Resilvered disks set to incorrect block size of 512B instead of 4096 like rest of pool
After replacing and resilvering two disks in a four disk zfs pool I noticed that the block size for the resilvered disks does not match the rest of the pool.
NAME STATE READ WRITE CKSUM
files ONLINE 0 0 0
raidz1-0 ONLINE 0 0 0
gptid/285e9b9e-17c2-11e6-8af4-d8d385b3ae72 ONLINE 0 0 0 block size: 512B configured, 4096B native
gptid/4c7c67e3-2475-11e6-b5f3-d8d385b3ae72 ONLINE 0 0 0
gptid/63a4fb1b-1ad2-11e6-8af4-d8d385b3ae72 ONLINE 0 0 0 block size: 512B configured, 4096B native
gptid/e8d101c5-2b4a-11e2-9357-d8d385b3ae72 ONLINE 0 0 0
The pool was originally created using block size of 4096 for all drives.
The resilvered drives have a block size of 512B. There was no option to choose the block during the resilvering process as best I can recall.
Please add a feature where the block size for resilvered disks will automatically match the oldest disk in the pool