Problems after machine and BOINC upgrade

Forums › Problems and Bug Reports

Robert Rubin
Robert Rubin
Joined: 24 Apr 12
Posts: 10
Credit: 1181898
RAC: 0
21 Jan 2013 16:28:45 UTC
Topic 196757
(moderation:
    )

    I have started receiving numerous "Lost task"messages and "unable to connect to server" messages in my event log after switching from a Apple Powermac G5 (PowerPC) to a new Intel-based iMac. I also see plenty of tasks running but only one completed upload.

    Perhaps I need to reset something I missed in switching to the new machine? In preparation, while still running the old Powermac, I requested no new work and then finished all pending tasks. I then upgraded to the newest version of BOINC (7.0.31) before restarting on the new iMac.

    Are there other things that need to be changed for the new machine to work? I never received any error messages in my event log when I was running the old Powermac.

    Thanks for any help/advice.

    Gary Roberts
    Gary Roberts
    Moderator
    Joined: 9 Feb 05
    Posts: 5877
    Credit: 118636903936
    RAC: 18397220

    Problems after machine and BOINC upgrade

    21 Jan 2013 21:38:10 UTC
    Message 114458
    (moderation:
      )

      Quote:
      I have started receiving numerous "Lost task"messages and "unable to connect to server" messages in my event log after switching from a Apple Powermac G5 (PowerPC) to a new Intel-based iMac. I also see plenty of tasks running but only one completed upload.


      Your computers are hidden so we can't see what returned tasks look like.

      Is your new machine on a wired or wireless connection? If wireless, have you checked to see if you have a good strong signal? If the scheduler is "resending lost tasks" it would indicate that you are not properly receiving all the data when the scheduler sends you a reply to a work request.

      Quote:
      Perhaps I need to reset something I missed in switching to the new machine? In preparation, while still running the old Powermac, I requested no new work and then finished all pending tasks. I then upgraded to the newest version of BOINC (7.0.31) before restarting on the new iMac.


      Is the installation on the iMac a new install complete with a new hostID or did you transfer the previous BOINC data directory and retain the previous identity for the new machine? If it's a new install, everything should be fine.

      Cheers,
      Gary.

      Robert Rubin
      Robert Rubin
      Joined: 24 Apr 12
      Posts: 10
      Credit: 1181898
      RAC: 0

      RE: RE: I have started

      1 Feb 2013 0:28:31 UTC
      Message 114459 in response to message 114458
      (moderation:
        )

        Quote:
        Quote:
        I have started receiving numerous "Lost task"messages and "unable to connect to server" messages in my event log after switching from a Apple Powermac G5 (PowerPC) to a new Intel-based iMac. I also see plenty of tasks running but only one completed upload.

        Your computers are hidden so we can't see what returned tasks look like.

        i will be happy to unhide them if you can tell me how to do it.

        Is your new machine on a wired or wireless connection? If wireless, have you checked to see if you have a good strong signal? If the scheduler is "resending lost tasks" it would indicate that you are not properly receiving all the data when the scheduler sends you a reply to a work request.

        I do have a very good signal - 970 mbs download and similar on upload

        Quote:
        Perhaps I need to reset something I missed in switching to the new machine? In preparation, while still running the old Powermac, I requested no new work and then finished all pending tasks. I then upgraded to the newest version of BOINC (7.0.31) before restarting on the new iMac.

        Is the installation on the iMac a new install complete with a new hostID or did you transfer the previous BOINC data directory and retain the previous identity for the new machine? If it's a new install, everything should be fine.

        The new installation is a completely new install. New Boinc (latest release) on new hardware.

        Thanks for any help. I'm going to shut down until this can be resolved. I can see tasks getting processed, buy no uploads and lots of errors, that never happened before on the old Powermac.


        Jord
        Joined: 26 Jan 05
        Posts: 2952
        Credit: 5893653
        RAC: 1

        RE: i will be happy to

        1 Feb 2013 0:36:57 UTC
        Message 114460 in response to message 114459
        (moderation:
          )

          Quote:
          i will be happy to unhide them if you can tell me how to do it.


