It does not appear that BM is adhering to the cpu_usage and gpu_usage settings stated in the app_config.xml file. I have also created a scenario (136) in the emulator for Collatz WUs as opposed to MilkyWay WUs.

Any suggestions/observations? Thanks in advance.

Environment:

1/20/2015 8:45:03 AM | | Starting BOINC client version 7.4.36 for windows_x86_64
1/20/2015 8:45:03 AM | | log flags: file_xfer, sched_ops, task, cpu_sched, unparsed_xml
1/20/2015 8:45:03 AM | | Libraries: libcurl/7.39.0 OpenSSL/1.0.1j zlib/1.2.8
1/20/2015 8:45:03 AM | | Data directory: C:\ProgramData\BOINC
1/20/2015 8:45:03 AM | | CUDA: NVIDIA GPU 0: NVS 4200M (driver version 341.05, CUDA version 6.5, compute capability 2.1, 1024MB, 971MB available, 156 GFLOPS peak)
1/20/2015 8:45:03 AM | | OpenCL: NVIDIA GPU 0: NVS 4200M (driver version 341.05, device version OpenCL 1.1 CUDA, 1024MB, 971MB available, 156 GFLOPS peak)
1/20/2015 8:45:03 AM | | OpenCL CPU: Intel(R) Core(TM) i5-2520M CPU @ 2.50GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 3.0.1.15216, device version OpenCL 1.2 (Build 80752))
1/20/2015 8:45:03 AM | | Processor: 4 GenuineIntel Intel(R) Core(TM) i5-2520M CPU @ 2.50GHz [Family 6 Model 42 Stepping 7]
1/20/2015 8:45:03 AM | | 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 htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt aes syscall nx lm avx vmx smx tm2 pbe
1/20/2015 8:45:03 AM | | OS: Microsoft Windows 7: Professional x64 Edition, Service Pack 1, (06.01.7601.00)
1/20/2015 8:45:03 AM | | Memory: 7.89 GB physical, 15.77 GB virtual
1/20/2015 8:45:03 AM | | Disk: 448.96 GB total, 252.15 GB free
1/20/2015 8:45:03 AM | | Local time is UTC -6 hours
1/20/2015 8:45:03 AM | Albert@Home | Found app_config.xml
1/20/2015 8:45:03 AM | Asteroids@home | Found app_config.xml
1/20/2015 8:45:03 AM | Collatz Conjecture | Found app_config.xml
1/20/2015 8:45:03 AM | Einstein@Home | Found app_config.xml
1/20/2015 8:45:03 AM | Milkyway@Home | Found app_config.xml
1/20/2015 8:45:03 AM | Moo! Wrapper | Found app_config.xml
1/20/2015 8:45:03 AM | primaboinca | Found app_config.xml
1/20/2015 8:45:03 AM | PrimeGrid | Found app_config.xml
1/20/2015 8:45:03 AM | | Config: report completed tasks immediately
1/20/2015 8:45:03 AM | | Config: use all coprocessors
1/20/2015 8:45:03 AM | | Config: event log limit disabled

"app_config.xml" for Milkyway:

<app_config>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway_separation__modified_fit</name>
<user_friendly_name>Milkyway@Home Separation (Modified Fit)</user_friendly_name>
</app>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway_nbody</name>
<user_friendly_name>MilkyWay@Home N-Body Simulation</user_friendly_name>
</app>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway</name>
<user_friendly_name>MilkyWay@Home</user_friendly_name>
</app>
</app_config>

Config File Re-read:

1/22/2015 2:38:31 PM | | Re-reading cc_config.xml
1/22/2015 2:38:31 PM | | Not using a proxy
1/22/2015 2:38:31 PM | | Config: report completed tasks immediately
1/22/2015 2:38:31 PM | | Config: use all coprocessors
1/22/2015 2:38:31 PM | | Config: event log limit disabled
1/22/2015 2:38:31 PM | | log flags: file_xfer, sched_ops, task, coproc_debug, cpu_sched, cpu_sched_debug
1/22/2015 2:38:31 PM | | log flags: statefile_debug, unparsed_xml, work_fetch_debug
1/22/2015 2:38:31 PM | Albert@Home | Found app_config.xml
1/22/2015 2:38:31 PM | Einstein@Home | Found app_config.xml
1/22/2015 2:38:31 PM | Asteroids@home | Found app_config.xml
1/22/2015 2:38:31 PM | Collatz Conjecture | Found app_config.xml
1/22/2015 2:38:31 PM | Milkyway@Home | Found app_config.xml
1/22/2015 2:38:31 PM | Moo! Wrapper | Found app_config.xml
1/22/2015 2:38:31 PM | primaboinca | Found app_config.xml
1/22/2015 2:38:31 PM | PrimeGrid | Found app_config.xml

