NIS latency causes NFS hang due to indeterminate resolution in export evaluation
Applies to
- ONTAP
- NFS w/ netgroup exports
- NIS providing netgroup
Issue
- NFS mounts appear to hang while ONTAP is unable to complete netgroup lookups in adequate time.
- ONTAP will fail to be able to send additional requests to NIS as the max parallel operations have been reached.
- sysctl sysvar nblade can see increasing counters:
sysvar.nblade.ngprocess.access.RpcNoSendParallel: 9563593847
sysvar.nblade.ngprocess.rewind.giveup.nfs.export: 5539997
- The number of pipes and the statistics around each of these pipes via
::> statistics. statistics show -object nblade_rpc_client -instance AccessCacheMgwdClient-*