Parity2 calculation, unlike parity1, is dependent on the order of the disks, simple as that. And, your syslog says it was correcting Q (parity2).
Feb 22 16:32:15 Server kernel: mdcmd (43): check
Feb 22 16:32:15 Server kernel: md: recovery thread: check P Q ...
Feb 22 16:32:15 Server kernel: md: recovery thread: Q corrected, sector=0
Feb 22 16:32:15 Server kernel: md: recovery thread: Q corrected, sector=8
...
Feb 22 16:32:15 Server kernel: md: recovery thread: Q corrected, sector=816
Feb 22 16:32:15 Server kernel: md: recovery thread: Q corrected, sector=824
Feb 22 16:32:15 Server kernel: md: recovery thread: stopped logging
So this is completely normal. Run a non-correcting parity check to verify and everything should be fine.