"cpu_sched and coproc" debug:

1/22/2015 2:39:32 PM | | [cpu_sched_debug] Request CPU reschedule: periodic CPU scheduling
1/22/2015 2:39:32 PM | | [cpu_sched_debug] schedule_cpus(): start
1/22/2015 2:39:32 PM | Milkyway@Home | [cpu_sched_debug] add to run list: de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0 (coprocessor job, FIFO) (prio 0.000000)
1/22/2015 2:39:32 PM | Milkyway@Home | [cpu_sched_debug] reserving 1.000000 of coproc NVIDIA
1/22/2015 2:39:32 PM | PrimeGrid | [cpu_sched_debug] add to run list: genefer_1048576_421768_3 (CPU job, EDF) (prio 0.000000)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] add to run list: dnetc_r72_1421936572_9_9_0 (CPU job, priority order) (prio 0.000000)
1/22/2015 2:39:32 PM | primaboinca | [cpu_sched_debug] add to run list: uc_1421654515_1036988121219_0 (CPU job, priority order) (prio 0.000000)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] add to run list: dnetc_r72_1421945189_9_9_0 (CPU job, priority order) (prio -1000.325290)
1/22/2015 2:39:32 PM | | [cpu_sched_debug] enforce_run_list(): start
1/22/2015 2:39:32 PM | | [cpu_sched_debug] preliminary job list:
1/22/2015 2:39:32 PM | Milkyway@Home | [cpu_sched_debug] 0: de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0 (MD: no; UTS: yes)
1/22/2015 2:39:32 PM | PrimeGrid | [cpu_sched_debug] 1: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] 2: dnetc_r72_1421936572_9_9_0 (MD: no; UTS: no)
1/22/2015 2:39:32 PM | primaboinca | [cpu_sched_debug] 3: uc_1421654515_1036988121219_0 (MD: no; UTS: no)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] 4: dnetc_r72_1421945189_9_9_0 (MD: no; UTS: yes)
1/22/2015 2:39:32 PM | | [cpu_sched_debug] final job list:
1/22/2015 2:39:32 PM | PrimeGrid | [cpu_sched_debug] 0: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 2:39:32 PM | Milkyway@Home | [cpu_sched_debug] 1: de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0 (MD: no; UTS: yes)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] 2: dnetc_r72_1421945189_9_9_0 (MD: no; UTS: yes)
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] 3: dnetc_r72_1421936572_9_9_0 (MD: no; UTS: no)
1/22/2015 2:39:32 PM | primaboinca | [cpu_sched_debug] 4: uc_1421654515_1036988121219_0 (MD: no; UTS: no)
1/22/2015 2:39:32 PM | Milkyway@Home | [coproc] NVIDIA instance 0; 1.000000 pending for de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0
1/22/2015 2:39:32 PM | Milkyway@Home | [coproc] NVIDIA instance 0: confirming 1.000000 instance for de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0
1/22/2015 2:39:32 PM | PrimeGrid | [cpu_sched_debug] scheduling genefer_1048576_421768_3 (high priority)
1/22/2015 2:39:32 PM | Milkyway@Home | [cpu_sched_debug] scheduling de_modfit_fast_15_3s_136_narrow2_2_1413455402_28306712_0
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] scheduling dnetc_r72_1421945189_9_9_0
1/22/2015 2:39:32 PM | Moo! Wrapper | [cpu_sched_debug] scheduling dnetc_r72_1421936572_9_9_0
1/22/2015 2:39:32 PM | primaboinca | [cpu_sched_debug] scheduling uc_1421654515_1036988121219_0
1/22/2015 2:39:32 PM | | [cpu_sched_debug] enforce_run_list: end

--
jon

I think we've had this one before.

'<user_friendly_name>' is not a valid tag specified in Application configuration.

Thanks, Richard. I'll purge all instances of the tag and try again. Report back shortly.

--
jon

Boinc Manager is the GUI, It doesn't and won't follow what in the app_config.xml, the Boinc client will through:

It does not appear that BM is adhering to the cpu_usage and gpu_usage settings stated in the app_config.xml file. I have also created a scenario (136) in the emulator for Collatz WUs as opposed to MilkyWay WUs.

