On 10/25/2017 01:13 PM, Bjørnar Ness wrote: > I am not asking for _A_ way to solve this, lets try to stick to the problem > reported: md writing to inactive spares on rebuild. I gave the most likely reason in my first reply. All member devices, including spares, carry array metadata in their superblock. That metadata includes event counts and rebuild progress information that MD will need to safely continue if your system crashes in the middle of that operation. If you strongly believe that shouldn't be the case, you'll need to prepare a patch that changes the behavior, post the patch on this list for review, and convince the maintainers that you're right and that your patch is well-formed. Part of your argument will undoubtedly have to be *why* this case needs special handling in the kernel when the obvious (userspace) solution is to not attach excessive numbers of spares. Phil -- 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
