On 11/30/20 10:31 AM, Guoqing Jiang wrote:
On 11/19/20 12:41, Zhao Heming wrote:
mkfs.xfs /dev/md0
This would make people think clustered raid can co-work with local fs well ...
This issue came from real customer issue. For easy understanding, I simplify the scenario. The real usage is customer exports clustered array to only one virtual-machine as storage (xfs on it). The host side follows clustered rules, use clustered lvm volumes on top of clustered array.
mdadm --manage --add /dev/md0 /dev/sdi mdadm --wait /dev/md0 mdadm --grow --raid-devices=3 /dev/md0
mdadm /dev/md0 --fail /dev/sdg mdadm /dev/md0 --remove /dev/sdg mdadm --grow --raid-devices=2 /dev/md0
The above is good fit for a new test case in mdadm.
A new test case may name: 02r1_Grow_add_remove When I have time, I will do this job.
Thanks, Guoqing
Thanks.