File Up/Download problems, hitch, and questions

Forums › Cafe Einstein

Redstorm [LIGURIA]
Redstorm [LIGURIA]
Joined: 8 Nov 04
Posts: 2
Credit: 33183122
RAC: 0
14 Nov 2004 11:03:30 UTC
Topic 187086
(moderation:
    )

    Hi all, out there.

    It appears that one of my PC got three times the same workunit from the server,and sent back the only result that has been calculated.
    I know that this problem di not happen only once, to me.
    While I know that after the stated time has elapsed, the server will do send out the same workunit to more crunchin' bloodhounds, I think this apparent misbehaviour of server side software is to be reported to sysadmins.
    Please note that up to now I am able to run only on windoze machines, my bunch of FreeBSD machines is still sleeping in carton boxes after a massive ( wife, child, pets and books...;) transfer a short time ago form Naples to Genoa.

    Now an example of three is company, too workunit:

    http://einsteinathome.org/workunit/54020

    Bye for now.

    Pietro Lauro

    genes
    genes
    Joined: 10 Nov 04
    Posts: 41
    Credit: 2810710
    RAC: 9535

    File Up/Download problems, hitch, and questions

    14 Nov 2004 13:09:56 UTC
    Message 257
    (moderation:
      )

      I have one too:

      http://einsteinathome.org/workunit/58205

      The interesting thing here is that with two out of the three runs finished, they are not claiming the same credit. Perhaps the differences in run time could be related to how many times the WU had to be stopped to share with another BOINC project... ? I know the machine is reliable, but I do actually *use* it, in addition to running BOINC projects. ;-) Maybe that has an influence as well. BTW, all of my machines run every BOINC project available, which at this point includes CPDN, Seti, and E@H. LHC seems to have gone down for a while, but perhaps Predictor may start up again soon.

      I would think that it is dangerous from a project point of view to allow only one system to process a particular result, considering that some people who participate in these projects overclock their systems to the point of making them unreliable. But wait -- maybe they are testing our machines to see if they *DO* produce reliable, repeatable results. After all, this is a test phase of the project, time to work out all of these issues.

      -Gene


      genes
      genes
      Joined: 10 Nov 04
      Posts: 41
      Credit: 2810710
      RAC: 9535

      Alas, I am guilty (again) of

      14 Nov 2004 21:55:03 UTC
      Message 258
      (moderation:
        )

        Alas, I am guilty (again) of not looking in the Questions and Problems area. There are 2 threads,

        http://einsteinathome.org/node/187047

        and

        http://einsteinathome.org/node/187043

        dealing with this issue.


        Comment viewing options

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

        Forums › Cafe Einstein