          Go to Einstein project preferences, edit them, set "Should Einstein@Home show your computers on its web site?" to Yes, save changes to the web site.

          It may take a bit of time before the computers are visible, it depends on whether or not these pages are cached.

          Robert Rubin
          Robert Rubin
          Joined: 24 Apr 12
          Posts: 10
          Credit: 1181898
          RAC: 0

          RE: RE: i will be happy

          3 Feb 2013 4:50:23 UTC
          Message 114461 in response to message 114460
          (moderation:
            )

            Quote:
            Quote:
            i will be happy to unhide them if you can tell me how to do it.

            Go to Einstein project preferences, edit them, set "Should Einstein@Home show your computers on its web site?" to Yes, save changes to the web site.

            It may take a bit of time before the computers are visible, it depends on whether or not these pages are cached.

            Many thanks. My computers will be visible once the changed preferences affect the public pages.

            Odd that my old (2005) Powermac G5 ran so trouble free, and I am now unable to run a fresh install of Boinc 7.0.31 on a new iMac!

            If anyone has any further suggestions on troubleshooting the problem (tasks run but never finish and are never uploaded), I'd appreciate it.

            Thanks.

            Gary Roberts
            Gary Roberts
            Moderator
            Joined: 9 Feb 05
            Posts: 5877
            Credit: 118636903936
            RAC: 18397220

            RE: If anyone has any

            4 Feb 2013 2:01:28 UTC
            Message 114462 in response to message 114461
            (moderation:
              )

              Quote:
              If anyone has any further suggestions on troubleshooting the problem (tasks run but never finish and are never uploaded), I'd appreciate it.


              Thanks for showing your computers.

              The odd thing is that in your tasks list, there is no evidence of any crunching being done. You say that crunching starts but never finishes but none of the old aborted tasks show even 1 second of crunch time. If you abort a task that has started, the crunch time used is normally reported. It really looks like crunching never starts for some reason. Are you sure BOINC is actually running and that the science apps are showing elapsed time?

              Do you use BOINC Manager (advanced view) to monitor how things are going? Can you actually see the progress percentage incrementing on the "Tasks" tab? Can you quote some values for "Elapsed time", "Progress" and "To completion" columns? You should also look at the event log and examine what BOINC is recording there.

              Another helpful thing you could do is stop and restart BOINC and copy the startup messages from the event log and post them into a new message here. You should check those messages yourself and see if you can spot a problem. If there is an obvious problem, just post the lines describing the problem if you don't understand it.

              Cheers,
              Gary.

              Robert Rubin
              Robert Rubin
              Joined: 24 Apr 12
              Posts: 10
              Credit: 1181898
              RAC: 0

              RE: RE: If anyone has any

              4 Feb 2013 3:45:17 UTC
              Message 114463 in response to message 114462
              (moderation:
                )

                Quote:
                Quote:
                If anyone has any further suggestions on troubleshooting the problem (tasks run but never finish and are never uploaded), I'd appreciate it.

                Thanks for showing your computers.

                The odd thing is that in your tasks list, there is no evidence of any crunching being done. You say that crunching starts but never finishes but none of the old aborted tasks show even 1 second of crunch time. If you abort a task that has started, the crunch time used is normally reported. It really looks like crunching never starts for some reason. Are you sure BOINC is actually running and that the science apps are showing elapsed time?

                Do you use BOINC Manager (advanced view) to monitor how things are going? Can you actually see the progress percentage incrementing on the "Tasks" tab? Can you quote some values for "Elapsed time", "Progress" and "To completion" columns? You should also look at the event log and examine what BOINC is recording there.

                Another helpful thing you could do is stop and restart BOINC and copy the startup messages from the event log and post them into a new message here. You should check those messages yourself and see if you can spot a problem. If there is an obvious problem, just post the lines describing the problem if you don't understand it.

