MetroCluster IP high write latency switch isl port errors
Applies to
- MetroCluster IP
- ONTAP 9
- Cisco backend switches
Issue
- Clients experience high write latency.
- Switch ISL port shows a high number of errors:
switch01# show interface counters errors
--------------------------------------------------------------------------------
Port Align-Err FCS-Err Xmit-Err Rcv-Err UnderSize OutDiscards
--------------------------------------------------------------------------------
Eth1/15 557 921949 0 922506 0 8452
- scsi.cmd.aborted and transport errors are seen in event logs:
[node01: scsi_cmdblk_strthr_admin: scsi.cmd.abortedByHost:error]: Disk device 0m.i2.2L3: Command aborted by host adapter: HA status 0x13: cdb 0x9a:000000015e4bc200:0002:0300.
[node01: doneq0: scsi.mcc.adt.ioTransportError:error]: mcc_adt[2] - Transport error during execution of command: HA status 0x13: CAM transport status 0x1b : cdb 0x9a:000000015e4c6600:0002:0300.
- Write commands are successful on retry:
[node01: scsi_cmdblk_strthr_admin: scsi.cmd.retrySuccess:debug]: Disk device 0v.i2.0L3: request successful after retry #1/#0: cdb 0x9a:000000015e4c3800:0002:0300 (13728).