Welcome to MilkyWay@home

N-body updated to 0.40

Message boards : News : N-body updated to 0.40
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · Next

AuthorMessage
Gumpokc

Send message
Joined: 7 Sep 09
Posts: 3
Credit: 147,071
RAC: 0
Message 46912 - Posted: 2 Apr 2011, 3:16:16 UTC - in response to Message 46863.  

I am having the exact same problem, for roughly the last 3-5 days. EVERY nbody_orphan test unit fails within 3-6 seconds.

running windows vista home premium, intel quad core 2.66

Everytime the workunits crash it leaves 32 closed error windows.
Tonight i checked here and found this thread.
First i reset the project...no change.
Then I manually downloaded the nbody, dll's, and pthreads.....no change
I do not have a app_info.xml file in any boinc directory, not even in hidden directories, and I am not sure on what all needs to be in it, so I did not try that as a fix.

I am considering shutting down M@H for a week or two to see if the workunit bugs can be worked out.
ID: 46912 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Justin
Avatar

Send message
Joined: 8 Jul 10
Posts: 16
Credit: 11,058,624
RAC: 0
Message 46913 - Posted: 2 Apr 2011, 3:44:36 UTC

Here's my problem:

I've ran into situation's where the multi-threaded N-body will take 7.80 of my CPU. This will place 7 other active CPU tasks (sse3) on hold, while it finishes the Nbody; and after the Nbody finishes... it grabs another Nbody... and it can cause a hit to my RAC as no GPU tasks get queued (so my GPU idles) and the remaining 7 sse3 tasks don't have enough cores accessible to finish or resume crunching.

help.

ID: 46913 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Toppie*

Send message
Joined: 28 Mar 09
Posts: 68
Credit: 1,003,982,681
RAC: 0
Message 46923 - Posted: 2 Apr 2011, 8:29:01 UTC - in response to Message 46913.  

Here's my problem:

I've ran into situation's where the multi-threaded N-body will take 7.80 of my CPU. This will place 7 other active CPU tasks (sse3) on hold, while it finishes the Nbody; and after the Nbody finishes... it grabs another Nbody... and it can cause a hit to my RAC as no GPU tasks get queued (so my GPU idles) and the remaining 7 sse3 tasks don't have enough cores accessible to finish or resume crunching.

help.


On my win7 64 with one gpu it works 100%
On my vista64 with two gpu's they ALL error out. I tried resetting, etc. but
no luck.
ID: 46923 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Len LE/GE

Send message
Joined: 8 Feb 08
Posts: 261
Credit: 104,050,322
RAC: 0
Message 46927 - Posted: 2 Apr 2011, 10:31:11 UTC - in response to Message 46905.  

after i make a app_info file and then copy this

<app><!-- std app for N-Body 0.40 mt 32bit -->
...
...
...
</app_version>

into it i keep getting a "parse error in app_info.xml; check XML synatx" in red in my messages ?


I only posted the part needed for the new nbody app.
You made a fresh app_info.xml and need the wrapping app_info tab.

Insert first line:
<app_info>
Insert last line:
</app_info>

Those tags are only needed once in this file (start and end of the file), and all settings for even different apps are done in between that wrapping tags.

So it should look like

<app_info>

<app><!-- std app for N-Body 0.40 mt 32bit -->
...
...
...
</app_version>

</app_info>


The version you used is for win32 (4 of your systems are win32); for your win64 system you need to change the exe and dll names like described in my original post.

If it's still not working, there must be something else wrong and we need to take another look.
ID: 46927 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ahj
Avatar

Send message
Joined: 11 Aug 10
Posts: 8
Credit: 20,861,275
RAC: 0
Message 46928 - Posted: 2 Apr 2011, 10:31:58 UTC

Every single one of my nbody units seem to be erroring out instantly. Can't tell if its a problem on my end or MW's :(

