SVM Raid 5 mirror stuck in resynching state

I could use some advice on a disk replacement issue for a T5220 SPARC server where a replaced disk is stuck in the resynching state. 

There is an SVM mirrored RAID 5 volume comprised of 4 disks and 2 hot spares.  Originally, there were 2 failed disks replaced by the hot spares and another disk which was in “MAINTENANCE” state.  An unexpected reboot cleared the error logs and there were no further errors on the disk in “MAINTENANCE” state.  My goal was to first replace the failed disks so the hot spares would be released just in case the third disk failed when I issued the metareplace command to repair this disk.