On Fri, 6 May 2011, Johnny Billquist wrote:
On 2011-05-06 20.32, Paul Goyette wrote:Conclusion: these disk drives suck! It's not a raidframe problem....What interest me is if this really is a drive problem, or some NetBSD problem. I have observed problems in the past that might be related, with rather bad I/O performance. Haven't had time to dig deeper into it, or properly even diagnose it, so it's mostly suspicions.Could you try booting some other OS and try the performance of the disks?
As several people pointed out, the poor drive performance was a direct result of using the block device vs raw device. With the raw device I have both drives simultaneously dd'ing at nearly 100M _each_. Erasure is nearly finished now, and I'll be trying to reassemble the raidset tomorrow morning.
------------------------------------------------------------------------- | Paul Goyette | PGP Key fingerprint: | E-mail addresses: | | Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com | | Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net | | Kernel Developer | | pgoyette at netbsd.org | -------------------------------------------------------------------------