3 ==========================================
4 cephfs-top -- Ceph Filesystem Top Utility
5 ==========================================
7 .. program:: cephfs-top
12 | **cephfs-top** [flags]
18 **cephfs-top** provides top(1) like functionality for Ceph Filesystem.
19 Various client metrics are displayed and updated in realtime.
21 Ceph Metadata Servers periodically send client metrics to Ceph Manager.
22 ``Stats`` plugin in Ceph Manager provides an interface to fetch these metrics.
29 Cluster: Ceph cluster to connect. Defaults to ``ceph``.
33 Id: Client used to connect to Ceph cluster. Defaults to ``fstop``.
35 .. option:: --selftest
37 Perform a selftest. This mode performs a sanity check of ``stats`` module.
39 .. option:: --conffile [CONFFILE]
41 Path to cluster configuration file
43 .. option:: -d [DELAY], --delay [DELAY]
45 Refresh interval in seconds (default: 1)
49 Dump the metrics to stdout
51 .. option:: --dumpfs <fs_name>
53 Dump the metrics of the given filesystem to stdout
55 Descriptions of fields
56 ======================
68 number of opened files
76 number of opened inodes
80 total size of read IOs
84 total size of write IOs
88 average size of read IOs
92 average size of write IOs
96 speed of read IOs compared with the last refresh
100 speed of write IOs compared with the last refresh
102 .. describe:: rlatavg
108 standard deviation (variance) for read latency
110 .. describe:: wlatavg
112 average write latency
116 standard deviation (variance) for write latency
118 .. describe:: mlatavg
120 average metadata latency
124 standard deviation (variance) for metadata latency
129 **cephfs-top** is part of Ceph, a massively scalable, open-source, distributed storage system. Please refer to the Ceph documentation at
130 http://ceph.com/ for more information.
136 :doc:`ceph <ceph>`\(8),
137 :doc:`ceph-mds <ceph-mds>`\(8)