Message boards : GPUs : Not getting any work for my Anonymous platform at Einstein@home
Message board moderation
Author | Message |
---|---|
Send message Joined: 17 Nov 16 Posts: 893 |
Hope someone can help. I am trying to get work for the BRP4 application at Einstein@home Not getting anything other than: Nano 131 Einstein@Home 4/27/2019 1:57:56 PM Sending scheduler request: To fetch work. 132 Einstein@Home 4/27/2019 1:57:56 PM Requesting new tasks for NVIDIA GPU 133 Einstein@Home 4/27/2019 1:58:00 PM Scheduler request completed: got 0 new tasks 134 Einstein@Home 4/27/2019 1:58:00 PM No work sent 135 Einstein@Home 4/27/2019 1:58:00 PM Your app_info.xml file doesn't have a version of Gamma-ray pulsar binary search #1 on GPUs. 136 Einstein@Home 4/27/2019 1:58:00 PM No work available for the applications you have selected. Please check your preferences on the web site. I don't have the Gamma-ray pulsar binary search #1 on GPUs project selected. I've been told that this is a flaw with Einstein and the message is sent to all and not to worry about. The problem is the no work is available for the application I've selected. The BRP4 application is selected at the website, saved for the Home location and the client has been updated multiple times as well as completely restarted. This is my app_info.xml <app_info> <app> <name>einsteinbinary_BRP4</name> </app> <file_info> <name>einsteinbinary_cuda64</name> <executable/> </file_info> <file_info> <name>einsteinbinary_cuda-db.dev</name> </file_info> <file_info> <name>einsteinbinary_cuda-dbhs.dev</name> </file_info> <app_version> <app_name>einsteinbinary_BRP4</app_name> <version_num>999</version_num> <api_version>7.2.2</api_version> <coproc> <type>CUDA</type> <count>1.0</count> </coproc> <file_ref> <file_name>einsteinbinary_cuda64</file_name> <main_program/> </file_ref> <file_ref> <file_name>einsteinbinary_cuda-db.dev</file_name> <open_name>db.dev</open_name> <copy_file/> </file_ref> <file_ref> <file_name>einsteinbinary_cuda-dbhs.dev</file_name> <open_name>dbhs.dev</open_name> <copy_file/> </file_ref> </app_version> </app_info> This is my app_version section in client_state.xml for the application <app_version> <app_name>einsteinbinary_BRP4</app_name> <version_num>999</version_num> <platform>aarch64-unknown-linux-gnu</platform> <avg_ncpus>1.000000</avg_ncpus> <flops>14732440740.064838</flops> <api_version>7.2.2</api_version> <file_ref> <file_name>einsteinbinary_cuda64</file_name> <main_program/> </file_ref> <file_ref> <file_name>einsteinbinary_cuda-db.dev</file_name> <open_name>db.dev</open_name> <copy_file/> </file_ref> <file_ref> <file_name>einsteinbinary_cuda-dbhs.dev</file_name> <open_name>dbhs.dev</open_name> <copy_file/> </file_ref> <coproc> <type>NVIDIA</type> <count>1.000000</count> </coproc> <dont_throttle/> </app_version> There are currently 10K BRP4 tasks available in the RTS buffer for that application. GAURAV KHANNA, the developer of the app I am using is getting work for his aarch64-unknown-linux-gnu Tegra hosts and is processing the BRP4 tasks fine. This my Event Log startup: 1 4/27/2019 11:54:58 AM Starting BOINC client version 7.9.3 for aarch64-unknown-linux-gnu 2 4/27/2019 11:54:58 AM log flags: file_xfer, sched_ops, task 3 4/27/2019 11:54:58 AM Libraries: libcurl/7.58.0 OpenSSL/1.1.0g zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3 4 4/27/2019 11:54:58 AM Data directory: /var/lib/boinc-client 5 4/27/2019 11:54:59 AM CUDA: NVIDIA GPU 0: NVIDIA Tegra X1 (driver version unknown, CUDA version 10.0, compute capability 5.3, 