Welcome to MilkyWay@home

Request help updating app_info.xml for Linux

Message boards : Number crunching : Request help updating app_info.xml for Linux
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58159 - Posted: 7 May 2013, 11:04:57 UTC

I was wondering if someone could look over my app_info.xml file (for Linux) and help me figure out what I'm doing wrong.

I'm not a programmer and am trying to update my previously working file but am wondering if there have been some fundamental changes to the syntax that I'm not picking up.

The two basic problems are: 1) I can no longer run multiple simultaneous WUs on the GPUs, and 2) when I try to run the MW with the app_info.xml file I get messages like MilkyWay_simulation 1.02 file cannot be found. If I delete the xml file, and re-start Boinc-client, there it is.

Here's the app_info.xml as it stands right now. Any help would be greatly appreciated.

I did try the bare-bones app_config.xml approach mentioned in another thread but I can't get it to run multiple simultaneous WUs on the GPUs which is why I went back to the app_info.xml which now doesn't work.

Thanks,
Steve

<app_info>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>milkyway_separation_1.01_x86_64-pc-linux-gnu</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>1.01</version_num>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_separation_1.01_x86_64-pc-linux-gnu</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>milkyway_separation_1.02_x86_64-pc-linux-gnu__opencl_nvidia</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>1.02</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>cuda</plan_class>
<coproc>
<type>CUDA</type>
<count>0.5</count>
</coproc>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_separation_1.02_x86_64-pc-linux-gnu__opencl_nvidia</file_name>
<main_program/>
</file_ref>
</app>
<name>milkyway_nbody</name>
</app>
<file_info>
<name>milkyway_nbody_1.08_x86_64-pc-linux-gnu__mt</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway_nbody</app_name>
<version_num>1.08</version_num>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_nbody_1.08_x86_64-pc-linux-gnu__mt</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>milkyway_nbody</name>
</app>
<file_info>
<name>milkyway_nbody_1.09_x86_64-pc-linux-gnu__opencl_nvidia</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway_nbody</app_name>
<version_num>1.09</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>cuda</plan_class>
<coproc>
<type>CUDA</type>
<count>0.5</count>
</coproc>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_nbody_1.09_x86_64-pc-linux-gnu__opencl_nvidia</file_name>
<main_program/>
</file_ref>
</app_version>
</app_info>
ID: 58159 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 8 May 09
Posts: 3315
Credit: 519,938,432
RAC: 22,812
Message 58160 - Posted: 7 May 2013, 11:40:41 UTC - in response to Message 58159.  

I was wondering if someone could look over my app_info.xml file (for Linux) and help me figure out what I'm doing wrong.

I'm not a programmer and am trying to update my previously working file but am wondering if there have been some fundamental changes to the syntax that I'm not picking up.

The two basic problems are: 1) I can no longer run multiple simultaneous WUs on the GPUs, and 2) when I try to run the MW with the app_info.xml file I get messages like MilkyWay_simulation 1.02 file cannot be found. If I delete the xml file, and re-start Boinc-client, there it is.

Here's the app_info.xml as it stands right now. Any help would be greatly appreciated.

I did try the bare-bones app_config.xml approach mentioned in another thread but I can't get it to run multiple simultaneous WUs on the GPUs which is why I went back to the app_info.xml which now doesn't work.

Thanks,
Steve


You are running too old of a Boinc version, 6.12.33, to use the new app_config files, that's why it didn't work. You MUST be using no earlier then 7.0.40 to use an app_config.xml file, and Boinc is now upto 7.0.65 as the release version for Linux platforms. I would upgrade Boinc then try the app_config file again as it eliminates the need to keep up with all the different file names.
ID: 58160 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58163 - Posted: 7 May 2013, 12:37:47 UTC - in response to Message 58160.  

Thanks for the help. I've always relied on the official Linux distro repositories for my Boinc apps. I just checked and they're still stuck there on version 6, or in the case of SETI, version 5.

Let me see about upgrading.

Regards,
Steve
ID: 58163 · 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 58173 - Posted: 8 May 2013, 12:09:12 UTC - in response to Message 58163.  

