Anyone else notice the high number of failure rates for all short runs? Was looking and have noticed I had a few resends to my machines that completed correctly. Was trying to figure out why the previous machines errored out. Couldn't find a common reason other than BOINC version. I hope there is a way for the scientist to look at these and see what might have been the reason. Since we can't access those error logs, only they can tell us if it's the work unit that errored or something with the computers.

Z
____________

Not sure if this is the same issue, but also ran into this:
http://www.gpugrid.net/result.php?resultid=20751166
Runs, for about 2 hrs, gets into the high 30% complete range, then crashes. Restarts, gets to about the same point, crashes again. Finally just aborted this, as it looks as though it would have looped forever.
Looking through the logs, can't see anything that would indicate I have an issue here.
Jim
____________

Couldn't find a common reason other than BOINC version.

Here's one. All users had an error in a couple of seconds.
https://www.gpugrid.net/workunit.php?wuid=16425136