ONTAP SAN tape backup job errors with SCSI aborts
Applies to
- ONTAP
- NDMP
- FC
- NetVault
Issue
- During backup using NDMP with FC SAN connected tape library errors occur , commands aborted , CRC errors , command timeouts
- The FC cable has been swapped with another port with no errors
- No errors when SFP is replaced with known good one
Example:
storage_cluster::> event log show -event scsi., tape., fci.
Time Node Severity Event
------------------- ---------------- ------------- ---------------------------
DEBUG fci.device.login.status: A login delay has occurred between Fibre Channel adapter 0f and tape device 1:2:3, with error subcode 0x703.
ERROR tape.device.cmdAbort: Tape device 1:2:3 command aborted by host, ha_status 0x5.
ERROR scsi.cmd.abortedByHost: Tape device 1:2:3: Command aborted by host adapter: HA status 0x4: cdb 0x1a.
DEBUG fci.device.quiesce: Adapter 0f encountered a command timeout on Tape device 1:2:3 LUN 0 cdb 0x1a retry: 0 Quiescing the device.
DEBUG fci.device.timeout: HBA 0f encountered a device timeout on Tape device 1:2:3 LUN 0 cdb 0x1a retry: 0
ERROR scsi.path.excessiveErrors: Excessive errors encountered by adapter 0f on tape device 1:2:3.
ERROR scsi.cmd.pastTimeToLive: Tape device 1:2:3: request failed after try #1: cdb 0x12.
CRC :
storage (SN: 123) - e0d had 1073 CRC errors in the past 13 days, 23:50:13. (9972 total, uptime of 207 days,