Welcome to MilkyWay@home

All Milkyway@Home 1.02 tasks ending in computation error on HD6950.

Message boards : Number crunching : All Milkyway@Home 1.02 tasks ending in computation error on HD6950.
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · Next

AuthorMessage
swiftmallard
Avatar

Send message
Joined: 18 Jul 09
Posts: 300
Credit: 303,562,776
RAC: 0
Message 61174 - Posted: 21 Feb 2014, 21:24:12 UTC - in response to Message 61173.  

I'm also gettting Computation error for all the M@H projects. Any solutions to this?
I'm on the latest boinc version (7.2.39 x64) and boinc keeps running as a service. All the other projects on the same CPU are okay.

What should I do? Keep on sending errors until someone fix it?

I only see one error from an n-body task. It is nothing to worry about. If it is the "validation inconclusive" tasks you are concerned about, do not be. They will validate when another cruncher verifies your result.
ID: 61174 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
guizalan

Send message
Joined: 20 Feb 14
Posts: 6
Credit: 20,780,697
RAC: 0
Message 61177 - Posted: 21 Feb 2014, 22:12:44 UTC - in response to Message 61174.  
Last modified: 21 Feb 2014, 22:13:15 UTC

No, on the event log it says the "Computational task for blablabla has finished".
But on the progress bar where it should say "100%", it shows "error".

This happens for all M@H tasks. The other projects are just fine.
ID: 61177 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
swiftmallard
Avatar

Send message
Joined: 18 Jul 09
Posts: 300
Credit: 303,562,776
RAC: 0
Message 61178 - Posted: 21 Feb 2014, 22:53:14 UTC
Last modified: 21 Feb 2014, 22:59:16 UTC

Can you post the lines from the Event Log?
ID: 61178 · 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 61179 - Posted: 22 Feb 2014, 0:59:03 UTC - in response to Message 61173.  

Your actual result list shows no errors for modified fit 1.28, no errors for n-body 1.40 and no errors for mw 1.02.

... boinc keeps running as a service.


Do not run boinc as a service. This might be the cause that you are seeing a confused local client showing errors while the results are valid.
ID: 61179 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
swiftmallard
Avatar

Send message
Joined: 18 Jul 09
Posts: 300
Credit: 303,562,776
RAC: 0
Message 61180 - Posted: 22 Feb 2014, 1:14:07 UTC - in response to Message 61179.  

Do not run boinc as a service. This might be the cause that you are seeing a confused local client showing errors while the results are valid.

I was wondering about that but I've never run Boinc as a service so I didn't want to comment about it.
ID: 61180 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
guizalan

Send message
Joined: 20 Feb 14
Posts: 6
Credit: 20,780,697
RAC: 0
Message 61181 - Posted: 22 Feb 2014, 2:48:34 UTC

The event logs doesn't says much...

21/02/2014 17:24:57 | Milkyway@Home | Starting task ps_separation_10_2s_sSgrFreeInertia_4_1392810302_699245_0
21/02/2014 17:26:05 | Milkyway@Home | Computation for task ps_separation_10_2s_sSgrFreeInertia_4_1392810302_699245_0 finished
21/02/2014 17:26:05 | Milkyway@Home | Starting task de_nbody_02_12_sim_orphan_narrow_3_1392144420_116204_3
21/02/2014 17:27:19 | Milkyway@Home | Computation for task de_nbody_02_12_sim_orphan_narrow_3_1392144420_116204_3 finished
21/02/2014 17:31:01 | Milkyway@Home | Starting task ps_separation_10_2s_sSgrFreeInertia_4_1392810302_699243_0
21/02/2014 17:32:09 | Milkyway@Home | Computation for task ps_separation_10_2s_sSgrFreeInertia_4_1392810302_699243_0 finished

The status bar is not '100%' (is 'ERROR'), and the tasks are uploaded.

I don't think the problem is because of "running as service"; look my computers below.



Comp. quarto is not running as service, and it works!
Comp. PCCASA is running as service, and it works!
Comp. sala is the computer in question, running as service, the tasks are finishing with ERROR, and I don't get points for them!
ID: 61181 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
swiftmallard
Avatar

Send message
Joined: 18 Jul 09
Posts: 300
Credit: 303,562,776
RAC: 0
Message 61182 - Posted: 22 Feb 2014, 3:41:26 UTC

This is the information needed to see the problem:

<core_client_version>7.2.39</core_client_version>
<![CDATA[
<message>
The handle is invalid.
(0x6) - exit code 6 (0x6)
</message>
<stderr_txt>
<search_application> milkyway_nbody 1.40 Windows x86_64 double OpenMP, Crlibm </search_application>
Using OpenMP 1 max threads on a system with 4 processors
Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'nbody_checkpoint_tmp_5532' to 'nbody_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to update checkpoint 'nbody_checkpoint' with temporary (2): No such file or directory
Failed to write checkpoint
Error running system: NBODY_CHECKPOINT_ERROR (64)
20:26:32 (5532): called boinc_finish