Any suggestions/observations? Thanks in advance.

"app_config.xml" for Milkyway

<app_config>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway_separation__modified_fit</name>
<user_friendly_name>Milkyway@Home Separation (Modified Fit)</user_friendly_name>
</app>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway_nbody</name>
<user_friendly_name>MilkyWay@Home N-Body Simulation</user_friendly_name>
</app>
<app>
<fraction_done_exact />
<gpu_versions>
<cpu_usage>0.20</cpu_usage>
<gpu_usage>0.50</gpu_usage>
</gpu_versions>
<max_concurrent>2</max_concurrent>
<name>milkyway</name>
<user_friendly_name>MilkyWay@Home</user_friendly_name>
</app>
</app_config>

Your app_config.xml looks to be non-standard, the format should follow this (Don't put extra lines in with standards that don't exist (<user_friendly_name>), don't change the order):

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

<app_config>
[<app>
<name>uppercase</name>
<max_concurrent>1</max_concurrent>
[<fraction_done_exact/>]
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>.4</cpu_usage>
</gpu_versions>
</app>]
...
[<app_version>
<app_name>uppercase</app_name>
[<plan_class>mt</plan_class>]
[<avg_ncpus>x</avg_ncpus>]
[<ngpus>x</ngpus>]
[<cmdline>--nthreads 7</cmdline>]
</app_version>]
...
[<project_max_concurrent>N</project_max_concurrent>]
</app_config>

Claggy

UPDATE: As it happens a new Collatz WU has recently started and is now running with what appears to be the same issue (I don't have the "user_friendly_name" tag in the Collatz app_config file.

BTW, this behavior began to appear after I upgraded to 7.4.36 -- I had not seen it in earlier releases with the same app_config file. I scanned the git files looking to see if there was anything explicitly targeting cpu and gpu usage or app_config processing but couldn't find anything in the 7.4.36 release code that seemed to fit.

Here's the app_config:

<app_config>
<app>
<fraction_done_exact/>
<gpu_versions>
<cpu_usage>1.00</cpu_usage>
<gpu_usage>0.75</gpu_usage>
</gpu_versions>
<max_concurrent>4</max_concurrent>
<name>solo_collatz</name>
</app>
<app>
<fraction_done_exact/>
<gpu_versions>
<cpu_usage>1.00</cpu_usage>
<gpu_usage>0.75</gpu_usage>
</gpu_versions>
<max_concurrent>4</max_concurrent>
<name>mini_collatz</name>
</app>
<app>
<fraction_done_exact/>
<gpu_versions>
<cpu_usage>1.00</cpu_usage>
<gpu_usage>1.00</gpu_usage>
</gpu_versions>
<max_concurrent>4</max_concurrent>
<name>large_collatz</name>
</app>
</app_config>

Here's the log scheduling snippet:

1/22/2015 3:46:22 PM | | [cpu_sched_debug] Request CPU reschedule: periodic CPU scheduling
1/22/2015 3:46:22 PM | | [cpu_sched_debug] schedule_cpus(): start
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: solo_collatz_2393014943811557401715_1649267441664_0 (coprocessor job, FIFO) (prio -0.820975)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] reserving 1.000000 of coproc NVIDIA
1/22/2015 3:46:22 PM | PrimeGrid | [cpu_sched_debug] add to run list: genefer_1048576_421768_3 (CPU job, EDF) (prio -1000.081342)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438268729894003_103079215104_0 (CPU job, priority order) (prio -0.851499)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438062571463795_103079215104_0 (CPU job, priority order) (prio -0.854508)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438990284399731_103079215104_0 (CPU job, priority order) (prio -0.857517)
1/22/2015 3:46:22 PM | | [cpu_sched_debug] enforce_run_list(): start
1/22/2015 3:46:22 PM | | [cpu_sched_debug] preliminary job list:
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 0: solo_collatz_2393014943811557401715_1649267441664_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | PrimeGrid | [cpu_sched_debug] 1: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 2: mini_collatz_2393020438268729894003_103079215104_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 3: mini_collatz_2393020438062571463795_103079215104_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 4: mini_collatz_2393020438990284399731_103079215104_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | | [cpu_sched_debug] final job list:
1/22/2015 3:46:22 PM | PrimeGrid | [cpu_sched_debug] 0: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 1: solo_collatz_2393014943811557401715_1649267441664_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 2: mini_collatz_2393020438062571463795_103079215104_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 3: mini_collatz_2393020438990284399731_103079215104_0 (MD: no; UTS: yes)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] 4: mini_collatz_2393020438268729894003_103079215104_0 (MD: no; UTS: no)
1/22/2015 3:46:22 PM | Collatz Conjecture | [coproc] NVIDIA instance 0; 1.000000 pending for solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 3:46:22 PM | Collatz Conjecture | [coproc] NVIDIA instance 0: confirming 1.000000 instance for solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 3:46:22 PM | PrimeGrid | [cpu_sched_debug] scheduling genefer_1048576_421768_3 (high priority)
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] scheduling solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438062571463795_103079215104_0
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438990284399731_103079215104_0
1/22/2015 3:46:22 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438268729894003_103079215104_0
1/22/2015 3:46:22 PM | | [cpu_sched_debug] enforce_run_list: end

