Self-healing fibre channel link

A system for verifying the integrity of a communication link, the system comprising,

• a first adapter to detect, during normal operation of a communication link, a condition indicative of a high failure rate (HFR) on the communication link;
• the first adapter further configured to place the communication link in an HFR state as a result of the condition, wherein placing the communication link in HFR state comprises suspending transmission of real I/Os over the communication link while the communication link is in HFR state;
• the first adapter further configured to repeatedly transmit test I/Os to a second adapter over the communication link while the communication link is in HFR state;
• and
• the first adapter further configured to automatically exit HFR state and resume normal operation on the communication link once at least one of the following has occurred:
• a selected number of test I/Os have been successfully transmitted over the communication link, and a number of test I/Os have been successfully transmitted over the communication link for a selected amount of time.

An apparatus for verifying the integrity of a communication link, the apparatus comprising,

• a detection module to detect, during normal operation of a communication link, a condition indicative of a high failure rate (HFR) on the communication link;
• a state module to place the communication link in an HFR state as a result of the condition, wherein placing the communication link in HFR state comprises suspending transmission of real I/Os over the communication link while the communication link is in HFR state;
• a test module to repeatedly transmit test I/Os over the communication link while the communication link is in HFR state;
• and
• an exit module configured to automatically exit HFR state and resume normal operation once at least one of the following has occurred:
• a selected number of test I/Os have been successfully transmitted over the communication link, and a number of test I/Os have been successfully transmitted over the communication link for a selected amount of time.