SAS expander reset on DS460C disk shelf drawer caused disks to fail

• MetroCluster IP
• DS460C

1. Plex failed after many disk failed at the same time
2. ::> run -node <nodename> -command sysconfig -r shows the failed disks
3. ::> storage shelf drawer show -fields shelf,drawer,firmware-a,firmware-b,serial-number,part-number,vendor, fru-type,error,mfg-date is missing information for some of the failed disks
4. ::> storage shelf drawer show-phy might shows enabled enabled enabled-rate-unknown enabled-12gbs
5. Some of the drives may recover and go back as a data disks and spare but some of them don't.
6. After the recovery the error messages intermittently still occur, only on the affected drawer:

cluster-02: scsi_cmdblk_strthr_admin: scsi.cmd.abortedByHost:error]: Disk device 0c.12.14: Command aborted by host adapter: HA status 0x4: cdb 0x28:bd9d57d8:0008......................................

.............................................................
[cluster-02: scsi_cmdblk_strthr_admin: scsi.cmd.retrySuccess:debug]: Disk device 4d.12.14: request successful after retry #1/#0: cdb 0x2f:04519800:0400 (8712).
[cluster-02: scsi_cmdblk_strthr_admin: scsi.cmd.retrySuccess:debug]: Disk device 4d.12.14: request successful after retry #1/#0: cdb 0x88:0000000197576710:00000020 (17071).

7. And another expander reset is seen in the event logs:

[cluster-02: acpa_interface_main: acp.exp.reset:info]: SAS expander reset issued to 4d.12.B (NA), (disk shelf serial number: SHJHU2048000006).
[cluster-02: acp_ib_response_thread: acp.exp.reset.success:info]: SAS expander reset command sent to 4d.12.B (NA) was successful, (disk shelf serial number: SHJHU2048000006).