</stderr_txt>
]]>

and

<core_client_version>7.2.39</core_client_version>
<![CDATA[
<message>
Incorrect function.
(0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
<search_application> milkyway_separation 1.00 Windows x86_64 double </search_application>
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Error reading astronomy parameters from file 'astronomy_parameters.txt'
Trying old parameters file
Using SSE4.1 path
Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to move file 'separation_checkpoint_tmp' to 'separation_checkpoint' (6832): This object is not allowed to be opened in a transaction.

Failed to update checkpoint file ('separation_checkpoint_tmp' to 'separation_checkpoint') (2): No such file or directory
Write checkpoint failed
17:26:02 (1556): called boinc_finish

</stderr_txt>
]]>

anybody?
ID: 61182 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 61205 - Posted: 23 Feb 2014, 16:24:25 UTC - in response to Message 61182.  

Not something to do with needing admin rights is it?

Btw this is clearly a different problem to the theme & title of this thread (errors with MW v1.02 WUs only), should of started a new 1 really
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 61205 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 61207 - Posted: 23 Feb 2014, 17:00:13 UTC - in response to Message 61168.  
Last modified: 23 Feb 2014, 17:00:42 UTC


Didn't know mod fit WUs gave less credit/hr, I wonder why they do that?


Partial answer to my own question lol, I noticed when the MW v1.02 WUs are running, power draw at the wall goes up a lot on my HD 5850.
Just now total power draw went from ~267w running MW sep. mod. fit to ~291w running MW v1.02! (think I'll start a new thread about that).
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 61207 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,938,332
RAC: 22,876
Message 61216 - Posted: 23 Feb 2014, 19:48:37 UTC - in response to Message 61205.  

Not something to do with needing admin rights is it?

Btw this is clearly a different problem to the theme & title of this thread (errors with MW v1.02 WUs only), should of started a new 1 really


It sounds like it, which would lead back to the Service install the person did.
ID: 61216 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Seraphim401

Send message
Joined: 3 Apr 10
Posts: 8
Credit: 27,124,250
RAC: 0
Message 61316 - Posted: 3 Mar 2014, 20:28:04 UTC - in response to Message 61207.  


Didn't know mod fit WUs gave less credit/hr, I wonder why they do that?


Partial answer to my own question lol, I noticed when the MW v1.02 WUs are running, power draw at the wall goes up a lot on my HD 5850.
Just now total power draw went from ~267w running MW sep. mod. fit to ~291w running MW v1.02! (think I'll start a new thread about that).


Yeah I noticed that to.I hope they lay off the aggression a bit.
There is no need for the gpu to be running at 99% for a WU.
Just my humble opinion.
ID: 61316 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 61336 - Posted: 6 Mar 2014, 19:12:54 UTC

The other WUs also run the GPU at ~99% too but still manage to draw less power, I guess they use less of the GPU.

Btw the GPU load is supposed to be ~99% to make the best use of our GPUs ;).
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 61336 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 61347 - Posted: 7 Mar 2014, 20:08:30 UTC
Last modified: 7 Mar 2014, 20:16:52 UTC

It seems there is a fix to get (at least) Cat 13.9 drivers to work with MW v1.02 afterall! :)

(some grammar edited ;) )

***** Now running on Radeon hd 4870 opencl boinc *****

Here is what happened with CAT 13.9 legacy installed

First checked graphics card with gpu-z, when first started gpu-z gave the following warning :-

quote
ATI OpenCL driver bug detected, skipping OpenCL detection.
Uninstall the AMD STREam SDK.................
end quote

I removed AMD app sdk
through CAT program in add remove programs then installed AMD app sdk 2.7
name - AMD-APP-SDK-v2.7-Windows-641

Same problem in gpu-z and boinc

put the above "quote" into search engine & it lead me to this website.

http://mrlithium.blogspot.com/2013/03/opencl-hassles-ati-opencl-driver-bug.html

It had the following fix

Quoting from Mrlithium's blog

THE FIX:
1. Uninstall any AMD Stream SDK you have right now.
2. Delete amdocl.dll and opencl.dll from C:\windows\system32 and C:\windows\Syswow64.
3. Reinstall the AMD Stream SDK you have. Does not matter which one (I think)
4. Check to make sure the amdocl.dll and opencl.dll files you just deleted have not come back.
5. Check in C:\Program Files (x86)\AMD APP\bin\x86 and C:\Program Files (x86)\AMD APP\bin\x86_64 to make sure the files are there in that location instead. This is where they should be.
6. Try using GPU-z to see if there is a tickmark next to OpenCL. If there is, all should be well, and you can continue to use Reaper/CGminer/GUIMiner to mine bitcoins/litecoins/etc. Or any OpenCL program for that matter.
7. For some reason I had to delete the kernel files that reaper/cgminer had generated such as litecoin-reaperv13.Cayman-256-5760-2.bin but I am not sure that this was due to OpenCL because I tried unlocking my Radeon 6950 GPU to a 6970 (and succeeded, link here) So it may have just been that the number of shaders or something was wrong.

