **, if not insecure operating system? May WCG expect reciprocity when it relates to the client?
** With some back door ways, there's still patches to get from MS, but how patched is that? http://home.bt.com/tech-gadgets/computing/is-...popular-os-11363958273351 / rant end |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Some beta units from "batch" 10 and 22 loaded on my machines recently. Estimated runtimes in 1.5 to 2.5 hour range.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Is WCG expected to continue build application for a no longer supported, un-patched **, if not insecure operating system? I would say that that was for WCG to decide. However, the article that you link to says: For most people, Windows XP was the PC high-point. And for many, it still is - which is why they’re still using it. In fact, Windows XP is still running on more than 7% of machines - way ahead of its successor Windows Vista on 0.7%. And this, for me, is the most important point. Does WCG really want to forgo the donated power of 7% of machines? Besides, more updates have been released by M$ since the article was written and, in my opinion, secure behaviour is (generally) more important than 'secure' machines. I should like to be able to donate the power of my XP machine to this project, whatever it is, when it goes live, especially as it seems that the fix is trivial. Please, |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
I seriously dislike select snip quoting, because the next reader is frequently not see the whole post for the context I presented the question in.
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Was just trying to make it clear what aspect of which post I was replying to, but will try to remember your (dis)likes.
|
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
That one was one of the larger input systems and time between checkpoints is longer. I looked at the resend and it ran for about 30 minutes between checkpoints and the average for that batch is even greater. The application takes a checkpoint every time a line like "Finished _0001 in 1710.61 seconds." shows up in the log. Your run never reached that point so that is why it restarted at 0%. The question is why was your run being restarted. If you get more work and see this again can you set some of the debug flags on your client to try and get more detail in the event log? If you can set cpu_sched, cpu_sched_debug, task_debug, suspend_debug. Thanks, armstrdj FYI, I think we can write this one off. I had four betas act this way. All were on machines running Ubuntu. It looks like I may have been getting some memory errors on some of my Linux boxes so, even with LAIM on, I got myself into a bit of a catch-22. I've rebooted all of my Linux machines and the betas I currently have seem to be running fine. Not sure how the science is set to handle memory errors but that's your monkey  |
||
|
|
|