Migration of data in response to write-failure of disk drive head

A method, comprising:

• detecting a failure of a head that reads from and writes to an affected surface of a disk of a disk drive, the disk drive being part of a drive array, the failure preventing the head from writing to the affected surface but not preventing the head from reading from the affected surface;
• performing a remediation initiated by at least one of a storage controller and a host coupled to the drive array, the remediation comprising:
• determining blocks of spare capacity of other drives of the drive array;
• and
• copying data from the affected surface to the blocks of spare capacity of the other drives via a peer-to-peer data transfer within the array, the peer-to-peer data transfer performed independently of the storage controller and the host;
• and
• using the blocks of spare capacity in place of the affected surface for data storage and retrieval subsequent to the failure.

A system comprising a drive array that includes a disk drive, the disk drive comprising:

• a read/write channel operable to communicate with a plurality of heads that read and write to a respective plurality of disk surfaces of the disk drive;
• and
• one or more processors coupled to the read/write channel and operable via instructions to perform:
• detect a write-failed head of the plurality of heads associated with an affected one of the plurality of disk surfaces, the write-failed head unable to write to the affected surface but able to read from the affected surface;
• in response to detecting the write-failed head, perform a remediation comprising:
• determining spare capacity blocks on other surfaces of the plurality of surfaces different than the affected surface;
• copying data from the affected surface to the spare capacity blocks via an internal copy function within the disk drive;
• determining that the spare capacity blocks have insufficient capacity to store all of the data from the affected surface;
• in coordination with at least one of the storage controller and the host, identifying blocks of spare capacity of other drives of the drive array;
• and
• copying at least part of the data from the affected surface of the disk drive to the blocks of spare capacity of the other drives via a peer-to-peer data transfer within the array, the copying initiated by at least one of the storage controller and the host coupled, wherein the peer-to-peer data transfer after the initiation is performed independently of the storage controller and the host;
• and
• use the spare capacity blocks in place of the affected surface for data storage and retrieval subsequent to detecting the write-failed head,
• the system further comprising a storage controller and a host coupled to the drive array, wherein the copying of the data from the affected surface to the spare capacity blocks is internally performed by the disk drive.