Sat 02 Apr 2011 09:15:34 PM EST		Starting BOINC client version 6.10.58 for x86_64-pc-linux-gnu
Sat 02 Apr 2011 09:15:34 PM EST		Config: GUI RPC allowed from:
Sat 02 Apr 2011 09:15:34 PM EST		log flags: file_xfer, sched_ops, task
Sat 02 Apr 2011 09:15:34 PM EST		Libraries: libcurl/7.21.0 OpenSSL/0.9.8o zlib/1.2.3.4 libidn/1.18
Sat 02 Apr 2011 09:15:34 PM EST		Data directory: /var/lib/boinc-client
Sat 02 Apr 2011 09:15:34 PM EST		Processor: 4 GenuineIntel Intel(R) Atom(TM) CPU D510   @ 1.66GHz [Family 6 Model 28 Stepping 10]
Sat 02 Apr 2011 09:15:34 PM EST		Processor: 512.00 KB cache
Sat 02 Apr 2011 09:15:34 PM EST		Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc arch_perfmon pebs bts rep_good aperfmperf pni dtes64 monitor ds_cpl tm2 ssse3 cx16 xtpr 
Sat 02 Apr 2011 09:15:34 PM EST		OS: Linux: 2.6.35-25-generic
Sat 02 Apr 2011 09:15:34 PM EST		Memory: 3.86 GB physical, 6.09 GB virtual
Sat 02 Apr 2011 09:15:34 PM EST		Disk: 140.72 GB total, 129.54 GB free
Sat 02 Apr 2011 09:15:34 PM EST		Local time is UTC +11 hours
Sat 02 Apr 2011 09:15:34 PM EST		No usable GPUs found
Sat 02 Apr 2011 09:15:34 PM EST		No general preferences found - using BOINC defaults
Sat 02 Apr 2011 09:15:34 PM EST		Reading preferences override file
Sat 02 Apr 2011 09:15:34 PM EST		Preferences:
Sat 02 Apr 2011 09:15:34 PM EST		   max memory usage when active: 1976.44MB
Sat 02 Apr 2011 09:15:34 PM EST		   max memory usage when idle: 3557.59MB
Sat 02 Apr 2011 09:15:34 PM EST		   max disk usage: 10.00GB
Sat 02 Apr 2011 09:15:34 PM EST		   don't use GPU while active
Sat 02 Apr 2011 09:15:34 PM EST		   suspend work if non-BOINC CPU load exceeds 25 %
Sat 02 Apr 2011 09:15:34 PM EST		   (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
Sat 02 Apr 2011 09:15:34 PM EST		Not using a proxy
Sat 02 Apr 2011 09:15:34 PM EST		This computer is not attached to any projects
Sat 02 Apr 2011 09:15:34 PM EST		Visit http://boinc.berkeley.edu for instructions
Sat 02 Apr 2011 09:16:23 PM EST		Fetching configuration file from http://milkyway.cs.rpi.edu/milkyway/get_project_config.php
Sat 02 Apr 2011 09:16:48 PM EST		Running CPU benchmarks
Sat 02 Apr 2011 09:16:48 PM EST		Suspending computation - running CPU benchmarks
Sat 02 Apr 2011 09:16:58 PM EST	Milkyway@home	work fetch suspended by user
Sat 02 Apr 2011 09:17:19 PM EST		Benchmark results:
Sat 02 Apr 2011 09:17:19 PM EST		   Number of CPUs: 4
Sat 02 Apr 2011 09:17:19 PM EST		   502 floating point MIPS (Whetstone) per CPU
Sat 02 Apr 2011 09:17:19 PM EST		   2372 integer MIPS (Dhrystone) per CPU
Sat 02 Apr 2011 09:17:20 PM EST		Resuming computation
Sat 02 Apr 2011 09:17:26 PM EST	Milkyway@home	Master file download succeeded
Sat 02 Apr 2011 09:17:31 PM EST	Milkyway@home	Sending scheduler request: Project initialization.
Sat 02 Apr 2011 09:17:31 PM EST	Milkyway@home	Requesting new tasks
Sat 02 Apr 2011 09:17:37 PM EST	Milkyway@home	Scheduler request completed: got 1 new tasks
Sat 02 Apr 2011 09:17:37 PM EST	Milkyway@home	General prefs: from Milkyway@home (last modified 29-Mar-2011 19:45:42)
Sat 02 Apr 2011 09:17:37 PM EST	Milkyway@home	Host location: none
Sat 02 Apr 2011 09:17:37 PM EST	Milkyway@home	General prefs: using your defaults
Sat 02 Apr 2011 09:17:37 PM EST		Reading preferences override file
Sat 02 Apr 2011 09:17:37 PM EST		Preferences:
Sat 02 Apr 2011 09:17:37 PM EST		   max memory usage when active: 1976.44MB
Sat 02 Apr 2011 09:17:37 PM EST		   max memory usage when idle: 3557.59MB
Sat 02 Apr 2011 09:17:37 PM EST		   max disk usage: 70.36GB
Sat 02 Apr 2011 09:17:37 PM EST		   suspend work if non-BOINC CPU load exceeds 25 %
Sat 02 Apr 2011 09:17:37 PM EST		   (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
Sat 02 Apr 2011 09:17:39 PM EST	Milkyway@home	Started download of milkyway_0.50_x86_64-pc-linux-gnu__sse2
Sat 02 Apr 2011 09:17:39 PM EST	Milkyway@home	Started download of parameters-13.txt
Sat 02 Apr 2011 09:17:41 PM EST	Milkyway@home	Finished download of parameters-13.txt
Sat 02 Apr 2011 09:17:41 PM EST	Milkyway@home	Started download of stars-13.txt
Sat 02 Apr 2011 09:20:31 PM EST	Milkyway@home	Finished download of milkyway_0.50_x86_64-pc-linux-gnu__sse2
Sat 02 Apr 2011 09:21:10 PM EST	Milkyway@home	Finished download of stars-13.txt
Sat 02 Apr 2011 09:25:49 PM EST		Not using a proxy
Sat 02 Apr 2011 09:26:24 PM EST	Milkyway@home	update requested by user
Sat 02 Apr 2011 09:26:25 PM EST	Milkyway@home	work fetch resumed by user
Sat 02 Apr 2011 09:26:26 PM EST	Milkyway@home	Sending scheduler request: Requested by user.
Sat 02 Apr 2011 09:26:26 PM EST	Milkyway@home	Reporting 1 completed tasks, requesting new tasks
Sat 02 Apr 2011 09:26:36 PM EST	Milkyway@home	Scheduler request completed: got 1 new tasks
Sat 02 Apr 2011 09:26:38 PM EST	Milkyway@home	File milkyway_nbody_0.40_x86_64-pc-linux-gnu__mt exists already, skipping download
Sat 02 Apr 2011 09:26:38 PM EST	Milkyway@home	Started download of orphan_test_2model.lua
Sat 02 Apr 2011 09:26:38 PM EST	Milkyway@home	Started download of histogram
Sat 02 Apr 2011 09:26:45 PM EST	Milkyway@home	Finished download of orphan_test_2model.lua
Sat 02 Apr 2011 09:26:45 PM EST	Milkyway@home	Finished download of histogram
Sat 02 Apr 2011 09:26:45 PM EST	Milkyway@home	Starting de_nbody_orphan_test_2model_4_9757_1301739233_0
Sat 02 Apr 2011 09:26:45 PM EST	Milkyway@home	Starting task de_nbody_orphan_test_2model_4_9757_1301739233_0 using milkyway_nbody version 40
Sat 02 Apr 2011 09:26:46 PM EST	Milkyway@home	Computation for task de_nbody_orphan_test_2model_4_9757_1301739233_0 finished
Sat 02 Apr 2011 09:27:46 PM EST	Milkyway@home	Sending scheduler request: To fetch work.
Sat 02 Apr 2011 09:27:46 PM EST	Milkyway@home	Reporting 1 completed tasks, requesting new tasks
Sat 02 Apr 2011 09:27:57 PM EST	Milkyway@home	Scheduler request completed: got 2 new tasks
Sat 02 Apr 2011 09:28:00 PM EST	Milkyway@home	Starting de_nbody_orphan_test_2model_4_9833_1301739233_0
Sat 02 Apr 2011 09:28:00 PM EST	Milkyway@home	Starting task de_nbody_orphan_test_2model_4_9833_1301739233_0 using milkyway_nbody version 40
Sat 02 Apr 2011 09:28:01 PM EST	Milkyway@home	Computation for task de_nbody_orphan_test_2model_4_9833_1301739233_0 finished
Sat 02 Apr 2011 09:28:01 PM EST	Milkyway@home	Starting de_nbody_orphan_test_2model_4_9815_1301739233_0
Sat 02 Apr 2011 09:28:01 PM EST	Milkyway@home	Starting task de_nbody_orphan_test_2model_4_9815_1301739233_0 using milkyway_nbody version 40
Sat 02 Apr 2011 09:28:02 PM EST	Milkyway@home	Computation for task de_nbody_orphan_test_2model_4_9815_1301739233_0 finished
Sat 02 Apr 2011 09:28:58 PM EST	Milkyway@home	work fetch suspended by user