3957MB, 3315MB available, 236 GFLOPS peak) 6 Einstein@Home 4/27/2019 11:54:59 AM Found app_info.xml; using anonymous platform 7 4/27/2019 11:55:00 AM [libc detection] gathered: 2.27, Ubuntu GLIBC 2.27-3ubuntu1 8 4/27/2019 11:55:00 AM Host name: Nano 9 4/27/2019 11:55:00 AM Processor: 4 ARM ARMv8 Processor rev 1 (v8l) [Impl 0x41 Arch 8 Variant 0x1 Part 0xd07 Rev 1] 10 4/27/2019 11:55:00 AM Processor features: fp asimd evtstrm aes pmull sha1 sha2 crc32 11 4/27/2019 11:55:00 AM OS: Linux Ubuntu: Ubuntu 18.04.2 LTS [4.9.140-tegra|libc 2.27 (Ubuntu GLIBC 2.27-3ubuntu1)] 12 4/27/2019 11:55:00 AM Memory: 3.86 GB physical, 0 bytes virtual 13 4/27/2019 11:55:00 AM Disk: 29.21 GB total, 17.33 GB free 14 4/27/2019 11:55:00 AM Local time is UTC -7 hours 15 4/27/2019 11:55:00 AM Config: GUI RPC allowed from any host 16 4/27/2019 11:55:00 AM Config: GUI RPCs allowed from: 17 4/27/2019 11:55:00 AM 192.168.2.34 18 4/27/2019 11:55:00 AM Config: report completed tasks immediately 19 Einstein@Home 4/27/2019 11:55:00 AM URL http://einstein.phys.uwm.edu/; Computer ID 12775352; resource share 25 20 SETI@home 4/27/2019 11:55:00 AM URL http://setiathome.berkeley.edu/; Computer ID 8707387; resource share 75 21 SETI@home 4/27/2019 11:55:00 AM General prefs: from SETI@home (last modified 24-Apr-2019 09:04:51) 22 SETI@home 4/27/2019 11:55:00 AM Computer location: home 23 4/27/2019 11:55:00 AM General prefs: using separate prefs for home 24 4/27/2019 11:55:00 AM Reading preferences override file 25 4/27/2019 11:55:00 AM Preferences: 26 4/27/2019 11:55:00 AM max memory usage when active: 1978.28 MB 27 4/27/2019 11:55:00 AM max memory usage when idle: 3560.91 MB 28 4/27/2019 11:55:00 AM max disk usage: 5.00 GB 29 4/27/2019 11:55:00 AM max CPUs used: 3 30 4/27/2019 11:55:00 AM don't use GPU while active 31 4/27/2019 11:55:00 AM suspend work if non-BOINC CPU load exceeds 25% 32 4/27/2019 11:55:00 AM (to change preferences, visit a project web site or select Preferences in the Manager) 33 4/27/2019 11:55:00 AM Setting up project and slot directories 34 4/27/2019 11:55:00 AM Checking active tasks 35 4/27/2019 11:55:00 AM Setting up GUI RPC socket 36 4/27/2019 11:55:00 AM gui_rpc_auth.cfg is empty - no GUI RPC password protection 37 4/27/2019 11:55:00 AM Checking presence of 49 project files 38 4/27/2019 11:55:00 AM Suspending computation - user request 39 Einstein@Home 4/27/2019 11:56:00 AM project resumed by user 40 SETI@home 4/27/2019 11:56:03 AM project resumed by user 41 Einstein@Home 4/27/2019 11:56:16 AM update requested by user 42 Einstein@Home 4/27/2019 11:56:21 AM Sending scheduler request: Requested by user. 43 Einstein@Home 4/27/2019 11:56:21 AM Requesting new tasks for NVIDIA GPU 44 Einstein@Home 4/27/2019 11:56:24 AM Scheduler request completed: got 0 new tasks 45 Einstein@Home 4/27/2019 11:56:24 AM No work sent 46 Einstein@Home 4/27/2019 11:56:24 AM Your app_info.xml file doesn't have a version of Gamma-ray pulsar binary search #1 on GPUs. 47 Einstein@Home 4/27/2019 11:56:24 AM No work available for the applications you have selected. Please check your preferences on the web site. Can anyone point out why I can't get work? Thanks in advance. [Edit] This is the scheduler response for last contact: 2019-04-27 21:35:21.5051 [PID=9189] Request: [USER#xxxxx] [HOST#12775352] [IP xxx.xxx.xxx.154] client 7.9.3 2019-04-27 21:35:21.5296 [PID=9189 ] [debug] have_master:1 have_working: 1 have_db: 1 2019-04-27 21:35:21.5296 [PID=9189 ] [debug] using working prefs 2019-04-27 21:35:21.5296 [PID=9189 ] [debug] have db 