7 The Ceph Filesystem (CephFS) is a POSIX-compliant filesystem that uses
8 a Ceph Storage Cluster to store its data. The Ceph filesystem uses the same Ceph
9 Storage Cluster system as Ceph Block Devices, Ceph Object Storage with its S3
10 and Swift APIs, or native bindings (librados).
12 .. note:: If you are evaluating CephFS for the first time, please review
13 the best practices for deployment: :doc:`/cephfs/best-practices`
16 +-----------------------+ +------------------------+
18 | | +------------------------+
20 | | +------------------------+
21 | CephFS Kernel Object | | CephFS Library |
22 | | +------------------------+
24 | | +------------------------+
26 +-----------------------+ +------------------------+
28 +---------------+ +---------------+ +---------------+
29 | OSDs | | MDSs | | Monitors |
30 +---------------+ +---------------+ +---------------+
36 Using the Ceph Filesystem requires at least one :term:`Ceph Metadata Server` in
37 your Ceph Storage Cluster.
43 <style type="text/css">div.body h3{margin:5px 0px 0px 0px;}</style>
44 <table cellpadding="10"><colgroup><col width="33%"><col width="33%"><col width="33%"></colgroup><tbody valign="top"><tr><td><h3>Step 1: Metadata Server</h3>
46 To run the Ceph Filesystem, you must have a running Ceph Storage Cluster with at
47 least one :term:`Ceph Metadata Server` running.
53 Add/Remove MDS(s) <add-remove-mds>
54 MDS states <mds-states>
55 MDS failover and standby configuration <standby>
56 MDS Configuration Settings <mds-config-ref>
57 Client Configuration Settings <client-config-ref>
58 Journaler Configuration <journaler>
59 Manpage ceph-mds <../../man/8/ceph-mds>
63 </td><td><h3>Step 2: Mount CephFS</h3>
65 Once you have a healthy Ceph Storage Cluster with at least
66 one Ceph Metadata Server, you may create and mount your Ceph Filesystem.
67 Ensure that your client has network connectivity and the proper
68 authentication keyring.
73 Create CephFS <createfs>
75 Mount CephFS as FUSE <fuse>
76 Mount CephFS in fstab <fstab>
77 Use the CephFS Shell <cephfs-shell>
78 Supported Features of Kernel Driver <kernel-features>
79 Manpage ceph-fuse <../../man/8/ceph-fuse>
80 Manpage mount.ceph <../../man/8/mount.ceph>
81 Manpage mount.fuse.ceph <../../man/8/mount.fuse.ceph>
86 </td><td><h3>Additional Details</h3>
91 Deployment best practices <best-practices>
92 MDS States <mds-states>
93 Administrative commands <administration>
94 Understanding MDS Cache Size Limits <cache-size-limits>
95 POSIX compatibility <posix>
96 Experimental Features <experimental-features>
98 Using Ceph with Hadoop <hadoop>
99 cephfs-journal-tool <cephfs-journal-tool>
100 File layouts <file-layouts>
101 Client eviction <eviction>
102 Handling full filesystems <full>
103 Health messages <health-messages>
104 Troubleshooting <troubleshooting>
105 Disaster recovery <disaster-recovery>
106 Client authentication <client-auth>
107 Upgrading old filesystems <upgrading>
108 Configuring directory fragmentation <dirfrags>
109 Configuring multiple active MDS daemons <multimds>
110 Export over NFS <nfs>
111 Application best practices <app-best-practices>
117 Advanced: Metadata repair <disaster-recovery-experts>
121 </td></tr></tbody></table>
129 Client's Capabilities <capabilities>
130 libcephfs <../../api/libcephfs-java/>