Node reporting false IO latency for partner owned disks

Last updated
Save as PDF
Share
1. Share
2. Tweet
3. Share

Views:: 20

Visibility:: Public

Votes:: 0

Category:: fas-systems

Specialty:: hw

Last Updated:

Applies to

ONTAP 9
FAS Systems
AFF Systems

Issue

High latency is reported against a disk, but the reporting node does not own the disk. EMS messages will look similar to:

[Node01: disk_latency_monitor:shm.threshold.ioLatency:debug]: Disk 11d.63.9 has exceeded the expected IO latency in the current window with average latency of 580 msecs and average utilization of 29 percent.

Node that owns the disk reports no latency issue with the drive.
The drive is the single data disk in a 3 disk root aggregate.

Aggregate aggr_root_node02 (online, raid_dp) (block checksums)
  Plex /aggr_root_node02/plex0 (online, normal, active, pool0)
    RAID group /aggr_root_node02/plex0/rg0 (normal, block checksums)

      RAID Disk	Device   	HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks)
      ---------	------   	------------- ---- ---- ---- ----- --------------    --------------
      dparity 	2a.65.0  	2a    65  0   SA:A   0  FSAS  7200 3807816/7798408704 3815447/7814037168 
      parity  	9a.51.0  	9a    51  0   SA:A   0  FSAS  7200 3807816/7798408704 3815447/7814037168 
      data    	9b.63.9  	9b    63  9   SA:B   0  FSAS  7200 3807816/7798408704 3815447/7814037168