pc is being drip fed with one task/day

Forums › Problems and Bug Reports

ashj
ashj
Joined: 21 Jun 06
Posts: 55
Credit: 8922931
RAC: 0
21 Jul 2009 12:33:31 UTC
Topic 194450
(moderation:
    )

    I have three boxes crunching for E@H. One of which, for the last week, can only download a single task.

    Unfortunately, the normal duration of a task is around 15 hours so I can only crunch one task a day and the box is then idle for up to 9 hours.

    Has anyone a similar experience?

    My other two boxes have no problems at all and yesterday my dual core obtained twelve new tasks so it is clearly not a shortage of work.

    Any comments or advice gratefully received.

    Gundolf Jahn
    Gundolf Jahn
    Joined: 1 Mar 05
    Posts: 1079
    Credit: 341280
    RAC: 0

    pc is being drip fed with one task/day

    21 Jul 2009 13:04:53 UTC
    Message 93844
    (moderation:
      )

      You could check the Long Term Debts for that machine. If there seems to be something amiss, try the clause in cc_config.xml. (Don't forget to remove it again, afterwards.)

      Or else the section from client_state.xml could give hints.

      Gruß,
      Gundolf

      Computer sind nicht alles im Leben. (Kleiner Scherz)

      Richard Haselgrove
      Richard Haselgrove
      Joined: 10 Dec 05
      Posts: 2142
      Credit: 2776688246
      RAC: 797106

      Won't be anything to do with

      21 Jul 2009 13:30:06 UTC
      Message 93845
      (moderation:
        )

        Won't be anything to do with debt, Gundolf, if the box is idle.

        At first I thought it sounded more like a continual error which had brought quota down to 1, but looking at host 1943707, that doesn't seem to be the case.

        Ashj, if you have a careful look at that list - and watch out, they aren't in chronological order - you'll see that you're actually being sent a new task at the same time at the previous one is reported - you should only be losing a few seconds while they're uploaded/downloaded.

        But even so, it shouldn't be happening, and it makes sense to try and sort it out. Have a look at the bottom of the computer details (you can see much more on that page than we can). What are the four figures at the bottom, and how do they compare with -

        Quote:
        % of time BOINC client is running 99.9841 %
        While BOINC running, % of time host has an Internet connection 99.9943 %
        While BOINC running, % of time work is allowed 99.9904 %
        Task duration correction factor 1.374623


        Post your figures here, and we can advise further.

        ashj
        ashj
        Joined: 21 Jun 06
        Posts: 55
        Credit: 8922931
        RAC: 0

        Hi Richard, These are my

        21 Jul 2009 13:46:05 UTC
        Message 93846 in response to message 93845
        (moderation:
          )

          Hi Richard,

          These are my figures:

          % of time BOINC client is running 98.9988 %
          While BOINC running, % of time host has an Internet connection 3.7519 %
          While BOINC running, % of time work is allowed 99.9071 %
          Task duration correction factor 1.721775

          I do not leave the box online until I need a connection.

          Unfortunatly I do need some shut eye so the nine hours lost is always past midnight (GMT +1).

          With Regards

          ashj

          Michael Karlinsky
          Michael Karlinsky
          Joined: 22 Jan 05
          Posts: 888
          Credit: 23502182
          RAC: 0

          Hi, taken from scheduler

          21 Jul 2009 13:54:26 UTC
          Message 93847
          (moderation:
            )

            Hi,

            taken from scheduler logs:

            2009-07-21 06:23:48.6127 [PID=22411]   Request:  [HOST#1943707]  client 6.6.36
2009-07-21 06:23:48.6135 [PID=22411] [debug]   [HOST#1943707] Resetting nresults_today
2009-07-21 06:23:48.7113 [PID=22411]    [send] CPU: req 30240.00 sec, 1.00 instances; est delay 0.00
2009-07-21 06:23:48.7114 [PID=22411]    [send] work_req_seconds: 30240.00 secs
2009-07-21 06:23:48.7114 [PID=22411]    [send] Not using matchmaker scheduling; Not using EDF sim
2009-07-21 06:23:48.7114 [PID=22411]    [send] available disk 9.87 GB, work_buf_min 8640
2009-07-21 06:23:48.7114 [PID=22411]    [send] active_frac 0.999071 on_frac 0.989988 DCF 1.721775
2009-07-21 06:23:48.7121 [PID=22411]    [send] [HOST#1943707] is reliable; OS: Microsoft Windows XP, error_rate: 0.004849, avg_turn_hrs: 17.837 max res/day 16
2009-07-21 06:23:48.7121 [PID=22411]    [send] set_trust: random choice for error rate 0.004849: no
2009-07-21 06:23:48.7121 [PID=22411]    [mixed] sending non-locality work first
2009-07-21 06:23:48.7135 [PID=22411] [debug]   Sorted list of URLs follows [host timezone: UTC+3600]
2009-07-21 06:23:48.7135 [PID=22411] [debug]   zone=+03600 url=http://einstein.aei.mpg.de
2009-07-21 06:23:48.7136 [PID=22411] [debug]   zone=+00000 url=http://einstein.astro.gla.ac.uk
2009-07-21 06:23:48.7136 [PID=22411] [debug]   zone=-28800 url=http://einstein.ligo.caltech.edu
2009-07-21 06:23:48.7138 [PID=22411]    [send] [HOST#1943707] Sending app_version einsteinbinary_ABP1 2 305 ; 1.60 GFLOPS
2009-07-21 06:23:48.7287 [PID=22411]    [send] est. duration for WU 56131340: unscaled 34429.05 scaled 59934.26
2009-07-21 06:23:48.7288 [PID=22411]    [HOST#1943707] Sending [RESULT#134390302 p2030_53614_05631_0048_G65.21+00.11.N_6.dm_118_1] (est. dur. 59934.26 seconds)
2009-07-21 06:23:48.7307 [PID=22411]    [send] don't need more work
2009-07-21 06:23:48.7307 [PID=22411]    [mixed] sending locality work second
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.05_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.05_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.10_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.10_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.15_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.15_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.20_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.20_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.25_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.25_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.30_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.30_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.35_S5R4
2009-07-21 06:23:48.7307 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.35_S5R4
2009-07-21 06:23:48.7308 [PID=22411]    [locality] [HOST#1943707] has file h1_0873.40_S5R4
2009-07-21 06:23:48.7308 [PID=22411]    [locality] [HOST#1943707] has file l1_0873.40_S5R4
2009-07-21 06:23:48.7308 [PID=22411]    [locality] [HOST#1943707] has file skygrid_0880Hz_S5R5.dat
2009-07-21 06:23:48.7308 [PID=22411]    [send] don't need more work
2009-07-21 06:23:48.7308 [PID=22411]    [send] don't need more work
2009-07-21 06:23:48.7308 [PID=22411]    [send] don't need more work
2009-07-21 06:23:48.7448 [PID=22411]    Sending reply to [HOST#1943707]: 1 results, delay req 60.00
2009-07-21 06:23:48.7450 [PID=22411]    Scheduler ran 0.661 seconds

