Orange Pi5 kernel

Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards

3 Commits   0 Branches   0 Tags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   1) ============================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   2) Kernel NFS Server Statistics
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   3) ============================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   4) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   5) :Authors: Greg Banks <gnb@sgi.com> - 26 Mar 2009
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   6) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   7) This document describes the format and semantics of the statistics
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   8) which the kernel NFS server makes available to userspace.  These
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300   9) statistics are available in several text form pseudo files, each of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  10) which is described separately below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  11) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  12) In most cases you don't need to know these formats, as the nfsstat(8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  13) program from the nfs-utils distribution provides a helpful command-line
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  14) interface for extracting and printing them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  15) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  16) All the files described here are formatted as a sequence of text lines,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  17) separated by newline '\n' characters.  Lines beginning with a hash
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  18) '#' character are comments intended for humans and should be ignored
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  19) by parsing routines.  All other lines contain a sequence of fields
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  20) separated by whitespace.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  21) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  22) /proc/fs/nfsd/pool_stats
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  23) ========================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  24) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  25) This file is available in kernels from 2.6.30 onwards, if the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  26) /proc/fs/nfsd filesystem is mounted (it almost always should be).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  27) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  28) The first line is a comment which describes the fields present in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  29) all the other lines.  The other lines present the following data as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  30) a sequence of unsigned decimal numeric fields.  One line is shown
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  31) for each NFS thread pool.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  32) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  33) All counters are 64 bits wide and wrap naturally.  There is no way
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  34) to zero these counters, instead applications should do their own
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  35) rate conversion.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  36) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  37) pool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  38) 	The id number of the NFS thread pool to which this line applies.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  39) 	This number does not change.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  40) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  41) 	Thread pool ids are a contiguous set of small integers starting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  42) 	at zero.  The maximum value depends on the thread pool mode, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  43) 	currently cannot be larger than the number of CPUs in the system.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  44) 	Note that in the default case there will be a single thread pool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  45) 	which contains all the nfsd threads and all the CPUs in the system,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  46) 	and thus this file will have a single line with a pool id of "0".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  47) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  48) packets-arrived
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  49) 	Counts how many NFS packets have arrived.  More precisely, this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  50) 	is the number of times that the network stack has notified the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  51) 	sunrpc server layer that new data may be available on a transport
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  52) 	(e.g. an NFS or UDP socket or an NFS/RDMA endpoint).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  53) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  54) 	Depending on the NFS workload patterns and various network stack
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  55) 	effects (such as Large Receive Offload) which can combine packets
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  56) 	on the wire, this may be either more or less than the number
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  57) 	of NFS calls received (which statistic is available elsewhere).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  58) 	However this is a more accurate and less workload-dependent measure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  59) 	of how much CPU load is being placed on the sunrpc server layer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  60) 	due to NFS network traffic.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  61) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  62) sockets-enqueued
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  63) 	Counts how many times an NFS transport is enqueued to wait for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  64) 	an nfsd thread to service it, i.e. no nfsd thread was considered
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  65) 	available.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  66) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  67) 	The circumstance this statistic tracks indicates that there was NFS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  68) 	network-facing work to be done but it couldn't be done immediately,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  69) 	thus introducing a small delay in servicing NFS calls.  The ideal
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  70) 	rate of change for this counter is zero; significantly non-zero
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  71) 	values may indicate a performance limitation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  72) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  73) 	This can happen because there are too few nfsd threads in the thread
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  74) 	pool for the NFS workload (the workload is thread-limited), in which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  75) 	case configuring more nfsd threads will probably improve the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  76) 	performance of the NFS workload.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  77) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  78) threads-woken
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  79) 	Counts how many times an idle nfsd thread is woken to try to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  80) 	receive some data from an NFS transport.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  81) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  82) 	This statistic tracks the circumstance where incoming
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  83) 	network-facing NFS work is being handled quickly, which is a good
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  84) 	thing.  The ideal rate of change for this counter will be close
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  85) 	to but less than the rate of change of the packets-arrived counter.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  86) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  87) threads-timedout
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  88) 	Counts how many times an nfsd thread triggered an idle timeout,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  89) 	i.e. was not woken to handle any incoming network packets for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  90) 	some time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  91) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  92) 	This statistic counts a circumstance where there are more nfsd
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  93) 	threads configured than can be used by the NFS workload.  This is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  94) 	a clue that the number of nfsd threads can be reduced without
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  95) 	affecting performance.  Unfortunately, it's only a clue and not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  96) 	a strong indication, for a couple of reasons:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  97) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  98) 	 - Currently the rate at which the counter is incremented is quite
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300  99) 	   slow; the idle timeout is 60 minutes.  Unless the NFS workload
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) 	   remains constant for hours at a time, this counter is unlikely
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) 	   to be providing information that is still useful.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) 	 - It is usually a wise policy to provide some slack,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) 	   i.e. configure a few more nfsds than are currently needed,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) 	   to allow for future spikes in load.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) Note that incoming packets on NFS transports will be dealt with in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) one of three ways.  An nfsd thread can be woken (threads-woken counts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) this case), or the transport can be enqueued for later attention
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) (sockets-enqueued counts this case), or the packet can be temporarily
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) deferred because the transport is currently being used by an nfsd
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) thread.  This last case is not very interesting and is not explicitly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) counted, but can be inferred from the other counters thus::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) 	packets-deferred = packets-arrived - ( sockets-enqueued + threads-woken )
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) More
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) ====
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) 
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) Descriptions of the other statistics file should go here.