                Odd indeed. And I appreciate your help very much in trying to figure this out.

                Here are the first few lines out of the running task list from Boinc Manager:

                Project Progress Status Elapsed .... Name

                Einstein@Home 93.131% Running, High Priority 05:12:37 h1_0444.85_S6GC1_415...
                Einstein@Home 6.44% Running 00:47:09 p2030.20121017.G49.05...

                Boinc Manager certainly seems to be reporting crunching, and the Activity Monitor on my iMac confirms a lot of activity.

                When the tasks hit 100%, they disappear from the list, but don't show up in any of the places where one can look for completed work.

                And here is the Event Log from Boinc Manager:

                Sun Feb 3 18:00:34 2013 | | No config file found - using defaults
                Sun Feb 3 18:00:34 2013 | | Starting BOINC client version 7.0.31 for x86_64-apple-darwin
                Sun Feb 3 18:00:34 2013 | | log flags: file_xfer, sched_ops, task
                Sun Feb 3 18:00:34 2013 | | Libraries: libcurl/7.26.0 OpenSSL/0.9.7l zlib/1.2.5 c-ares/1.9.1
                Sun Feb 3 18:00:34 2013 | | Data directory: /Library/Application Support/BOINC Data
                Sun Feb 3 18:00:34 2013 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz [x86 Family 6 Model 58 Stepping 9]
                Sun Feb 3 18:00:34 2013 | | Processor features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 PCLMULQDQ DTES64 MON DSCPL VMX SMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 SSE4.2 xAPIC POPCNT AES PCID XSAVE OSXSAVE TSCTMR AVX1.0 RDRAND F16C
                Sun Feb 3 18:00:34 2013 | | OS: Mac OS X 10.8.2 (Darwin 12.2.1)
                Sun Feb 3 18:00:34 2013 | | Memory: 32.00 GB physical, 2.78 TB virtual
                Sun Feb 3 18:00:34 2013 | | Disk: 2.83 TB total, 2.78 TB free
                Sun Feb 3 18:00:34 2013 | | Local time is UTC -8 hours
                Sun Feb 3 18:00:34 2013 | | OpenCL: NVIDIA GPU 0: GeForce GTX 680MX (driver version CLH 1.0, device version OpenCL 1.1, 2048MB, 2048MB available)
                Sun Feb 3 18:00:34 2013 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 6426110; resource share 100
                Sun Feb 3 18:00:34 2013 | Einstein@Home | General prefs: from Einstein@Home (last modified 27-Apr-2012 20:03:23)
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Computer location: home
                Sun Feb 3 18:00:34 2013 | Einstein@Home | General prefs: no separate prefs for home; using your defaults
                Sun Feb 3 18:00:34 2013 | | Reading preferences override file
                Sun Feb 3 18:00:34 2013 | | Preferences:
                Sun Feb 3 18:00:34 2013 | | max memory usage when active: 16384.00MB
                Sun Feb 3 18:00:34 2013 | | max memory usage when idle: 29491.20MB
                Sun Feb 3 18:00:34 2013 | | max disk usage: 50.00GB
                Sun Feb 3 18:00:34 2013 | | don't use GPU while active
                Sun Feb 3 18:00:34 2013 | | suspend work if non-BOINC CPU load exceeds 25 %
                Sun Feb 3 18:00:34 2013 | | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0426.25_S6GC1__790_S6LV1B_1 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0426.25_S6GC1__789_S6LV1B_0 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__672_S6LV1B_1 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__671_S6LV1B_0 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__670_S6LV1B_0 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0429.35_S6GC1__1075_S6LV1B_1 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0433.25_S6GC1__769_S6LV1B_1 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Task h1_0444.85_S6GC1__415_S6LV1B_1 is 3.85 days overdue; you may not get credit for it. Consider aborting it.
