David and others,
I'd like to add that evms ( http://evms.sourceforge.net/ ) already has a bad-block management layer, maybe it could be merged inside md to have write errors management as well, maybe :)
Good point.
When I set up my RAID (albeit a relatviely non-critical domestic system) I considered whether or not to use EVMS because of the bad block relocation feature.
In the end, I went with md, lvm2, and mdadm, primarily as I recall because at the time EVMS did not support lvm2 and also required additional kernel patches to support bad block relocation. Plus, there were issues about having the root filesystem managed by EVMS (it requires the set up of a custom initrd).
All things considered, I chose the path of least resistance and went with md/lvm2/mdadm. Perhaps it's time to reconsider that and look at EVMS again?
R. -- http://robinbowes.com - 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