In case you want or have to stick with BOINC 6

There is at least 1 obvious error in your app_info:

<cmdline></cmdline>
<file_ref>
<file_name>milkyway_separation_1.02_x86_64-pc-linux-gnu__opencl_nvidia</file_name>
<main_program/>
</file_ref>
</app>
<name>milkyway_nbody</name>
</app>
<file_info>
<name>milkyway_nbody_1.08_x86_64-pc-linux-gnu__mt</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway_nbody</app_name>
<version_num>1.08</version_num>

The red marked part should be

</app_version>
<app>


Cleaning it from old versions and structure it a bit (i.e. empty lines between the different blocks) helps to find those wrong/missing tags quicker.
ID: 58173 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58208 - Posted: 11 May 2013, 14:51:32 UTC - in response to Message 58173.  

Thanks for the feedback. I'll make the change. As far as running the apps, the problem has resolved itself on the v.6 machines and the new v.7 installation HOWEVER, in all cases, I still cannot run multiple simultaneous GPU WUs.

I'll see if this change helps with the v.6 machines.

Thanks again for the suggestion.

Regards,
Steve
ID: 58208 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58209 - Posted: 11 May 2013, 15:17:03 UTC - in response to Message 58173.  

LEN LE/GE,

Thank you, thank you, thank you, thank you, thank you, .....

That one change looks like its solved all of the problems including getting the GPUs to process multiple, simultaneous WUs on one of the machines so far. I'll have to wait until the change cycles through on the other machines but hopefully it will work there as well.

Thanks so much for the help.

Regards,
Steve
ID: 58209 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58226 - Posted: 12 May 2013, 13:17:57 UTC

While the machine with the Fermi 460 and the 420 cards is now running great with multiple, simultaneous GPU WUs being processed using the corrected app_info.xml, I'm still having a problem with the other two machines and the GPUs.

To recap - all machines are running Linux Mint with NVIDIA GPUs. All machines are seeing and apparently running all of the types of WUs sent without problem. The snag on the remaining two machines is that they can't process multiple simultaneous GPU WUs. I've tried both the fixed app_info.xml that's working on Machine one as well as an app_config.xml mentioned on the message board (both with and without the <max_concurrent>4</max_concurrent> line) but no luck.

One machine has a single Fermi 460 while the other has a Fermi 550I (dual GPU but only one seems to be seen no matter what I do.

The machine with the Fermi 550I is still on BOINC v.6 but I'm trying to find a reliable repository to upgrade from. The machine with the single Fermi 460 (running Linux Mint Debian Edition) does offer Boinc V.7 and that's what's installed but it doesn't make any difference.

Is there another config file somewhere that still has to be checked?

FWIW, Einstein is working with no problems at all on the GPU front although they went to a different approach some months back where now you just check off on your E@Home prerences page if you want to run simultaneous GPU WUs and it's taken care of there. No app_anything.xml files.

Thanks for any help you can provide.

Regards,
Steve
ID: 58226 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 58227 - Posted: 12 May 2013, 14:06:31 UTC - in response to Message 58226.  
Last modified: 12 May 2013, 14:12:20 UTC

On the problem machines, check to see if the line below is pasted into cc_config in the BOINC directory that's below Program Data parent directory.

<use_all_gpus>1</use_all_gpus>

Probably will not fix both, but hopefully fixes one, move onto the last after this is inserted into cc_config

If cc_config.xml does not exist in the BOINC directory, create it with notepad (make sure its saved as .xml not .txt). As an example mine is below:

<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<save_stats_days>180</save_stats_days>
<max_file_xfers>48</max_file_xfers>
<max_file_xfers_per_project>36</max_file_xfers_per_project>
</options>
</cc_config>

Regards
Zy
ID: 58227 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile arkayn
Avatar

Send message
Joined: 14 Feb 09
Posts: 999
Credit: 74,932,619
RAC: 0
Message 58233 - Posted: 12 May 2013, 20:09:33 UTC - in response to Message 58226.  



The machine with the Fermi 550I is still on BOINC v.6 but I'm trying to find a reliable repository to upgrade from.
Regards,
Steve


This is the one that keeps up to date with the current builds put out by Berkeley.

sudo add-apt-repository ppa:costamagnagianfranco/boinc
sudo apt-get update
sudo apt-get upgrade

I used it to install on my Linux test machine, which spends most of its time in Windows.
ID: 58233 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58240 - Posted: 13 May 2013, 12:06:52 UTC - in response to Message 58227.  

Zy,

Thanks for the suggestions. Unfortunately, no joy. I'm not getting any error messages anywhere. I guess the good news is that at least I am processing MW GPU WUs, just not multiple ones simultaneously.

Thanks again for effort.

Regards,
Steve
ID: 58240 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58241 - Posted: 13 May 2013, 12:12:50 UTC - in response to Message 58233.  