                Sun Feb 3 18:00:34 2013 | | Not using a proxy
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0426.25_S6GC1__790_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 1
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0426.25_S6GC1__789_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 2
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__672_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 3
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__671_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 4
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__670_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 5
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0429.35_S6GC1__1075_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 6
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0433.25_S6GC1__769_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 7
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Restarting task h1_0444.85_S6GC1__415_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 0
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Sending scheduler request: To fetch work.
                Sun Feb 3 18:00:34 2013 | Einstein@Home | Requesting new tasks for NVIDIA
                Sun Feb 3 18:00:40 2013 | Einstein@Home | Scheduler request completed: got 0 new tasks
                Sun Feb 3 18:00:40 2013 | Einstein@Home | No work sent
                Sun Feb 3 18:00:40 2013 | Einstein@Home | see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6426/6426110
                Sun Feb 3 18:12:42 2013 | Einstein@Home | Sending scheduler request: To fetch work.
                Sun Feb 3 18:12:42 2013 | Einstein@Home | Requesting new tasks for NVIDIA
                Sun Feb 3 18:12:44 2013 | Einstein@Home | Scheduler request completed: got 0 new tasks
                Sun Feb 3 18:12:44 2013 | Einstein@Home | No work sent
                Sun Feb 3 18:12:44 2013 | Einstein@Home | see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6426/6426110
                Sun Feb 3 18:42:20 2013 | Einstein@Home | Sending scheduler request: To fetch work.
                Sun Feb 3 18:42:20 2013 | Einstein@Home | Requesting new tasks for CPU and NVIDIA
                Sun Feb 3 18:42:30 2013 | Einstein@Home | Scheduler request completed: got 1 new tasks
                Sun Feb 3 18:42:32 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3120.bin4
                Sun Feb 3 18:42:32 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3121.bin4
                Sun Feb 3 18:42:39 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3120.bin4
                Sun Feb 3 18:42:39 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3122.bin4
                Sun Feb 3 18:42:43 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3121.bin4
                Sun Feb 3 18:42:43 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3123.bin4
                Sun Feb 3 18:42:44 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3122.bin4
                Sun Feb 3 18:42:44 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3124.bin4
                Sun Feb 3 18:42:46 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3123.bin4
                Sun Feb 3 18:42:46 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3124.bin4
                Sun Feb 3 18:42:46 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3125.bin4
                Sun Feb 3 18:42:46 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3126.bin4
                Sun Feb 3 18:42:50 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3125.bin4
