NFS server not responding - Resolution Guide

Applies to

• ONTAP 9
• NFS

Description

• The "NFS server not responding" message  is logged when a client does not receive a timely response to a request.

Example:

Sep 18 08:38:02 server kernel: [2681704.006022] nfs: server 10.xx.xx.233 not responding, still trying

• A single missed/delayed response is enough to trigger the message, even if subsequent requests receive timely responses.
  • The NFS operation that did not get a response was likely sent 3 minutes prior.
    • This assumes that the client is using a mount timeo of 600 (60 seconds) and retrans of 2.
    • For more information see RHEL mount hangs: nfs: server [...] not responding, still trying - Red Hat Customer Portal.
• If the issue is not resolved, it may present as a mount hang.
• The issue typically requires a packet trace to diagnose, especially if ONTAP does not log related events
  • Capture dual packet traces on both storage and client: Capturing packet traces in ONTAP and clients - Resolution Guide
• PerfArchive data may indicate no change in IOPs during the errors, but low latency
• The messages may be due to:
  • Network partition
  • Transient error
  • Client failure
• If ONTAP is receiving the request but not responding via NFS before the client logs NFS server not responding, the following EMS message may appear during the hour following to indicate the number of NFS responses during the previous hour which have received that have exceeded the defined trigger time (default of 60 seconds).
  • Nblade.NfsResponseTraceTriggerHourly:debug]: params: {'responseCount': '14', 'trigger': '60'}