I see some contention in network processing node according to Unified Manager. The description says the following:
Network Processing Represents the software component in the cluster involved with I/O processing between the protocol layer and the cluster. The node handling network processing might have changed since the event was detected. If the network processing component is in contention, it means high utilization at the network processing node is impacting the latency of one or more workloads.
Anyone else have more detail what exactly is this "software component" being referred to?
"Network Processing" for an IO request occurs on the node that owns the LIF and includes the effort to translate the IO from a protocol one (NFS/SMB/iSCSI/FCP) to a cluster one (wafl/spinnp). If you have contention on this component either (a) your CPU utilization was very high on the node owning the LIF (i.e. high queue time waiting for CPU) or (b) the request was very expensive to process (i.e. high service time on CPU).
In the case of (a) you should see that CPU was very high on the node owning the LIF at the of the incident.
In the case of (b) you may see errors/retries on some component (check event log, check network error counters), or a software bug that caused some excessive looping could be to blame. To research this deeper I would contact technical support.
Hope this helps!
Cheers, Chris Madden
Storage Architect, NetApp EMEA (and author of Harvest)