                Sun Feb 3 18:42:50 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3126.bin4
                Sun Feb 3 18:42:50 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3127.bin4
                Sun Feb 3 18:42:50 2013 | Einstein@Home | Started download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000.zap
                Sun Feb 3 18:42:52 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000.zap
                Sun Feb 3 18:42:54 2013 | Einstein@Home | Finished download of p2030.20121009.G193.74-02.02.S.b4s0g0.00000_3127.bin4
                Sun Feb 3 18:43:30 2013 | Einstein@Home | Sending scheduler request: To fetch work.
                Sun Feb 3 18:43:30 2013 | Einstein@Home | Requesting new tasks for NVIDIA
                Sun Feb 3 18:43:33 2013 | Einstein@Home | Scheduler request completed: got 0 new tasks
                Sun Feb 3 18:43:33 2013 | Einstein@Home | No work sent
                Sun Feb 3 18:43:33 2013 | Einstein@Home | see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6426/6426110
                Sun Feb 3 18:53:40 2013 | Einstein@Home | Computation for task h1_0429.35_S6GC1__1075_S6LV1B_1 finished
                Sun Feb 3 18:53:40 2013 | Einstein@Home | Starting task p2030.20121017.G49.05-01.51.C.b6s0g0.00000_3040_1 using einsteinbinary_BRP4 version 133 (BRP4SSE) in slot 6
                Sun Feb 3 18:53:42 2013 | Einstein@Home | Started upload of h1_0429.35_S6GC1__1075_S6LV1B_1_0
                Sun Feb 3 18:53:48 2013 | Einstein@Home | Finished upload of h1_0429.35_S6GC1__1075_S6LV1B_1_0
                Sun Feb 3 18:53:50 2013 | Einstein@Home | Sending scheduler request: To report completed tasks.
                Sun Feb 3 18:53:50 2013 | Einstein@Home | Reporting 1 completed tasks
                Sun Feb 3 18:53:50 2013 | Einstein@Home | Not requesting tasks: don't need
                Sun Feb 3 18:53:53 2013 | Einstein@Home | Scheduler request completed
                Sun Feb 3 19:01:39 2013 | Einstein@Home | Computation for task h1_0426.25_S6GC1__789_S6LV1B_0 finished
                Sun Feb 3 19:01:39 2013 | Einstein@Home | Starting task h1_0447.00_S6GC1__S6BucketLVEa_447.113682292Hz_648_1 using einstein_S6BucketLVE version 104 in slot 2
                Sun Feb 3 19:01:41 2013 | Einstein@Home | Started upload of h1_0426.25_S6GC1__789_S6LV1B_0_0
                Sun Feb 3 19:01:44 2013 | Einstein@Home | Finished upload of h1_0426.25_S6GC1__789_S6LV1B_0_0
                Sun Feb 3 19:01:45 2013 | Einstein@Home | Sending scheduler request: To report completed tasks.
                Sun Feb 3 19:01:45 2013 | Einstein@Home | Reporting 1 completed tasks
                Sun Feb 3 19:01:45 2013 | Einstein@Home | Not requesting tasks: don't need
                Sun Feb 3 19:01:47 2013 | Einstein@Home | Scheduler request completed
                Sun Feb 3 19:01:53 2013 | Einstein@Home | Computation for task h1_0426.25_S6GC1__790_S6LV1B_1 finished
                Sun Feb 3 19:01:53 2013 | Einstein@Home | Starting task h1_0447.00_S6GC1__S6BucketLVEa_447.113682292Hz_649_1 using einstein_S6BucketLVE version 104 in slot 1
                Sun Feb 3 19:01:56 2013 | Einstein@Home | Started upload of h1_0426.25_S6GC1__790_S6LV1B_1_0
                Sun Feb 3 19:02:01 2013 | Einstein@Home | Finished upload of h1_0426.25_S6GC1__790_S6LV1B_1_0
                Sun Feb 3 19:02:48 2013 | Einstein@Home | Sending scheduler request: To report completed tasks.
                Sun Feb 3 19:02:48 2013 | Einstein@Home | Reporting 1 completed tasks
                Sun Feb 3 19:02:48 2013 | Einstein@Home | Not requesting tasks: don't need
                Sun Feb 3 19:02:50 2013 | Einstein@Home | Scheduler request completed