Thanks again for the assistance.
--
jon

Thanks, Claggy.

Point taken on BM, my mistake.

I'll refactor the file to follow the standard you referenced and try again.

--
jon

I recoded the app_config file per the standard but it appears the client continues to use the default cpu and gpu usage values. I am attempting to allow two solo_collatz WUs to run concurrently and reserve one (1) full cpu in support of the gpu WUs.

BTW, I also tried stopping the client (and WUs) and restarting to make sure the config files were being re-read, just in case.

Refactored app_config:

<app_config>
<app>
<name>solo_collatz</name>
<fraction_done_exact/>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.50</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
<app>
<name>mini_collatz</name>
<fraction_done_exact/>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.50</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
<app>
<name>large_collatz</name>
<fraction_done_exact/>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>1.00</gpu_usage>
<cpu_usage>1.00</cpu_usage>
</gpu_versions>
</app>
</app_config>

Log file snippet:

1/22/2015 4:23:17 PM | | Re-reading cc_config.xml
1/22/2015 4:23:17 PM | | Not using a proxy
1/22/2015 4:23:17 PM | | Config: report completed tasks immediately
1/22/2015 4:23:17 PM | | Config: use all coprocessors
1/22/2015 4:23:17 PM | | Config: event log limit disabled
1/22/2015 4:23:17 PM | | log flags: file_xfer, sched_ops, task, coproc_debug, cpu_sched, cpu_sched_debug
1/22/2015 4:23:17 PM | | log flags: statefile_debug, unparsed_xml, work_fetch_debug
1/22/2015 4:23:17 PM | Collatz Conjecture | Found app_config.xml
1/22/2015 4:23:17 PM | primaboinca | Found app_config.xml
1/22/2015 4:23:17 PM | Milkyway@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Moo! Wrapper | Found app_config.xml
1/22/2015 4:23:17 PM | Albert@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Einstein@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Asteroids@home | Found app_config.xml
1/22/2015 4:23:17 PM | PrimeGrid | Found app_config.xml
1/22/2015 4:23:17 PM | | [cpu_sched_debug] Request CPU reschedule: Core client configuration
1/22/2015 4:23:17 PM | | [work_fetch] Request work fetch: Core client configuration
1/22/2015 4:23:18 PM | | [cpu_sched_debug] schedule_cpus(): start
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: solo_collatz_2393014943811557401715_1649267441664_0 (coprocessor job, FIFO) (prio -0.821280)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] reserving 1.000000 of coproc NVIDIA
1/22/2015 4:23:18 PM | PrimeGrid | [cpu_sched_debug] add to run list: genefer_1048576_421768_3 (CPU job, EDF) (prio -1000.081312)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438268729894003_103079215104_0 (CPU job, priority order) (prio -0.851804)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438062571463795_103079215104_0 (CPU job, priority order) (prio -0.854814)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] add to run list: mini_collatz_2393020438990284399731_103079215104_0 (CPU job, priority order) (prio -0.857823)
1/22/2015 4:23:18 PM | | [cpu_sched_debug] enforce_run_list(): start
1/22/2015 4:23:18 PM | | [cpu_sched_debug] preliminary job list:
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 0: solo_collatz_2393014943811557401715_1649267441664_0 (MD: no; UTS: yes)
1/22/2015 4:23:18 PM | PrimeGrid | [cpu_sched_debug] 1: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 2: mini_collatz_2393020438268729894003_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 3: mini_collatz_2393020438062571463795_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 4: mini_collatz_2393020438990284399731_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | | [cpu_sched_debug] final job list:
1/22/2015 4:23:18 PM | PrimeGrid | [cpu_sched_debug] 0: genefer_1048576_421768_3 (MD: yes; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 1: solo_collatz_2393014943811557401715_1649267441664_0 (MD: no; UTS: yes)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 2: mini_collatz_2393020438268729894003_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 3: mini_collatz_2393020438062571463795_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] 4: mini_collatz_2393020438990284399731_103079215104_0 (MD: no; UTS: no)
1/22/2015 4:23:18 PM | Collatz Conjecture | [coproc] NVIDIA instance 0; 1.000000 pending for solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 4:23:18 PM | Collatz Conjecture | [coproc] NVIDIA instance 0: confirming 1.000000 instance for solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 4:23:18 PM | PrimeGrid | [cpu_sched_debug] scheduling genefer_1048576_421768_3 (high priority)
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] scheduling solo_collatz_2393014943811557401715_1649267441664_0
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438268729894003_103079215104_0
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438062571463795_103079215104_0
1/22/2015 4:23:18 PM | Collatz Conjecture | [cpu_sched_debug] scheduling mini_collatz_2393020438990284399731_103079215104_0
1/22/2015 4:23:18 PM | | [cpu_sched_debug] enforce_run_list: end

