LHC@home

So my client seem to start crunching CMS after sixtrack ran out of WUs. I got the VM setup and it works fine if the taks runs to completion while the pc is on. But even though I follow the recommendation to exit boinc and also wait will the vm finish offloading the info by monitoring the ram usage before shutdown, It seem like the task go to show computation error after turning the pc back on. 6 of my 8 task error almost immediately while the other 2 took another 20 or so min to error out in the last group, is there a way to troubleshoot this? [/img]

CMS- and ATLAS-tasks don't like shutdowns at all. When a VM is properly saved to disk by BOINC, you could be lucky when a shutdown/reboot takes place within about 20 minutes.
They want a continuous internet connection.
The only VM application that can survive longer suspends, stops, shutdowns etc. is Theory.

Ahh well, I'll tried get theory WUs or crunch somethingelse while sixtracks are out then as I cant keep the system up all night. Thanks for the clarifications.