High VMware or application latency but ONTAP and UM show low due to bad upstream SFP
Applies to
- ONTAP 9
- Any user environment such as ESXi
- NAS protocols and iSCSI
Issue
- High latency is seen in VMware ESXi or application response time
- ONTAP latency is low
- Active IQ Unified Manager shows low response time as well
Example: vSphere reports high-volume latency or the ESX host reports latency in the log:
2020-09-17T01:31:14.178Z cpu44:2098185)WARNING: ScsiDeviceIO: 1564: Device naa.600a0920002002020202020202020202 performance has deteriorated. I/O latency increased from average value of 744 microseconds to 1515626 microseconds.
Example: qos statistics volume latency show
command shows low latency:
Cluster::> qos statistics volume latency show -volume Example_Vol -vserver VS1 Workload ID Latency Network Cluster Data Disk QoS Max QoS Min NVRAM Cloud FlexCache SM Sync VA --------------- ------ ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- -total- - 2.21ms 101.00us 75.00us 400.00us 1.62ms 0ms 0ms 7.00us 0ms 0ms 0ms 0ms Example_Vol.. 3195 1295.00us 478.00us 0ms 777.00us 26.00us 0ms 0ms 14.00us 0ms 0ms