--
jon

I recoded the app_config file per the standard but it appears the client continues to use the default cpu and gpu usage values. I am attempting to allow two solo_collatz WUs to run concurrently and reserve one (1) full cpu in support of the gpu WUs.

BTW, I also tried stopping the client (and WUs) and restarting to make sure the config files were being re-read, just in case.

I think you need to double check you're editing the right app_config.xml, and in the right directory, it should be:

C:\ProgramData\BOINC\projects\boinc.thesonntags.com_collatz

I took your app_config.xml and made one in my collatz project directory, did a read config, re-enabled Use CPU and ATI/AMD & NV GPUs, made my host only use 4 cores,
and grabbed some work from collatz,

I'm running two Solo Collatz Conjenture 6.04 (1 CPUs + 0.5 AMD/ATI GPUs) tasks,
and one Large Collatz Conjenture (1 CPUs + 1 Nvidia GPU) task,
and one Mini Collatz Conjenture 6.04 CPU task, making a total of 4 tasks/cores utilised.

increasing the allowed cores from 4 to 6 only started one more Mini Collatz Conjenture 6.04 CPU task,
showing that as well as 'Solo Collatz' being limited to two tasks running at once, that Mini Collatz is too.

Claggy

Claggy,

Thanks for double checking my app_config settings on your system. I checked (and double checked) the location of my collatz app_config file and it is in the location that you listed. I do run collatz on a OS X machine (7.4.36 client) as well with no issues so I reported this under the possibility that there was an issue unique to the Windows version of the client.

Given your testing I will close-down this thread until I can find a more defensible root cause -- I'll report back here when I find something.

Thanks again for the troubleshoot assistance.

--
jon

I deleted and rebuilt the app_config.xml file from scratch and it is now appears to be properly processed by the client.

It appears that the file was corrupted in some way but not to the point that the client didn't recognize it as an app_config file:

1/22/2015 4:23:17 PM | Collatz Conjecture | Found app_config.xml
1/22/2015 4:23:17 PM | primaboinca | Found app_config.xml
1/22/2015 4:23:17 PM | Milkyway@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Moo! Wrapper | Found app_config.xml
1/22/2015 4:23:17 PM | Albert@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Einstein@Home | Found app_config.xml
1/22/2015 4:23:17 PM | Asteroids@home | Found app_config.xml
1/22/2015 4:23:17 PM | PrimeGrid | Found app_config.xml

It seems to have been treating it as an empty file. I have the unparsed_xml flag turned on and didn't see any log messages related to it. Assuming that it would produce error log messages for the app_config file as it does for an account manager such as:

1/23/2015 11:20:36 AM | | Contacting account manager at http://bam.boincstats.com/
1/23/2015 11:20:38 AM | | [unparsed_xml] ACCT_MGR_OP::parse: unrecognized tag <version>
1/23/2015 11:20:38 AM | | [unparsed_xml] ACCT_MGR_OP::parse: unrecognized tag <server>

Just wanted to provide a public mea culpa and reasonable closure to my original reported issue. Thanks again to everyone for your review, troubleshooting, and suggestions.

--
jon

7.4.38 would've had client: always show unparsed tags in config files. Problem is, 7.4.38 was either never built, or never released for testing, so we're still waiting for this code to come to a real live bouncing baby BOINC.

---