            HTH

            Michael

            Team Linux Users Everywhere

            Richard Haselgrove
            Richard Haselgrove
            Joined: 10 Dec 05
            Posts: 2142
            Credit: 2776688246
            RAC: 797106

            The time fractions are OK

            21 Jul 2009 14:05:18 UTC
            Message 93848 in response to message 93847
            (moderation:
              )

              The time fractions are OK (except for internet connection, and you've explained that).

              It'll be

              Quote:
              2009-07-21 06:23:48.7113 [PID=22411] [send] CPU: req 30240.00 sec, 1.00 instances; est delay 0.00


              (0.35 days) - probably 0.1 days connect interval, 0.25 days additional cache.

              If you're only seeing this for one out of three hosts, maybe you accidentally set those as local preferences? Check 'Preferences' both on your account pages and in BOINC Manager.

              ashj
              ashj
              Joined: 21 Jun 06
              Posts: 55
              Credit: 8922931
              RAC: 0

              I have reset preferences

              21 Jul 2009 15:26:53 UTC
              Message 93849 in response to message 93848
              (moderation:
                )

                I have reset preferences to:

                Internet connection to every one day and enough work for an additional 3 days.

                I updated on line with E@H and then requested more work but do far without any joy.

                Will try again at the next submission.

                Thanks for the input from all contributors.

                Gundolf Jahn
                Gundolf Jahn
                Joined: 1 Mar 05
                Posts: 1079
                Credit: 341280
                RAC: 0

                RE: I updated on line with

                21 Jul 2009 17:37:17 UTC
                Message 93850 in response to message 93849
                (moderation:
                  )

                  Quote:
                  I updated on line with E@H and then requested more work but do far without any joy.


                  You didn't say anything about local preferences. Did you try the "Clear" button?

                  What about the different venues (home, school, work, default)?

                  Computer sind nicht alles im Leben. (Kleiner Scherz)

                  Michael Karlinsky
                  Michael Karlinsky
                  Joined: 22 Jan 05
                  Posts: 888
                  Credit: 23502182
                  RAC: 0

                  Hi, are you getting "won't

                  22 Jul 2009 10:45:13 UTC
                  Message 93851
                  (moderation:
                    )

                    Hi,

                    are you getting "won't finish in time" messages? There seems to be a problem with BOINC 6.6.36, consider upgrading, see SETI thread too.

                    HTH

                    Michael

                    Team Linux Users Everywhere

                    Richard Haselgrove
                    Richard Haselgrove
                    Joined: 10 Dec 05
                    Posts: 2142
                    Credit: 2776688246
                    RAC: 797106

                    RE: Hi, are you getting

                    22 Jul 2009 11:12:40 UTC
                    Message 93852 in response to message 93851
                    (moderation:
                      )

                      Quote:

                      Hi,

                      are you getting "won't finish in time" messages? There seems to be a problem with BOINC 6.6.36, consider upgrading, see SETI thread too.

                      HTH

                      Michael


                      That would need v6.6.38, only compiled yesterday evening: I wouldn't upgrade to that until some of the people moaning on the BOINC message boards about 'High Priority' running get their debugger logs out and post some metrics.

                      I think that your server log about work request, plus the fact that he's been getting a new task at each manual connection, is enough of an answer for this thread.

                      Jord
                      Joined: 26 Jan 05
                      Posts: 2952
                      Credit: 5779100
                      RAC: 0

                      RE: [I wouldn't upgrade to

                      22 Jul 2009 12:19:10 UTC
                      Message 93853 in response to message 93852
                      (moderation:
                        )

                        Quote:
                        [I wouldn't upgrade to that until some of the people moaning on the BOINC message boards


                        That's just a side effect of the fix, since the bug wrecked the long running estimate of deadlines (one that's stored on the server, no less!). I noticed it goes away after either running halfway through a task, or running at least one task.

                        But moaning is correct, people expect a bug to be fixed and their BOINC to magically reset all numbers to pre-bug time.

                        Comment viewing options

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

                        Forums › Problems and Bug Reports