And another of which both copies failed quickly, this time receiving a validate error: CASP9_ax_benchmark_hybridization_run25_T0593_SAVE_ALL_OUT_IGNORE_THE_REST_41774_720

CPU time 155.5573

# cpu_run_time_pref: 28800
======================================================
DONE :: 99 starting structures 1201 cpu seconds
This process generated 99 decoys from 99 attempts
======================================================
BOINC :: WS_max 0

CASP9_bk_benchmark_hybridization_run37_T0629_SAVE_ALL_OUT_IGNORE_THE_REST_44182_265_0

Maximum disk usage exceeded

CPU time 18551.66


Do these need a higher maximum disk usage or low model limit?


Best,
Snags

Jobs T0_ga had bad inputs. (Except T0515_ga.) I've removed these jobs from boinc and will do a better job testing them before submitting as T0_gb later today.

Thanks for your support,
-TJ

My last three 3.22 workunits each quickly gave a computation error, on both of my desktops.

https://boinc.bakerlab.org/rosetta/result.php?resultid=489997894

Setting up graphics native ...
BOINC:: Worker startup.
Starting watchdog...
Watchdog active.
======================================================
DONE :: 0 starting structures 1201 cpu seconds
This process generated 0 decoys from 0 attempts
======================================================
BOINC :: WS_max 0

BOINC :: Watchdog shutting down...
BOINC :: BOINC support services shutting down cleanly ...
called boinc_finish



upload failure:
T0597_ga_rs_stg0_lrlx_t000__casp9_SAVE_ALL_OUT_44122_335_0_0
-161



]]>

application version 3.22


https://boinc.bakerlab.org/rosetta/result.php?resultid=489997887

https://boinc.bakerlab.org/rosetta/result.php?resultid=489988179


Any known reason for this? Both computers were returning valid results before these.

Hi.

I had five of these fail with the same error, i was the second to get them.

T0609_ga_rs_stg0_lrlx_t000__casp9_SAVE_ALL_OUT_44131_394_1

</stderr_txt>
<message>
<file_xfer_error>
<file_name>T0609_ga_rs_stg0_lrlx_t000__casp9_SAVE_ALL_OUT_44131_394_1_0</file_name>
<error_code>-161</error_code>
</file_xfer_error>

</message>

---

Just joined Rosetta@Home, and so far all of my WU's are ending with client errors. I'm using version 6.12.34 (x64) of BOINC Manager, and Rosetta Mini 3.22.
I'm running the 64-bit version of Windows 7 Pro, and I have a NVidia GT 560 Ti (with GPU enabled).

Anyone got any ideas?

Just joined Rosetta@Home, and so far all of my WU's are ending with client errors. I'm using version 6.12.34 (x64) of BOINC Manager, and Rosetta Mini 3.22.
I'm running the 64-bit version of Windows 7 Pro, and I have a NVidia GT 560 Ti (with GPU enabled).

Anyone got any ideas?

first thought is that the GPU might be causing interference with the processing.
this project is a strictly CPU project.
also if you are OC'ing the cpu to much the project code does not like that either.
could also be a memory problem as well, not enough memory? but if your running gpu I don't think thats the issue.

Just joined Rosetta@Home, and so far all of my WU's are ending with client errors. I'm using version 6.12.34 (x64) of BOINC Manager, and Rosetta Mini 3.22.
I'm running the 64-bit version of Windows 7 Pro, and I have a NVidia GT 560 Ti (with GPU enabled).

Anyone got any ideas?

first thought is that the GPU might be causing interference with the processing.
this project is a strictly CPU project.
also if you are OC'ing the cpu to much the project code does not like that either.
could also be a memory problem as well, not enough memory? but if your running gpu I don't think thats the issue.

edit - checked your results. the casp9 units has problems and were pulled from the server. you ran some of them before they were pulled. you had some file transfer problems, nothing you could do about that. also some of the tasks beginning with T are buggy, i have a small handful of them I have been running that have gone buggy on me. so other than gpu causing issues, which I have never heard of before, and maybe some overclocking (which also does not seem to be a problem I think)issues, your just unlucky and getting alot of buggy units. things will change as time progresses. be patient.

first thought is that the GPU might be causing interference with the processing.
this project is a strictly CPU project.
also if you are OC'ing the cpu to much the project code does not like that either.
could also be a memory problem as well, not enough memory? but if your running gpu I don't think thats the issue.

edit - checked your results. the casp9 units has problems and were pulled from the server. you ran some of them before they were pulled. you had some file transfer problems, nothing you could do about that. also some of the tasks beginning with T are buggy, i have a small handful of them I have been running that have gone buggy on me. so other than gpu causing issues, which I have never heard of before, and maybe some overclocking (which also does not seem to be a problem I think)issues, your just unlucky and getting alot of buggy units. things will change as time progresses. be patient.