End quote from Mrlithium's blog

**** I only did step 2 alone. ******

Have not restarted computer yet.
Shut down boinc
Restarted boinc it then showed opencl 1.0 support for 4870 in the start up log.
It seems I now also have opencl cpu support version 2.0 sse2 now, I don't recall seeing that before
I think that happened from the developer amd app sdk 2.7 install

Able to run opencl in MW
Able to run opencl in Seti Astropulse opencl 1.00
Able to run opencl primegrid opencl
that is so far.......
None of which could be ran before

Running CAT 13.9
with developer SDK 2.7
gpu-z running with opencl checked.

Thanks to Mr.Lithium's blog and his efforts for trying to make some coins.

Why those drivers stopped things, and how they got there I don't know (maybe from previous graphics card, from igp before that, from previous driver installs, from original win7, don't know). I now know that step 2 alone has worked for my scenario.

Everything so far seems to be running normal, will inform if things change.

Venz out


From this thread http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3500&postid=61344
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 61347 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Darren

Send message
Joined: 13 Jun 09
Posts: 1
Credit: 372,474
RAC: 0
Message 61442 - Posted: 25 Mar 2014, 23:54:22 UTC - in response to Message 61347.  

I'm using Catalyst 13.12 (clean install) and still getting the same issue.

I posted a topic on the AMD Drivers forum... maybe that will help get some traction on this issue :(

http://forums.amd.com/game/messageview.cfm?catid=454&threadid=172625&enterthread=y
ID: 61442 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,938,332
RAC: 22,876
Message 61443 - Posted: 26 Mar 2014, 11:24:50 UTC - in response to Message 61442.  

I'm using Catalyst 13.12 (clean install) and still getting the same issue.

I posted a topic on the AMD Drivers forum... maybe that will help get some traction on this issue :(

http://forums.amd.com/game/messageview.cfm?catid=454&threadid=172625&enterthread=y


Have you tried the 13.9 set of drivers? If not why not, as it is reported that they do in fact work.
ID: 61443 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 61445 - Posted: 26 Mar 2014, 18:42:26 UTC - in response to Message 61442.  
Last modified: 26 Mar 2014, 18:43:51 UTC

I'm using Catalyst 13.12 (clean install) and still getting the same issue.

I posted a topic on the AMD Drivers forum... maybe that will help get some traction on this issue :(

http://forums.amd.com/game/messageview.cfm?catid=454&threadid=172625&enterthread=y


Linkified :P

Doesn't hurt to try, although I can't help thinking it's a MW issue.
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 61445 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 62268 - Posted: 7 Sep 2014, 11:31:04 UTC - in response to Message 61445.  

TTT

This thread should be stickified.
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 62268 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 63316 - Posted: 2 Apr 2015, 18:25:25 UTC

Anyone know if this MW problem is fixed with more recent drivers?
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 63316 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
grumpy

Send message
Joined: 14 Dec 07
Posts: 9
Credit: 10,208,774
RAC: 614
Message 63330 - Posted: 7 Apr 2015, 21:02:47 UTC - in response to Message 63316.  
Last modified: 7 Apr 2015, 21:03:33 UTC

No!...still problems
HD5850 + cat driver 14.5
ID: 63330 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[TA]Assimilator1
Avatar

Send message
Joined: 22 Jan 11
Posts: 375
Credit: 64,657,871
RAC: 0
Message 63334 - Posted: 8 Apr 2015, 17:02:22 UTC - in response to Message 63330.  

FGS!

Thx for the info anyway :)
Team AnandTech - SETI@H, DPAD, F@H, MW@H, A@H, LHC, POGS, R@H, Einstein@H, DHEP, WCG

Main rig - Ryzen 5 3600, MSI B450 G.Pro C. AC, RTX 3060Ti 8GB, 32GB DDR4 3200, Win 10 64bit
2nd rig - i7 4930k @4.1 GHz, HD 7870 XT 3GB(DS), 16GB DDR3 1866, Win7
ID: 63334 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · Next

Message boards : Number crunching : All Milkyway@Home 1.02 tasks ending in computation error on HD6950.

©2024 Astroinformatics Group