                Gary Roberts
                Gary Roberts
                Moderator
                Joined: 9 Feb 05
                Posts: 5877
                Credit: 118636903936
                RAC: 18397220

                RE: Boinc Manager certainly

                4 Feb 2013 7:52:33 UTC
                Message 114464 in response to message 114463
                (moderation:
                  )

                  Quote:

                  Boinc Manager certainly seems to be reporting crunching, and the Activity Monitor on my iMac confirms a lot of activity.

                  When the tasks hit 100%, they disappear from the list, but don't show up in any of the places where one can look for completed work.


                  I'm not sure what you mean by this last sentence?? When tasks finish, they get uploaded and then they get 'reported' at which point they get removed from your machine because they are now safely in the online database. If you look at your tasks list there (link provided previously) you will see them.

                  As a result of whatever you've done, there are now 7 completed tasks showing in the database. They were all a few days past deadline so most of them (5) were not awarded credit because someone else had been sent a copy of the task when it timed out and had returned it before you. However in two cases you beat the other person with your finished tasks and so they were awarded credit. You have one remaining 'old task' to go before you get fully into new work. Its quorum is already complete so you wont receive any credit so if you can identify it you should abort it. By the time you read this it will probably be crunched anyway.

                  By the look of things, there is nothing wrong with CPU crunching on your machine. It looks like BOINC was simply not running for most of the time previously. Having restarted BOINC, you have to be careful not to shut down BOINC itself when you close BOINC Manager. I suspect that's what has been happening. Do you have the BOINC Manager icon sitting in the dock when you close the Boinc Manager window? Right click the icon and select 'Options'. Have you got 'Open at Login' and 'Keep in dock' both selected?

                  Your startup messages look quite normal. I've selected a couple of things to comment on.

                  Sun Feb 3 18:00:34 2013 | | OpenCL: NVIDIA GPU 0: GeForce GTX 680MX (driver version CLH 1.0, device version OpenCL 1.1, 2048MB, 2048MB available) BOINC seems to be recognising your GPU but is not giving any information about CUDA compute capability so I'm guessing you need some sort of CUDA driver. I had a quick look on the nvidia website and found this page where there is a link about CUDA for MAC OS X. I know nothing about this but this is where I would start reading if I wanted to find out more :-). Hopefully, there might be someone reading this who has CUDA tasks running under MAC OS X who can tell you what to do if you want to use your GPU.

                  Sun Feb  3 18:00:34 2013 |  | suspend work if non-BOINC CPU load exceeds 25 %
Sun Feb  3 18:00:34 2013 |  | (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)


                  I would change this preference to zero rather than 25%. CPU tasks run at low priority and seem to get interrupted (suspended) at regular intervals if you leave this at 25%. Just a personal opinion. Maybe it's better in more recent versions of BOINC (it was quite a while ago when it annoyed me :-).)

                  Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0426.25_S6GC1__790_S6LV1B_1 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0426.25_S6GC1__789_S6LV1B_0 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__672_S6LV1B_1 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__671_S6LV1B_0 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0427.75_S6GC1__670_S6LV1B_0 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0429.35_S6GC1__1075_S6LV1B_1 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0433.25_S6GC1__769_S6LV1B_1 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Task h1_0444.85_S6GC1__415_S6LV1B_1 is 3.85 days overdue; you may not get credit for it.  Consider aborting it.


                  You are being advised of 8 tasks that are past the deadline.

                  Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0426.25_S6GC1__790_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 1
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0426.25_S6GC1__789_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 2
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__672_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 3
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__671_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 4
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0427.75_S6GC1__670_S6LV1B_0 using einstein_S6LV1 version 113 (SSE2) in slot 5
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0429.35_S6GC1__1075_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 6
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0433.25_S6GC1__769_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 7
Sun Feb  3 18:00:34 2013 | Einstein@Home | Restarting task h1_0444.85_S6GC1__415_S6LV1B_1 using einstein_S6LV1 version 113 (SSE2) in slot 0


                  These same 8 tasks are now being restarted. Note the use of 'restarting' rather than 'starting'. This means that all these tasks had at some stage previously been started and were now restarting from a saved checkpoint. This is fairly strong evidence that, essentially, BOINC had not been running very much (if at all) during the previous 17+ days. The maximum BOINC could have been running is only a few hours, otherwise they would have been completed and returned.

                  Sun Feb  3 18:00:34 2013 | Einstein@Home | Sending scheduler request: To fetch work.
Sun Feb  3 18:00:34 2013 | Einstein@Home | Requesting new tasks for NVIDIA
Sun Feb  3 18:00:40 2013 | Einstein@Home | Scheduler request completed: got 0 new tasks
Sun Feb  3 18:00:40 2013 | Einstein@Home | No work sent
Sun Feb  3 18:00:40 2013 | Einstein@Home | see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6426/6426110


                  BOINC is looking for work for your GPU but is not getting any. You will be able to work out the reason if (as suggested) you look at the scheduler log messages. The easiest way to do that is to go to your account page on the website and click on the 'Computers' link there. On the page that comes up in the far right hand column you will see a 'last contact' link. Click it and you will get something like this

