


I have also played with the number of cores to use, no change.

WU are set to "leave in mem" since yesterday, unfortunately, i still get the same error. I have already played around with some computing preferences. You have 3.5 gig of memory and are probably trying to run 8 units all at once, right? Are you running other projects too? If so stop those for a bit and see if the Rosie units stop erroring out. Now I think you SHOULD have the setting set to keep swapped units in memory. I´m getting this error (0xc0000005) too on my new machine Boinc tries to put the whole workunit in memory when it runs, with only 512meg of memory for the whole machine and running Windows too, you have very little left over for workunits, then you are trying to save them in memory when they get swapped out too, I think that is your problem. I would turn that off until you add more memory, and maybe even run only one unit at a time. I think this could be memory related, it says the machine only has 512mb of memory yet you are also "leaving in memory" all paused units. we are just bringing a boinc expert on board who should be able to help with this kind of error this seems pretty independent from the actual rosetta calculations.

Good to know-this kind of machine specificity should help us track down the problems. Wikipedia Join Distributed Computing projects that benefit humanity I assume these errors don't affect the science part, right? Could they be due to the PC shutting down (not being up 24/7)? Why am I getting a 'Reason: Access Violation (0xc0000005) error'? # This process generated 58 decoys from 58 attemptsįAQ says to ignore it Q. # DONE :: 1 starting structures built 58 (nstruct) times Reason: Access Violation (0xc0000005) at address 0x005100FF read attempt to address 0x9F1E623C Reason: Access Violation (0xc0000005) at address 0x0077A2BC write attempt to address 0x1BD34008 I wonder if it might be related to computer shutdown (BOINC runs as user process, not as service). I've never encountered this error on my other PCs, but the others are all running 24/7. This PC only runs with "leave in mem"=yes. I have recently added a WinXPsp2/P4/512MB PC ( hostid ) to which seems to encounter this error for most WUs it crunches, e.g.
