Not getting work

Forums › Problems and Bug Reports

John
John
Joined: 25 Feb 14
Posts: 2
Credit: 115403152
RAC: 32972
28 May 2018 14:39:15 UTC
Topic 215198
(moderation:
    )

    I haven't been getting any work since around the 16th of May. Had some trouble with nvidia driver around then so probably had a bunch of work units return errors in a row, but everything is working now. But can't get any work.

    Gavin
    Gavin
    Joined: 21 Sep 10
    Posts: 191
    Credit: 40644337738
    RAC: 2

    According to your host's last

    28 May 2018 14:51:49 UTC
    Message 165508
    (moderation:
      )

      According to your host's last scheduler contact your host doesn't support openCL. I suspect you have had a Windows update recently that has forced its version of GPU driver on you.

      You need to install or re-install a driver direct form nVidea to get crunching again :-)

      John
      John
      Joined: 25 Feb 14
      Posts: 2
      Credit: 115403152
      RAC: 32972

      That was it.  Thanks a bunch!

      28 May 2018 22:05:43 UTC
      Message 165509 in response to message 165508
      (moderation:
        )

        That was it.  Thanks a bunch!

        SuperSluether
        SuperSluether
        Joined: 1 Sep 14
        Posts: 4
        Credit: 54809605
        RAC: 0

        I also just had this issue on

        9 Jun 2018 5:52:02 UTC
        Message 165650
        (moderation:
          )

          I also just had this issue on Ubuntu 18.04. Apparently, the Nvidia driver that is available from the "Additional Drivers" section in the "Software and Updates" window only installs CUDA.

           

          To get OpenCL, I had to install the "boinc-client-opencl" package. (Technically this just installs the "ocl-icd-libopencl1" package, but it's a nice shortcut)

          Uzzy Booboo
          Uzzy Booboo
          Joined: 24 Mar 09
          Posts: 1
          Credit: 3088075
          RAC: 2038

          I'm having the same problem. 

          20 Jul 2018 3:00:36 UTC
          Message 166201
          (moderation:
            )

            I'm having the same problem.  5 of 6 projects are not downloading new work.  I've reinstalled BOINC and the drivers.  I've removed and re-added the projects.  Nothing is working.

            Gary Roberts
            Gary Roberts
            Moderator
            Joined: 9 Feb 05
            Posts: 5872
            Credit: 117494213735
            RAC: 35449090

            You can't be having the same

            20 Jul 2018 10:14:28 UTC
            Message 166207
            (moderation:
              )

              You can't be having the same problem (graphics driver not having the needed OpenCL compute libraries installed) because neither of the two hosts you have attached to this project have an nVidia or an AMD graphics card.  Only one of those hosts has made a recent contact with the project.

              Perhaps you might like to tell us what particular search or searches you have selected in your project preferences? When was the last time you got work for any search?  Have you made any changes to your project preferences recently?

              You mention you have 6 projects that you support but only one is getting work.  Can you mention all your projects and tell us which is the one that gets work.  Could it be that that particular project is behind in its share of your resources and BOINC is favouring it in order to help it catch up?

              Have you looked at the scheduler logs to see what happened the last time your active host contacted the servers?  In the 'Getting Started' forum there is this pinned thread that explains how to do that.  Here is a snip from the last contact for the active one of your hosts.  I've trimmed time, date, process ID (ie. this bit -> 2018-07-20 02:56:15.5835 [PID=18127]) from the beginning of each line so there's no line wrapping in the middle of messages.  The lines labelled [debug] can be ignored (info for the Devs).

              Request: [USER#xxxxx] [HOST#12654379] [IP xxx.xxx.xxx.69] client 7.12.1
[debug] have_master:1 have_working: 1 have_db: 1
[debug] using working prefs
[debug] have db 1; dbmod 1504032588.000000; global mod 1504032588.000000
[send] effective_ncpus 4 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
[send] effective_ngpus 1 max_jobs_on_host_gpu 999999
[send] Not using matchmaker scheduling; Not using EDF sim
[send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
[send] Intel GPU: req 0.00 sec, 0.00 instances; est delay 0.00
[send] work_req_seconds: 0.00 secs
[send] available disk 99.25 GB, work_buf_min 86400
[send] active_frac 0.860548 on_frac 0.772257 DCF 1.000000
Sending reply to [HOST#12654379]: 0 results, delay req 60.00
Scheduler ran 0.027 seconds

              If you look at the parameter 'work_req_seconds' you can see that your BOINC client wasn't requesting any work.  Either you have preference settings that don't allow work to be requested or perhaps your client thinks that some other project is more worthy at this time.