Thanks for the response -- I saw the thread about possible problems running with the GPU engaged, but I'm getting such good results with GPU-targeted SETI@Home WU that I really don't want to disengage it. I'm hoping that, if it is a GPU issue, that someone will solve the problem with an update.

In the meantime, I'll just keep cranking and keep an eye on the results. If the problem persists after a week or so, I'll just turn off Rosetta@Home until it gets fixed.

first thought is that the GPU might be causing interference with the processing.
this project is a strictly CPU project.
also if you are OC'ing the cpu to much the project code does not like that either.
could also be a memory problem as well, not enough memory? but if your running gpu I don't think thats the issue.

edit - checked your results. the casp9 units has problems and were pulled from the server. you ran some of them before they were pulled. you had some file transfer problems, nothing you could do about that. also some of the tasks beginning with T are buggy, i have a small handful of them I have been running that have gone buggy on me. so other than gpu causing issues, which I have never heard of before, and maybe some overclocking (which also does not seem to be a problem I think)issues, your just unlucky and getting alot of buggy units. things will change as time progresses. be patient.

Thanks for the response -- I saw the thread about possible problems running with the GPU engaged, but I'm getting such good results with GPU-targeted SETI@Home WU that I really don't want to disengage it. I'm hoping that, if it is a GPU issue, that someone will solve the problem with an update.

In the meantime, I'll just keep cranking and keep an eye on the results. If the problem persists after a week or so, I'll just turn off Rosetta@Home until it gets fixed.

That could be a long time.
Those kinds of fixes are a long time in coming.
If you got another system that is cpu only, run that on rosetta.
There might be other projects out there that have better coding for cpu/gpu where could run both of them at the same time.

Anyone got any ideas?

Looks like you're experiencing the same successful completion (BOINC support services shutting down cleanly ... called boinc_finish)/Exit status: 0x0/Outcome: Client error issue that others have experienced. Best we can tell, there's an issue with boinc transferring the results to the server.

I unfortunately don't have a solution to give, but if you're willing to turn on some debugging flags and tell us about the output, it may help us track down the issue. (See this post.)

William Blakemore asked:

OK ... it's really dumb question time ....

I've looked in both the "program files" and "ProgramData/BOINC" directories on my Win 7 machine, and I can't find any files named cc_config.xml. Where should I be looking for it?

The file is optional. So if it does not exist all of the flags get their default values.

---

Rosetta Moderator: Mod.Sense

William Blakemore asked:

OK ... it's really dumb question time ....

I've looked in both the "program files" and "ProgramData/BOINC" directories on my Win 7 machine, and I can't find any files named cc_config.xml. Where should I be looking for it?

The file is optional. So if it does not exist all of the flags get their default values.

I've completed the first phase of my testing, and I've conclusively proved that I don't know .xml well enough to create my debug cc_config.xml file.

Would it possible for you to post either a listing of the file, or a link to one, containing all the debug flags you want turned on?

William & anyone else that needs or wants to do a config xml file have a read of this page, get back to us if you need more help with it.

Sorry tried to make it a link, but no go.

http://boinc.berkeley.edu/wiki/Client_configuration

---

I've completed the first phase of my testing, and I've conclusively proved that I don't know .xml well enough to create my debug cc_config.xml file.

Would it possible for you to post either a listing of the file, or a link to one, containing all the debug flags you want turned on?

Here is a copy of one of my cc_config.xml files:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>

To use it just open NOTEPAD in Windows and copy and paste the text above into a new file. When it comes time to save it save it as a txt file type but with the .xml extension. Using Word or anything else adds hidden coding to the file and it won't work. When it is saved just copy it to the /ProgramData/Boinc/Data directory. ProgramData is normally hidden so you will need to set your pc to view hidden folders.

BTW the above file just tells Boinc to use ALL the gpu's it finds in the machine. It is irrelevant here at Rosetta, but works fine on projects that do use the gpu for crunching.