On Sat, Jul 30, 2011 at 8:03 AM, Peter Humphrey <pe...@humphrey.ukfsn.org> wrote: > On Saturday 30 July 2011 11:17:52 Dale wrote: > >> What could have caused this? Could it be a file system problem? I >> don't think it is a physical failure since it is working now after >> giving it a fresh start. I just don't get how this could have caused a >> kernel panic. This is plain weird. > > One possibility is that, having now written to almost every location on the > disk, its controller has marked some faulty blocks that used to contain code > in the disk subsystem. If it was reading damaged data, there's no surprise > in anything that happened next!
Well, except that it should have thrown some errors when the on-platter reed-solomon encoding didn't quite match. If the controller flagged some faulty blocks, it should show up via smartctl -A $DEVICE_NODE -- :wq
