On Thu, Aug 25, 2016 at 8:33 PM, Phil Turmel <philip@xxxxxxxxxx> wrote: > On 08/25/2016 06:32 PM, Chris Murphy wrote: > >>> It's possible, but why would you ever end up with a GPT in a partition? >> >> In every case I've seen, it was user error. I haven't heard of things >> putting GPTs in partitions, and in a sense I'd say it's a bug if any >> utility lets a user do that. Nesting GPT's in partitions, bad idea, >> although it *should* be innocuous because it shouldn't be seen/honored >> by anything that doesn't go looking for it because it doesn't belong >> there. > > It is possible to run gdisk or parted on /dev/sdX1 accidentally instead > of /dev/sdX. Pretty simple user error. > > It is also possible and appropriate if using v0.90 or v1.0 metadata on > an array and you partition the array itself. Then it'll show up on > member 0, any mirror of member 0, and possibly on a parity disk (if > intervening blocks are zero). Right, so something like GPT on /dev/sda and /dev/sdb to create sda1 and sdb1, then mdadm -C /dev/md0 --metadata=1.0 ... /dev/sda1 /dev/sdb1, and then create a GPT on /dev/md0. The result is /dev/md0, /dev/sda1, and /dev/sda2 will all appear to have the same GPT on them. I would say that's probably a bad idea, I know some tools allow it, but it creates an ambiguity. It could be argued to be inconsistent with the UEFI spec. The only nesting it describes is MBR on a GPT partition, not GPT nested in a GPT partition. This is probably also better done using LVM. Otherwise we get nutty things... -- Chris Murphy -- 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