                  2013-02-04 03:24:20.3288 [PID=9891]   Request: [USER#xxxxx] [HOST#6426110] [IP xxx.xxx.xxx.32] client 7.0.31
2013-02-04 03:24:20.3344 [PID=9891 ]    [send] effective_ncpus 8 max_jobs_on_host_cpu 999999 max_jobs_on_host 999999
2013-02-04 03:24:20.3344 [PID=9891 ]    [send] effective_ngpus 1 max_jobs_on_host_gpu 999999
2013-02-04 03:24:20.3344 [PID=9891 ]    [send] Not using matchmaker scheduling; Not using EDF sim
2013-02-04 03:24:20.3344 [PID=9891 ]    [send] CPU: req 0.00 sec, 0.00 instances; est delay 0.00
2013-02-04 03:24:20.3345 [PID=9891 ]    [send] CUDA: req 86580.00 sec, 1.00 instances; est delay 0.00
2013-02-04 03:24:20.3345 [PID=9891 ]    [send] work_req_seconds: 0.00 secs
2013-02-04 03:24:20.3345 [PID=9891 ]    [send] available disk 48.85 GB, work_buf_min 0
2013-02-04 03:24:20.3345 [PID=9891 ]    [send] active_frac 0.981976 on_frac 0.104438 DCF 1.943656
2013-02-04 03:24:20.3364 [PID=9891 ]    [send] [HOST#6426110] is reliable
2013-02-04 03:24:20.3365 [PID=9891 ]    [send] set_trust: error rate 0.085737 > 0.050000, don't trust
2013-02-04 03:24:20.3485 [PID=9891 ]    [version] Checking plan class 'BRP4cuda32OSX'
2013-02-04 03:24:20.3489 [PID=9891 ]    [version] reading plan classes from file '/BOINC/projects/EinsteinAtHome/plan_class_spec.xml'
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] CUDA compute capability required min: 100, supplied: 0
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] Checking plan class 'BRP4SSE'
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] plan class ok
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] Don't need CPU jobs, skipping version 133 for einsteinbinary_BRP4 (BRP4SSE)
2013-02-04 03:24:20.3490 [PID=9891 ]    [version] Checking plan class 'opencl-ati-lion'
2013-02-04 03:24:20.3491 [PID=9891 ]    [version] parsed project prefs setting 'gpu_util_brp': 1.000000
2013-02-04 03:24:20.3491 [PID=9891 ]    [version] No ATI devices found
2013-02-04 03:24:20.3491 [PID=9891 ]    [version] no app version available: APP#19 (einsteinbinary_BRP4) PLATFORM#6 (i686-apple-darwin) min_version 0
2013-02-04 03:24:20.3491 [PID=9891 ]    [version] Don't need CPU jobs, skipping version 1 for hsgamma_FGRP2 ()
2013-02-04 03:24:20.3491 [PID=9891 ]    [version] no app version available: APP#21 (hsgamma_FGRP2) PLATFORM#6 (i686-apple-darwin) min_version 0
2013-02-04 03:24:20.3536 [PID=9891 ]    [send] [HOST#6426110] is looking for work from a non-preferred application
2013-02-04 03:24:21.3250 [PID=9891 ]    [version] Don't need CPU jobs, skipping version 104 for einstein_S6BucketLVE ()
2013-02-04 03:24:21.3250 [PID=9891 ]    [version] no app version available: APP#22 (einstein_S6BucketLVE) PLATFORM#6 (i686-apple-darwin) min_version 0
2013-02-04 03:24:21.3268 [PID=9891 ] [debug]   [HOST#6426110] MSG(high) No work sent
2013-02-04 03:24:21.3268 [PID=9891 ] [debug]   [HOST#6426110] MSG(high) see scheduler log messages on http://einstein.phys.uwm.edu//host_sched_logs/6426/6426110
2013-02-04 03:24:21.3268 [PID=9891 ]    Sending reply to [HOST#6426110]: 0 results, delay req 60.00
2013-02-04 03:24:21.3271 [PID=9891 ]    Scheduler ran 1.005 seconds


