Hello, We monitor sockets-enqueued on our NFS el6 servers which gives us an idea if our servers are being overloaded. On el7 3.10.0-1062.4.1.el7.x86_64 sockets-enqueued is always = packets-arrived making it appear that nfs threads are never immediately available to service requests. Example output EL7: /proc/fs/nfsd/pool_stats: # pool packets-arrived sockets-enqueued threads-woken threads-timedout 0 9919427 9919427 9867288 40774 Thread usage does not move while sockets-enqueued are incrementing: /proc/net/rpc/nfsd: ... th 32 0 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 ... Is this a bug?