http://www.worldcommunitygrid.org/forums/wcg/viewthread?thread=10733
Another bug to fix. Cancel the work unit if it happens again.

Lawrence

Did you try to clic on "[Show Thread Printable Version]" that is at the bottom of each thread and then print the page from your navigator ('file' and after 'print') ?

An other way is to select the part of the thread you want to print and then clic on 'file' of your navigator and then 'print' and choose 'selected'.

You can also select what you want to print and then 'cut' and 'past' it to a word document.

It's strange but the 1.0.1.1 version of HCMD for UD Agent should have resolved the > 100% issue. It just sitting at 100% without completing is maybe the same and does reek like a similar thing to HPF2, where the process was not releasing a certain file(s). We thus need to know which of the UD processes is using CPU time:

1. Is there still a WCG_xxxx.exeprocess in taskmanager and is it using CPU time?
2. Is there a UD_xxxxxx.exe process and is it using CPU time?
3. Is there a UD.exe process (Yes of course), and is it using CPU time?

For closure kill the WCG_xxxx.exe process. It will send the rsults back and fetch new work. Killing UD.exe is likely only causing a restart and let it complete (if similar to the HPF2 'racing' event).

Edit:Corrected version number

For UD agent there should during the calculation process be 3 processes visible as outlined in my earlier post. The WCGridHCMD.exe is the one taking 50% (100 % of 1 core on a dual core machine). When the computation is finished, the WCGrid.HCMD.exe (aka the science) is closed and the UD_xxxxxxxx.exe performing the wrap up work and UD.exe showing the result of the device rating and the sending result and retrieval of new work.

Anyway, if the next 3 completed successfully, it just confirms the intermittence. Please report again if it shows its ugly face again.