              I notice that your BOINC client is version 7.12.1 which (I think) is very recently released.  Your problem could always be some sort of new client bug.  Did the problem coincide with the install of 7.12.1?

               

              Cheers,
              Gary.

              richard
              richard
              Joined: 29 Oct 12
              Posts: 2
              Credit: 207799489
              RAC: 0

              i have two pcs at home both

              18 Aug 2018 4:59:41 UTC
              Message 166504
              (moderation:
                )

                i have two pcs at home both running boinc but only one will use the gpu. Ive tried uninstalling and reinstalling boinc, updating gpu drivers, but i have a few projects that could use gpu but on one rig they wont. boinc is detecting my gpu in both rigs but it will not accept work for the gpu in the gaming rig. please help. here is the boinc event log

                 

                8/17/2018 8:57:51 PM | | cc_config.xml not found - using defaults
                8/17/2018 8:57:51 PM | | Starting BOINC client version 7.12.1 for windows_x86_64
                8/17/2018 8:57:51 PM | | log flags: file_xfer, sched_ops, task
                8/17/2018 8:57:51 PM | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
                8/17/2018 8:57:51 PM | | Data directory: C:\ProgramData\BOINC
                8/17/2018 8:57:51 PM | | Running under account rdhii
                8/17/2018 8:57:51 PM | | CUDA: NVIDIA GPU 0: GeForce GTX 1080 Ti (driver version 398.82, CUDA version 9.2, compute capability 6.1, 4096MB, 3550MB available, 12064 GFLOPS peak)
                8/17/2018 8:57:51 PM | | Host name: DeepThought
                8/17/2018 8:57:51 PM | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz [Family 6 Model 158 Stepping 9]
                8/17/2018 8:57:51 PM | | 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 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 pbe fsgsbase bmi1 hle smep bmi2
                8/17/2018 8:57:51 PM | | OS: Microsoft Windows 10: Core x64 Edition, (10.00.17134.00)
                8/17/2018 8:57:51 PM | | Memory: 31.96 GB physical, 36.71 GB virtual
                8/17/2018 8:57:51 PM | | Disk: 465.16 GB total, 189.92 GB free
                8/17/2018 8:57:51 PM | | Local time is UTC -8 hours
                8/17/2018 8:57:51 PM | | No WSL found.
                8/17/2018 8:57:51 PM | | VirtualBox version: 5.2.8
                8/17/2018 8:57:51 PM | Cosmology@Home | URL http://www.cosmologyathome.org/; Computer ID 357163; resource share 100
                8/17/2018 8:57:51 PM | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12652136; resource share 100
                8/17/2018 8:57:51 PM | LHC@home | URL https://lhcathome.cern.ch/lhcathome/; Computer ID 10548509; resource share 100
                8/17/2018 8:57:51 PM | Milkyway@Home | URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 775830; resource share 100
                8/17/2018 8:57:51 PM | Rosetta@home | URL http://boinc.bakerlab.org/rosetta/; Computer ID 3411950; resource share 100
                8/17/2018 8:57:51 PM | Universe@Home | URL https://universeathome.pl/universe/; Computer ID 495429; resource share 100
                8/17/2018 8:57:51 PM | World Community Grid | URL http://www.worldcommunitygrid.org/; Computer ID 4391083; resource share 100
                8/17/2018 8:57:51 PM | Einstein@Home | General prefs: from Einstein@Home (last modified 17-Aug-2018 20:35:17)
                8/17/2018 8:57:51 PM | Einstein@Home | Host location: none
                8/17/2018 8:57:51 PM | Einstein@Home | General prefs: using your defaults
                8/17/2018 8:57:51 PM | | Reading preferences override file
                8/17/2018 8:57:51 PM | | Preferences:
                8/17/2018 8:57:51 PM | | max memory usage when active: 16363.36 MB
                8/17/2018 8:57:51 PM | | max memory usage when idle: 29454.05 MB
                8/17/2018 8:57:51 PM | | max disk usage: 23.26 GB
                8/17/2018 8:57:51 PM | | suspend work if non-BOINC CPU load exceeds 25%
                8/17/2018 8:57:51 PM | | (to change preferences, visit a project web site or select Preferences in the Manager)
                8/17/2018 8:57:51 PM | | Setting up project and slot directories
                8/17/2018 8:57:51 PM | | Checking active tasks
                8/17/2018 8:57:51 PM | | Using account manager GridRepublic
                8/17/2018 8:57:51 PM | | Setting up GUI RPC socket
                8/17/2018 8:57:51 PM | | Checking presence of 328 project files
                8/17/2018 8:57:51 PM | Milkyway@Home | Starting task de_nbody_7_23_2018_v170_20k__data_3_1533467104_121103_1
                8/17/2018 8:57:51 PM | Milkyway@Home | Sending scheduler request: To fetch work.
                8/17/2018 8:57:51 PM | Milkyway@Home | Requesting new tasks for NVIDIA GPU
                8/17/2018 8:57:52 PM | Milkyway@Home | Scheduler request completed: got 0 new tasks
                8/17/2018 8:57:58 PM | Einstein@Home | Sending scheduler request: To fetch work.
                8/17/2018 8:57:58 PM | Einstein@Home | Requesting new tasks for NVIDIA GPU
                8/17/2018 8:58:00 PM | Einstein@Home | Scheduler request completed: got 0 new tasks
                8/17/2018 8:58:00 PM | Einstein@Home | No work sent
                8/17/2018 8:58:00 PM | Einstein@Home | see scheduler log messages on https://einsteinathome.org/host/12652136/log

