2010/6/19 Miles Fidelman <[email protected]>: > > - beyond the complexity and load, can you say a little more about why "RAID6 > on software is not a good idea, even worse with that hardware." >
First you have the load issue, just one processor (at most two threads with HT) is used for the real work and RAID maintenance (stripping and double parity calculation). Second, with RAID5/RAID6 you need the data to get to the disks (with RAID1 in the worse case, you just do a full sync), if your PSU dies under high load of the system, there's a good change to get a broken RAID. That's why dedicated hardware uses battery backed cache... > >> This deserves a deeper analysis with collected data, but if you are >> sure the drive is bad, replace it and check if it still happens (it >> will hit you back when adding the new disk again though) >> >> > > I'm just more confused at this point. Any other suggestions as to what to > look at, and how, would be very much appreciated. > Check for IOPS to disk when i/o wait triggers (IOPS and service time should increase), that way you can tell if it's an application or if it's a hardware issue, to know which applications is the one to blame, use collectl... Syslog should show something if it's a hardware issue (operation retry or just plain i/o error). > I'm also wondering if it would make sense to drop the disk from the array, > doing a complete diagnostic and reformat, then bringing it backup. I would just replace the disk, if that solves the issue after rebuild, the taken disk should go to a trashcan... > Thoughts? Suggestions as to procedure to minimize impact on the the array > and rebuild (yes I know RAID6 degrades below 4 disks - not quite sure of the > implications though for running with 3 drives for a few hours). Every read block that's supposed to live in the missing disk will be calculated from surviving disks. > > Thanks again, > > Miles Regards, -- Ciro Iriarte http://cyruspy.wordpress.com -- _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
