On 2002-10-06 (21:48, -0400), Daniel Finn DFinn@studentadvantage.com wrote:
More often than not the disk scrubbing on the netapps that I have don't finish. I assume this is because most of them see pretty high I/O and the netapp puts that I/O in higher order than the disk scrub. The disk scrub can be though of as a parity check correct?
Disk scrub is meant to find data/physical errors on disks and will be started regulary - here on sunday nights. You can run the disk scrub manualy if you want to and think that your filer isn't to busy for it - e.g. during the night via 'disk scrub start'.
Warnings and errors will be mailed via autosupport and written to /etc/messages. Take a look into the logfiles to get further informations about reasons why it stops. However, you can set the priority of the disk scrub via 'options raid.scrub.*' - read the documentation for more informations about it.
raid.scrub.duration 0 raid.scrub.enable on raid.scrub.perf_impact low (value might be overwritten in takeover)
Greetings,