                  Note the line that says "CUDA compute capability required min: 100, supplied: 0". This is why the scheduler wont send you any work for your GPU. It needs to see a sensible CUDA compute capability. Later on (back in your event log that you posted), your client asked for work for both the CPU and the GPU. No GPU work was supplied but you did get a BRP4 task for the CPU. This task will take a lot longer on a CPU core than it would if sent to a properly recognised GPU. If you get your GPU working you may want to stop these tasks going to the CPU because of how long they take there. You can do that with preference settings.

                  Cheers,
                  Gary.

                  Robert Rubin
                  Robert Rubin
                  Joined: 24 Apr 12
                  Posts: 10
                  Credit: 1181898
                  RAC: 0

                  Thank you for taking the time

                  9 Feb 2013 3:10:03 UTC
                  Message 114465 in response to message 114464
                  (moderation:
                    )

                    Thank you for taking the time to analyze my issues, Gary.

                    What I meant by the tasks disappearing after reaching 100% completion, is that when I selected "Your Results" in Boinc Manager, nothing showed up. Whereas completed tasks did show up there after crunching was done and the report was uploaded on the old configuration. This was not happening for about a week after I moved to the new machine and the new Boinc Manager release. I did shut everything down after about three days of this, which is why there has not been a lot of run time.

                    I will follow up on your suggestions.

                    I very much appreciate your help.

                    Best regards,

                    Rob

                    Richard Haselgrove
                    Richard Haselgrove
                    Joined: 10 Dec 05
                    Posts: 2143
                    Credit: 2990696345
                    RAC: 703858

                    RE: What I meant by the

                    9 Feb 2013 10:16:01 UTC
                    Message 114466 in response to message 114465
                    (moderation:
                      )

                      Quote:
                      What I meant by the tasks disappearing after reaching 100% completion, is that when I selected "Your Results" in Boinc Manager, nothing showed up. Whereas completed tasks did show up there after crunching was done and the report was uploaded on the old configuration. This was not happening for about a week after I moved to the new machine and the new Boinc Manager release. I did shut everything down after about three days of this, which is why there has not been a lot of run time.


                      The newer versions of BOINC tend to use the word 'Tasks', where older ones said 'Results'.

                      If you go to the 'tasks' tab in BOINC Manager (Advanced view), there's a group of command buttons to the top left of your screen. The top one toggles between 'Show active tasks' and 'Show all tasks'. If it currently says 'Show all tasks', try clicking it: that might reveal the missing results.

                      Robert Rubin
                      Robert Rubin
                      Joined: 24 Apr 12
                      Posts: 10
                      Credit: 1181898
                      RAC: 0

                      RE: RE: What I meant by

                      12 Feb 2013 15:22:56 UTC
                      Message 114467 in response to message 114466
                      (moderation:
                        )

                        Quote:
                        Quote:
                        What I meant by the tasks disappearing after reaching 100% completion, is that when I selected "Your Results" in Boinc Manager, nothing showed up. Whereas completed tasks did show up there after crunching was done and the report was uploaded on the old configuration. This was not happening for about a week after I moved to the new machine and the new Boinc Manager release. I did shut everything down after about three days of this, which is why there has not been a lot of run time.

                        The newer versions of BOINC tend to use the word 'Tasks', where older ones said 'Results'.

                        If you go to the 'tasks' tab in BOINC Manager (Advanced view), there's a group of command buttons to the top left of your screen. The top one toggles between 'Show active tasks' and 'Show all tasks'. If it currently says 'Show all tasks', try clicking it: that might reveal the missing results.

                        You were correct! Thank you very much for the suggestion. Mystery solved and BOINC Manager now seems to be working as expected - and with a very substantial performance improvement on the new iMac over the old PowerPC Powermac I used to run.

                        Comment viewing options

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

                        Forums › Problems and Bug Reports