On Thu, Mar 26, 2009 at 03:12:52PM -0500, Billy Crook wrote: ... > Would a SATA drive behind vblade, or behind the LayerWalker 16800 chip > be a good fit? If so, I'd be glad to beta it.
I don't think that it's a good fit for the vblade. The vblade is a userspace process. It interacts with a file, whether that's /tmp/foo or /dev/hda. Looking "under the hood" to see what's behind the file abstraction would be a design flaw for vblade. In the case of vblade exporting a SATA disk, it would be cleaner to talk SMART to the disk vblade is exporting and then make that information available as needed through another mechanism. The LayerWalker chip would always be talking directly to a real disk, so it would make sense for it to give SMART commands to the disk. I don't know whether it would, but I can't think of a reason why it wouldn't, since it's no extra effort for it to do so. It doesn't have to understand the ATA commands it passes to the disk. -- Ed Cashin <ecas...@coraid.com> Find experimental aoe Linux driver patches at http://coraid.typepad.com/aoe_linux_proving_grounds/ ------------------------------------------------------------------------------ _______________________________________________ Aoetools-discuss mailing list Aoetools-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