1; dbmod 1556121891.000000; global mod 1556121891.000000 2019-04-27 21:35:21.5296 [PID=9189 ] [send] effective_ncpus 3 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999 2019-04-27 21:35:21.5296 [PID=9189 ] [send] effective_ngpus 1 max_jobs_on_host_gpu 999999 2019-04-27 21:35:21.5296 [PID=9189 ] [send] Not using matchmaker scheduling; Not using EDF sim 2019-04-27 21:35:21.5296 [PID=9189 ] [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00 2019-04-27 21:35:21.5296 [PID=9189 ] [send] CUDA: req 51840.00 sec, 1.00 instances; est delay 0.00 2019-04-27 21:35:21.5296 [PID=9189 ] [send] work_req_seconds: 51840.00 secs 2019-04-27 21:35:21.5296 [PID=9189 ] [send] available disk 4.98 GB, work_buf_min 43200 2019-04-27 21:35:21.5296 [PID=9189 ] [send] active_frac 0.997416 on_frac 0.983078 DCF 1.000000 2019-04-27 21:35:21.5296 [PID=9189 ] Anonymous platform app versions: 2019-04-27 21:35:21.5297 [PID=9189 ] app: einsteinbinary_BRP4 ver: 999 2019-04-27 21:35:21.5305 [PID=9189 ] [mixed] sending locality work first (0.4812) 2019-04-27 21:35:21.5310 [PID=9189 ] [mixed] sending non-locality work second 2019-04-27 21:35:21.5525 [PID=9189 ] [send] Didn't find anonymous platform app for hsgamma_FGRPB1G 2019-04-27 21:35:21.5599 [PID=9189 ] [debug] [HOST#12775352] MSG(high) No work sent 2019-04-27 21:35:21.5600 [PID=9189 ] [debug] [HOST#12775352] MSG(high) Your app_info.xml file doesn't have a version of Gamma-ray pulsar binary search #1 on GPUs. 2019-04-27 21:35:21.5600 [PID=9189 ] [debug] [HOST#12775352] MSG(high) No work available for the applications you have selected. Please check your preferences on the web site. 2019-04-27 21:35:21.5600 [PID=9189 ] Sending reply to [HOST#12775352]: 0 results, delay req 60.00 2019-04-27 21:35:21.5601 [PID=9189 ] Scheduler ran 0.059 seconds |
Send message Joined: 17 Nov 16 Posts: 893 |
Well I finally managed to get work. Turns out BOINC only allows one BETA application per project. I had the BETA cpu app running already for the RPi3+ for Home venue. Once I moved the Nano to School venue and turned off cpu work, I was finally able to get some gpu work for the beta application. Still debugging the Nano platform. Errored out all of my daily allotment of tasks when libraries weren't found. Having to wait out the 24 hour penalty box before trying again. |
Send message Joined: 27 Jun 08 Posts: 641 |
Still debugging the Nano platform. Errored out all of my daily allotment of tasks when libraries weren't found. Having to wait out the 24 hour penalty box before trying again. I have this happen all too often and usually worst is 600+ errored tasks return to milkyway. ====wish list==== Was wondering if there was a boinc client option such as "Allow new tasks for 10 seconds" however, that might be a problem on projects such as milkway so.. "Allow 1 new task" It seems to me that the 10 second time could easily be implemented but I do not know enough about the client to know if it can even ask for a single task or even if the project would honor a request for just 1. |
Send message Joined: 23 Feb 08 Posts: 2498 |
"Allow 1 new task" Perhaps better still "send test task" so people could verify their build. Once it gets the task it would have to set no new tasks so it wouldn't go into a loop. |
Send message Joined: 5 Oct 06 Posts: 5134 |
Try doing a command line boinc --helpto see what the start-up options are (I can't find them documented in the Wiki just now) '--fetch_minimal_work' sounds like what you're looking for. |
Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.