SCSI command TimeOuts, checkCondition errors during reboot of a back-end device
Applies to
- Fabric MetroCluster
Issue
SCSI command TimeOuts and checkCondition errors from the Storage Enclosure Services (SES) devices reported during a reboot of an ATTO Bridge or a back-end switch on the same path.
With the reboot during a firmware upgrade of a Bridge we see the target and enclosure services devices report not ready:
Sat Jun 20 06:55:59 MSK [cluster_01-02: isp2400_asyncd_1: scsi.cmd.selectionTimeout:error]: Processor device FC_switch_A_2:7.126L0: Adapter/target error: HA status 0x7: cdb 0x3c. Targeted device did not respond to requested I/O. I/O will be retried.
Sat Jun 20 06:56:46 MSK [cluster_01-02: isp2400_intrd: scsi.cmd.checkCondition:error]: Enclosure services device FC_Switch_A_1:7.126L1025: Check Condition: CDB 0x3b: Sense Data SCSI:not ready - (0x2 - 0x35 0x2 0x0)(51).
...
We see only 3 seconds later the Enclosure services device reports request successful after retry over the redundant path:
Sat Jun 20 06:56:49 MSK [cluster_01-02: isp2400_intrd: scsi.cmd.retrySuccess:debug]: Enclosure services device FC_switch_A_2:7.126L25: request successful after retry #2/#0: cdb 0x3b (2023).