NFSHung and NFSStale explanation for CDM probe

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

The CDM probe has the ability to detect "Hung" and "Stale" states for NFS filesystems.  What do these states mean and how is it determined when a filesystem is Hung or Stale?

Environment

DX UIM - Any Version
cdm probe - 8.01+
NFS Filesystems

Resolution

Hung State:

• Occurs when the NFS server is unreachable
• Typically due to network issues or server down
• Detected by operation timeout
• Error code: -2
• The threshold on this is determined by the timeout set in the <disk> section of the cdm.cfg (default 90 seconds)

Stale State:

• Occurs when the NFS file handle is no longer valid
• Typically happens when the exported directory on the server was removed/changed while mounted
• Detected by specific NFS errors
• Error code: -3
• This does not have a threshold - the threshold value in the config file is a meaningless placeholder

Additional Information

For the 'Stale' state, the threshold value as seen below is not actually used for any functionality - the structure of the config file requires a threshold to be present since it relates to an alarm, but the actual value is not considered.  A filesystem is either stale or not - the value set for threshold does not matter as long as the value is set.