Arkayn,

Thanks for the suggestion. Everything looked like it went fine with adding the repository and I can see it listed in Synaptic after the update/upgrade, except that when I have Synaptic search its listings for "BOINC", it still shows the old v.6 ones.

I think I'm going to try a new installation on a different version of Linux and see what happens since all of my machines are virtually identical down to the brands and models of the components.

The good news is that, at least, I am processing MW GPU WUs, just not multiple ones simultaneously.

Thanks again.

Regards,
Steve
ID: 58241 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 58242 - Posted: 13 May 2013, 12:20:22 UTC - in response to Message 58240.  

Could try two other things if not all ready done so - strap the two together with a Bridge Strap, and/or enable Crossfire (cant remember the CUDA word for it - escapes me at present :) .... ) in the GPU Control Panel (Catalyst Control Centre for AMDs).

If the previous and these two don't work - buy some incense and beg forgiveness to the gods for upsetting them ..... because there is little else that will stop it given a clean uptodate set of drivers :)
ID: 58242 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mmstick
Avatar

Send message
Joined: 23 Nov 09
Posts: 29
Credit: 17,119,258
RAC: 0
Message 58255 - Posted: 14 May 2013, 6:20:34 UTC
Last modified: 14 May 2013, 7:13:42 UTC

I'm having similar issues. I can't get app_config.xml to work on Linux. BOINC manager says it detects app_config.xml but it is ignoring all the parameters.

Update: Found solution

<app_config>
<app>
<name>milkyway</name>
<max_concurrent>4</max_concurrent>
<gpu_versions>
<gpu_usage>.25</gpu_usage>
<cpu_usage>.25</cpu_usage>
</gpu_versions>
</app>
</app_config>
ID: 58255 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58259 - Posted: 14 May 2013, 11:19:36 UTC - in response to Message 58242.  

At this point, the amount of incense that I would have to burn would violate numerous environmental laws!

Regards,
Steve
ID: 58259 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58260 - Posted: 14 May 2013, 12:13:55 UTC - in response to Message 58255.  



Thanks for the suggestion but still no luck. I'm wondering how I'm able to run multiple simultaneous GPU WUs on that one machine given that we're talking about the same brand of Fermi 460 card in two of the machines. One works, the other only processes one GPU WU at a time; a real waste of HW power.

I'm curious as to how many others are having this problem, how many have solved it, or how many have given up and whether OS makes a difference.

Oh well, time to go light more incense!

Regards,
Steve
ID: 58260 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 58268 - Posted: 15 May 2013, 3:11:32 UTC

Straw clutch time .......

Are you creating these ones inside Notepad, and not another editor? Notepad is needed to make sure hidden control characters are not inserted aka word processor software - as these can and are misinterpreted by a BOINC program, and it all falls over.

Use of Notepad is essential - its not optional.
ID: 58268 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58273 - Posted: 15 May 2013, 12:14:44 UTC - in response to Message 58268.  

Since I'm on Linux, the choices are gedit or Pluma which are basic text editors. Since I've used them in the past when I had working app_info.xml's, I don't think they're the problem.

Just on the off chance, I just tried things again using app_config.xml and then an app_info.xml boyh created with Notepad under Win2000. No change.

Thanks anyway.

Regards,
Steve
ID: 58273 · 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 58275 - Posted: 15 May 2013, 18:55:27 UTC

When you try running with an app_config file, do you open the Boinc manager and watch it to see how many tasks it is actually crunching?
ID: 58275 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
europa

Send message
Joined: 29 Oct 10
Posts: 89
Credit: 39,246,947
RAC: 0
Message 58278 - Posted: 15 May 2013, 20:35:10 UTC

Oh,yes. The cpu wu's are all happy on their cores and the GPU is crunching just one of the available cuda wu's at a time.

Regards,
Steve
ID: 58278 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 58280 - Posted: 15 May 2013, 20:51:00 UTC
Last modified: 15 May 2013, 20:59:32 UTC

Could you cut & paste your complete cc_config in here to have a peek at?

- I assume both cards are of the same type and version?

- Does Catalyst Control Centre show hardware enabled for both cards (information-hardware tab)....

- Inside the Performance - Overdrive option at the top of the options showing up mid-bottom screen, do both cards appear inside a drop down box above the small illustration pics?
ID: 58280 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
1 · 2 · Next

Message boards : Number crunching : Request help updating app_info.xml for Linux

©2024 Astroinformatics Group