On Tue, Jan 19, 2016 at 09:55:26AM +0100, Hannes Reinecke wrote: > RAID10 delays the write until the bitmap has been updated. > So it really should check if the device is still working > before sending requests, otherwise it'll happily sending > I/O to a known faulty device. Hi, __make_request already checks the faulty bit. That's possible the disk becomes faulty after __make_request check though. But faulty bit can be set any time, for example, after the check with your patch. There is no guarantee pending writes get canceled immediately after the disk is marked faulty. I'm wondering what's the real problem. Thanks, Shaohua -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
