NFS disconnections due to inode cache shrinking
Applies to
- ONTAP 9.7P9
- NFSv3
Issue
- High suspend on LOAD_INODE_NO_WIPS
Nblade.nfsLongRunningOp
messages might be present in EMS logs:
Sun Jul 18 21:11:49 +0200 [Cluster-02: kernel: Nblade.nfsLongRunningOp:debug]: Detected a long running network process operation. The client IP address:port is XXX.XXX.XXX.XXX:844. The local IP address:port is XXX.XXX.XXX.XXX:2049. The protocol requesting the operation is NFS3. The RPC Program Number for the operation is 100003. The RPC Procedure Number for the operation is 4. The disk process UUID is 53d962e8f59a11e9a00700a098ef419e. The Vserver identifier is 11.
- NFS clients might show:
Jul 18 19:18:14 client03 kernel: nfs: server XXX.16.230.71 not responding, still trying
wafl.inode.cache.highLoad
messages might appear in EMS logs:
Wed Aug 25 05:55:07 +0200 [wspfas9krm-01: wafl_exempt00: wafl.inode.cache.highLoad:info]:
Wed Aug 25 05:55:07 +0200 [wspfas9krm-01: wafl_exempt00: ems.engine.suppressed:debug]: Event 'wafl.inode.cache.highLoad' suppressed 101 times in last 115163 seconds.
df
command from the storage system might take long time to complete