how to avoid fc-al loop down when there is one disk failure

VMAX monitor and control leverages redundant LCCs and dual path to each disk in loop, and usebypass methods to control the failure disks. Once any of disk getting failed, the VMAX will bypass the disk in the loop and invoke sparing within Raid protection.

See following figure:

Each disk enclosure contains a pair of link control cards (LCC A and B) that provide redundant connections to the disk. The LCCs monitor the Fibre Channel layer for connection and protocol errors. If errors occur, the status is reported and, if necessary, bypass circuits are used to switch out of the drive.

Symmetrix systems also include point-to-point back-end switches that enhance fault detection and isolation. The back-end has an independent relationship with each drive on the loop that optimizes problem isolation and improves serviceability. This dedicated relationship between a back-end controller and each disk drive also allows Symmetrix systems to analyze drive health prior to adding new disks to an existing configuration. Symmetrix systems will not add a faulty drive to an existing loop.

To ensure the highest level of availability, the communications link to the monitoring and control functions is not carried by the Fibre Channel loop. It is implemented through a separate path. And, if the Fibre Channel loop is not operating, the director can also use the out-of-band control path to access the LCC and reconfigure the loop into a working state