The underlying problem is a real bad disk. I was pleasantly surprised that zfs could still cope very well and even tell me which files so far are corrupt. There is one particular file which is causing me trouble. If I try to do anything, e.g., type the beginning of the filename and hit tab, then I get
[ 357.7550409] panic: solaris assert: remove_reference(hdr, ((void *)0), tag) =7 [ 357.7850073] cpu11: Begin traceback... [ 357.7950078] vpanic() at netbsd:vpanic+0x156 [ 357.8050083] panic() at netbsd:panic+0x3c [ 357.8150080] arc_buf_destroy() at zfs:arc_buf_destroy+0x19d [ 357.8250082] arc_read_done() at zfs:arc_read_done+0x1ab [ 357.8350084] zio_done() at zfs:zio_done+0x38b [ 357.8450097] zio_execute() at zfs:zio_execute+0xe6 [ 357.8450097] task_executor() at solaris:task_executor+0x67 [ 357.8650091] threadpool_thread() at netbsd:threadpool_thread+0x1a1 [ 357.8750095] cpu11: End traceback... The trouble is when I try to rsync -P as much as possible away, that file will be touched and I get the panic. Any ideas, or has my luck run out? Cheers, Patrick