Goal: 1 BILLION cobblestones

Progress: 2%
ID: 46928 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
DJStarfox

Send message
Joined: 29 Sep 10
Posts: 54
Credit: 1,342,886
RAC: 0
Message 46930 - Posted: 2 Apr 2011, 13:41:19 UTC - in response to Message 46928.  
Last modified: 2 Apr 2011, 13:41:34 UTC

Every single one of my nbody units seem to be erroring out instantly. Can't tell if its a problem on my end or MW's :(


This is the error you're getting:
<core_client_version>6.10.58</core_client_version>
<![CDATA[
<message>
process exited with code 22 (0x16, -234)
</message>
<stderr_txt>
execv: No such file or directory

</stderr_txt>
]]>


I would check/reset file system permissions!
ID: 46930 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
JoeyJoJo

Send message
Joined: 19 Jan 11
Posts: 2
Credit: 10,226
RAC: 0
Message 46940 - Posted: 2 Apr 2011, 23:00:26 UTC

Hi there,

I've had a fair few of these "stop working" (vista x64).

Is it necessary to take action? Or will it work itself out?

Cheers
ID: 46940 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Toppie*

Send message
Joined: 28 Mar 09
Posts: 68
Credit: 1,003,982,681
RAC: 0
Message 46947 - Posted: 3 Apr 2011, 9:01:13 UTC - in response to Message 46940.  

Hi there,

I've had a fair few of these "stop working" (vista x64).

Is it necessary to take action? Or will it work itself out?

Cheers


Same here. I think it is a Vista 64 problem. In the Windows
error reporting it says "Kernel32.dll!Key,etc" On my
Win7 64 machine it runs flawlessly. My problem is that while
Win waits for me to close the program it stays in the 7.80
cpu usage and prevents further cpu crunching.
ID: 46947 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Gumpokc

Send message
Joined: 7 Sep 09
Posts: 3
Credit: 147,071
RAC: 0
Message 46971 - Posted: 5 Apr 2011, 5:20:08 UTC - in response to Message 46947.  

That may be, but if so it's a really new problem.

I've woken up to check things several times, and gotten home from work to find error windows spammed all over the place. I'm just going to shut MW@H down for abit. If the problem gets corrected, then I'll pick it backup, if not, I'll drop MW@H, I don't see any other choices.
ID: 46971 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
JoeyJoJo

Send message
Joined: 19 Jan 11
Posts: 2
Credit: 10,226
RAC: 0
Message 47016 - Posted: 6 Apr 2011, 10:39:57 UTC - in response to Message 46971.  

That may be, but if so it's a really new problem.

I've woken up to check things several times, and gotten home from work to find error windows spammed all over the place. I'm just going to shut MW@H down for abit. If the problem gets corrected, then I'll pick it backup, if not, I'll drop MW@H, I don't see any other choices.


FWIW my machine hasn't complained in quite a few days. Perhaps it's fixed.
ID: 47016 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
pstehno
Avatar

Send message
Joined: 16 Jun 10
Posts: 6
Credit: 7,402,186
RAC: 0
Message 47037 - Posted: 6 Apr 2011, 15:15:17 UTC

Finally saw it run without erroring out! It took about 18 minutes to run while hogging all 4 available CPUs.
ID: 47037 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile [AF>EDLS] Polynesia
Avatar

Send message
Joined: 5 Apr 09
Posts: 71
Credit: 6,120,786
RAC: 0
Message 47121 - Posted: 7 Apr 2011, 6:36:45 UTC

this does not work with me .....

<app_info>
<app>
<name>milkyway_nbody</name>
<user_friendly_name>MilkyWay@Home nbody Simulation</user_friendly_name>
</app>
<file_info>
<name>milkyway_nbody_0.40_windows_x86_64__mt.exe</name>
<executable/>
</file_info>
<file_info>
<name>libgomp_64-1.dll</name>
<executable/>
</file_info>
<file_info>
<name>pthreadGC2_64.dll</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway_nbody</app_name>
<version_num>40</version_num>
<plan_class>mt</plan_class>
<avg_ncpus>4</avg_ncpus>
<max_ncpus>4</max_ncpus>
<cmdline>--nthreads=4</cmdline>
<file_ref>
<file_name>milkyway_nbody_0.40_windows_x86_64__mt.exe</file_name>
<main_program/>
</file_ref>
<file_info>
<name>libgomp_64-1.dll</name>
<executable/>
</file_info>
<file_info>
<name>pthreadGC2_64.dll</name>
<executable/>
</file_info>
</app_version>
<app>
<name>milkyway</name>
<user_friendly_name>Milkyway@home Separation</user_friendly_name>
</app>
<file_info>
<name>milkyway_0.52_windows_intelx86__cuda_opencl.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>52</version_num>
<plan_class>cuda_opencl</plan_class>
<avg_ncpus>0.15</avg_ncpus>
<max_ncpus>0.30</max_ncpus>
<flops>1.0e11</flops>
<coproc>
<type>CUDA</type>
<count>1</count>
</coproc>
<file_ref>
<file_name>milkyway_0.52_windows_intelx86__cuda_opencl.exe</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>
Team Alliance francophone, boinc: 7.0.18

