On 1/25/2012 12:15 PM, Seth Jennings wrote: > The write performance on my raid5 took a nose dive yesterday Simply identify what changed between the 24th and 25th, if indeed this problem started on the 25th. Odds are you're currently attacking this from the wrong angle, as neither your mdraid setup, your HBA, nor the drives themselves have changed in a 24 hour period, correct? More likely is that in this time frame you wrote a large file(s) filling up the EXT4 filesystem sufficiently to force writing into smaller and smaller free spaces, which is likely causing excessive RAID RMW operations. Show df output for all filesystems on the mdraid device to confirm or reject this theory. One of the downsides of parity RAID is that RMW kicks in with a vengeance when filesystems start getting full. -- Stan -- 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