                 

                and here is the einstein scheduler log

                 

                2018-08-18 04:58:01.5904 [PID=27958]   Request: [USER#xxxxx] [HOST#12652136] [IP xxx.xxx.xxx.2] client 7.12.1
2018-08-18 04:58:01.6522 [PID=27958] [debug]   have_master:1 have_working: 1 have_db: 1
2018-08-18 04:58:01.6523 [PID=27958] [debug]   using working prefs
2018-08-18 04:58:01.6523 [PID=27958] [debug]   have db 1; dbmod 1534566917.000000; global mod 1534566917.000000
2018-08-18 04:58:01.6523 [PID=27958]    [send] effective_ncpus 8 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2018-08-18 04:58:01.6523 [PID=27958]    [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2018-08-18 04:58:01.6523 [PID=27958]    [send] Not using matchmaker scheduling; Not using EDF sim
2018-08-18 04:58:01.6523 [PID=27958]    [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2018-08-18 04:58:01.6523 [PID=27958]    [send] CUDA: req 51840.00 sec, 1.00 instances; est delay 0.00
2018-08-18 04:58:01.6523 [PID=27958]    [send] work_req_seconds: 0.00 secs
2018-08-18 04:58:01.6523 [PID=27958]    [send] available disk 18.56 GB, work_buf_min 8640
2018-08-18 04:58:01.6523 [PID=27958]    [send] active_frac 0.975688 on_frac 0.818989 DCF 1.269770
2018-08-18 04:58:01.6533 [PID=27958]    [mixed] sending non-locality work first (0.2846)
2018-08-18 04:58:01.6721 [PID=27958]    [version] Checking plan class 'opencl-intel_gpu'
2018-08-18 04:58:01.6751 [PID=27958]    [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2018-08-18 04:58:01.6751 [PID=27958]    [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2018-08-18 04:58:01.6751 [PID=27958]    [version] No Intel GPU devices found
2018-08-18 04:58:01.6751 [PID=27958]    [version] Checking plan class 'opencl-intel_gpu-Beta'
2018-08-18 04:58:01.6751 [PID=27958]    [version] beta test app versions not allowed in project prefs.
2018-08-18 04:58:01.6751 [PID=27958]    [version] Checking plan class 'opencl-intel_gpu-new'
2018-08-18 04:58:01.6751 [PID=27958]    [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2018-08-18 04:58:01.6751 [PID=27958]    [version] No Intel GPU devices found
2018-08-18 04:58:01.6752 [PID=27958]    [version] Checking plan class 'opencl-intel_gpu'
2018-08-18 04:58:01.6752 [PID=27958]    [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2018-08-18 04:58:01.6752 [PID=27958]    [version] No Intel GPU devices found
2018-08-18 04:58:01.6752 [PID=27958]    [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#9 (windows_x86_64) min_version 0
2018-08-18 04:58:01.6752 [PID=27958]    [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#2 (windows_intelx86) min_version 0
2018-08-18 04:58:01.6752 [PID=27958]    [version] Checking plan class 'FGRPopencl-ati'
2018-08-18 04:58:01.6752 [PID=27958]    [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2018-08-18 04:58:01.6752 [PID=27958]    [version] No ATI devices found
2018-08-18 04:58:01.6752 [PID=27958]    [version] Checking plan class 'FGRPopencl-nvidia'
2018-08-18 04:58:01.6752 [PID=27958]    [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2018-08-18 04:58:01.6753 [PID=27958]    [version] NVidia device (or driver) doesn't support OpenCL
2018-08-18 04:58:01.6753 [PID=27958]    [version] Checking plan class 'FGRPopencl1K-ati'
2018-08-18 04:58:01.6753 [PID=27958]    [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2018-08-18 04:58:01.6753 [PID=27958]    [version] No ATI devices found
2018-08-18 04:58:01.6753 [PID=27958]    [version] Checking plan class 'FGRPopencl1K-nvidia'
2018-08-18 04:58:01.6753 [PID=27958]    [version] parsed project prefs setting 'gpu_util_fgrp': 1.000000
2018-08-18 04:58:01.6753 [PID=27958]    [version] NVidia device (or driver) doesn't support OpenCL
2018-08-18 04:58:01.6753 [PID=27958]    [version] no app version available: APP#40 (hsgamma_FGRPB1G) PLATFORM#9 (windows_x86_64) min_version 0
2018-08-18 04:58:01.6753 [PID=27958]    [version] no app version available: APP#40 (hsgamma_FGRPB1G) PLATFORM#2 (windows_intelx86) min_version 0
2018-08-18 04:58:01.6754 [PID=27958]    [version] Checking plan class 'FGRPSSE'
2018-08-18 04:58:01.6755 [PID=27958]    [version] plan class ok
2018-08-18 04:58:01.6755 [PID=27958]    [version] Don't need CPU jobs, skipping version 108 for hsgamma_FGRP5 (FGRPSSE)
2018-08-18 04:58:01.6755 [PID=27958]    [version] no app version available: APP#46 (hsgamma_FGRP5) PLATFORM#9 (windows_x86_64) min_version 0
2018-08-18 04:58:01.6755 [PID=27958]    [version] no app version available: APP#46 (hsgamma_FGRP5) PLATFORM#2 (windows_intelx86) min_version 0
2018-08-18 04:58:01.6809 [PID=27958]    [mixed] sending locality work second
2018-08-18 04:58:01.6830 [PID=27958] [debug]   [HOST#12652136] MSG(high) No work sent
2018-08-18 04:58:01.6830 [PID=27958] [debug]   [HOST#12652136] MSG(high) see scheduler log messages on https://einsteinathome.org/host/12652136/log
2018-08-18 04:58:01.6831 [PID=27958]    Sending reply to [HOST#12652136]: 0 results, delay req 60.00
2018-08-18 04:58:01.6831 [PID=27958]    Scheduler ran 0.096 seconds
                Gary Roberts
                Gary Roberts
                Moderator
                Joined: 9 Feb 05
                Posts: 5872
                Credit: 117494213735
                RAC: 35449090