GA-P55-UD5, i7 860, Win 7 64 bits, 8g DDR3, GTX 470
ID: 47121 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Len LE/GE

Send message
Joined: 8 Feb 08
Posts: 261
Credit: 104,050,322
RAC: 0
Message 47127 - Posted: 7 Apr 2011, 10:52:28 UTC

ID: 47127 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile [AF>EDLS] Polynesia
Avatar

Send message
Joined: 5 Apr 09
Posts: 71
Credit: 6,120,786
RAC: 0
Message 47151 - Posted: 7 Apr 2011, 19:02:43 UTC

ok thanks
Team Alliance francophone, boinc: 7.0.18

GA-P55-UD5, i7 860, Win 7 64 bits, 8g DDR3, GTX 470
ID: 47151 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile [AF>EDLS] Polynesia
Avatar

Send message
Joined: 5 Apr 09
Posts: 71
Credit: 6,120,786
RAC: 0
Message 47938 - Posted: 16 Apr 2011, 22:13:16 UTC

hello,

when will work for this application? thank you
Team Alliance francophone, boinc: 7.0.18

GA-P55-UD5, i7 860, Win 7 64 bits, 8g DDR3, GTX 470
ID: 47938 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Matt Arsenault
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 8 May 10
Posts: 576
Credit: 15,979,383
RAC: 0
Message 47941 - Posted: 16 Apr 2011, 22:23:51 UTC - in response to Message 47938.  

hello,

when will work for this application? thank you
Travis is working on fixing the nbody assimilator to use a new credit policy or something; whenever that's done.
ID: 47941 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
techbird

Send message
Joined: 15 Jan 11
Posts: 16
Credit: 40,731,537
RAC: 0
Message 48403 - Posted: 1 May 2011, 16:25:34 UTC

All was well with n_body WUs last night.

Now they all error out.
ID: 48403 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
aad

Send message
Joined: 30 Mar 09
Posts: 63
Credit: 621,582,726
RAC: 4
Message 48410 - Posted: 1 May 2011, 17:00:22 UTC - in response to Message 48403.  

All was well with n_body WUs last night.

Now they all error out.


Looks like you (and your wingmen)had the -177 problem.
You can find that in the Stderr output;
<message>
Maximum elapsed time exceeded
</message>
ID: 48410 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
techbird

Send message
Joined: 15 Jan 11
Posts: 16
Credit: 40,731,537
RAC: 0
Message 48416 - Posted: 1 May 2011, 18:22:07 UTC - in response to Message 48410.  

All was well with n_body WUs last night.

Now they all error out.


Looks like you (and your wingmen)had the -177 problem.
You can find that in the Stderr output;
<message>
Maximum elapsed time exceeded
</message>


Max elapsed time exceeded after

Run time 1.14 sec.
CPU time 2.56 sec.

Run times were around 25min each and completing successfully now I'm excedding max time I don't really understand how this can be.

as for the -177 problem what is it and how can I fix it? A reboot didn't help. changing the app_info file --nthread=2 from 1 thru 4 has had no effect. I'm about to try to detach and reattch and see if that will help, as soon as my gpu finishes its tasks.

ID: 48416 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Matt Arsenault
Volunteer moderator
Project developer
Project tester
Project scientist

Send message
Joined: 8 May 10
Posts: 576
Credit: 15,979,383
RAC: 0
Message 48417 - Posted: 1 May 2011, 18:35:43 UTC - in response to Message 48416.  

All was well with n_body WUs last night.

Now they all error out.


Looks like you (and your wingmen)had the -177 problem.
You can find that in the Stderr output;

Maximum elapsed time exceeded



Max elapsed time exceeded after

Run time 1.14 sec.
CPU time 2.56 sec.

Run times were around 25min each and completing successfully now I'm excedding max time I don't really understand how this can be.

as for the -177 problem what is it and how can I fix it? A reboot didn't help. changing the app_info file --nthread=2 from 1 thru 4 has had no effect. I'm about to try to detach and reattch and see if that will help, as soon as my gpu finishes its tasks.

It's a workunit problem. The flops estimates in the workunits I'm getting are 0, so the limits are immediately exceeded.
ID: 48417 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · Next

Message boards : News : N-body updated to 0.40

©2024 Astroinformatics Group