                Here is the relevant bit from

                18 Aug 2018 8:43:39 UTC
                Message 166506
                (moderation:
                  )

                  Here is the relevant bit from the above scheduler log excerpt.

                  NVidia device (or driver) doesn't support OpenCL

                  Windows apparently has a habit of updating graphics drivers and leaving out the compute libraries.  Gavin answered this same situation earlier in this thread.  This problem comes up quite regularly.  I don't use Windows at all so I'm just repeating what others have said.  The solution is to go to the nVidia website and download the full package.  You then have to completely remove the Microsoft driver and do a clean install of the full package from nVidia.

                   

                  Cheers,
                  Gary.

                  richard
                  richard
                  Joined: 29 Oct 12
                  Posts: 2
                  Credit: 207799489
                  RAC: 0

                  I get my drivers from nvidia

                  19 Aug 2018 19:08:25 UTC
                  Message 166522
                  (moderation:
                    )

                    I get my drivers from nvidia and have tried this several times and tried again this weekend but still no success. I get the same error NVidia device (or driver) doesn't support OpenCL no matter what i do.

                    Gary Roberts
                    Gary Roberts
                    Moderator
                    Joined: 9 Feb 05
                    Posts: 5872
                    Credit: 117494213735
                    RAC: 35449090

                    Apparently, there is a bit of

                    19 Aug 2018 22:20:39 UTC
                    Message 166523
                    (moderation:
                      )

                      Apparently, there is a bit of a knack as to how you install the drivers.  As I said, I don't use Windows so I have no direct experience.  Your problem will most likely be incorrect installation of the OpenCL libraries that are part of the full driver package.  There have been lots of previous examples of this problem over the last couple of years.

                      Have you tried a search here for some of these?  I just did a search for "nvidia AND driver AND clean AND install" so as to only find messages that contain all 4 words.  There were lots of hits.  As I recall, there was quite a lot of emphasis on doing a 'clean install', even to the point of running special apps to clean out any old driver remnants that might be interfering with the new install.  I also recall comments about selecting particular options in the install process itself.  I'm sure you'll find reference to these things if you look through some of the search results.

                       

                      Cheers,
                      Gary.

                      Comment viewing options

                      Select your preferred way to display the comments and click "Save settings" to activate your changes.

                      Forums › Problems and Bug Reports