1.. SPDX-License-Identifier: GPL-2.0
   4Ceph Distributed File System
   7Ceph is a distributed network file system designed to provide good
   8performance, reliability, and scalability.
  10Basic features include:
  12 * POSIX semantics
  13 * Seamless scaling from 1 to many thousands of nodes
  14 * High availability and reliability.  No single point of failure.
  15 * N-way replication of data across storage nodes
  16 * Fast recovery from node failures
  17 * Automatic rebalancing of data on node addition/removal
  18 * Easy deployment: most FS components are userspace daemons
  22 * Flexible snapshots (on any directory)
  23 * Recursive accounting (nested files, directories, bytes)
  25In contrast to cluster filesystems like GFS, OCFS2, and GPFS that rely
  26on symmetric access by all clients to shared block devices, Ceph
  27separates data and metadata management into independent server
  28clusters, similar to Lustre.  Unlike Lustre, however, metadata and
  29storage nodes run entirely as user space daemons.  File data is striped
  30across storage nodes in large chunks to distribute workload and
  31facilitate high throughputs.  When storage nodes fail, data is
  32re-replicated in a distributed fashion by the storage nodes themselves
  33(with some minimal coordination from a cluster monitor), making the
  34system extremely efficient and scalable.
  36Metadata servers effectively form a large, consistent, distributed
  37in-memory cache above the file namespace that is extremely scalable,
  38dynamically redistributes metadata in response to workload changes,
  39and can tolerate arbitrary (well, non-Byzantine) node failures.  The
  40metadata server takes a somewhat unconventional approach to metadata
  41storage to significantly improve performance for common workloads.  In
  42particular, inodes with only a single link are embedded in
  43directories, allowing entire directories of dentries and inodes to be
  44loaded into its cache with a single I/O operation.  The contents of
  45extremely large directories can be fragmented and managed by
  46independent metadata servers, allowing scalable concurrent access.
  48The system offers automatic data rebalancing/migration when scaling
  49from a small cluster of just a few nodes to many hundreds, without
  50requiring an administrator carve the data set into static volumes or
  51go through the tedious process of migrating data between servers.
  52When the file system approaches full, new nodes can be easily added
  53and things will "just work."
  55Ceph includes flexible snapshot mechanism that allows a user to create
  56a snapshot on any subdirectory (and its nested contents) in the
  57system.  Snapshot creation and deletion are as simple as 'mkdir
  58.snap/foo' and 'rmdir .snap/foo'.
  60Ceph also provides some recursive accounting on directories for nested
  61files and bytes.  That is, a 'getfattr -d foo' on any directory in the
  62system will reveal the total number of nested regular files and
  63subdirectories, and a summation of all nested file sizes.  This makes
  64the identification of large disk space consumers relatively quick, as
  65no 'du' or similar recursive scan of the file system is required.
  67Finally, Ceph also allows quotas to be set on any directory in the system.
  68The quota can restrict the number of bytes or the number of files stored
  69beneath that point in the directory hierarchy.  Quotas can be set using
  70extended attributes 'ceph.quota.max_files' and 'ceph.quota.max_bytes', eg::
  72 setfattr -n ceph.quota.max_bytes -v 100000000 /some/dir
  73 getfattr -n ceph.quota.max_bytes /some/dir
  75A limitation of the current quotas implementation is that it relies on the
  76cooperation of the client mounting the file system to stop writers when a
  77limit is reached.  A modified or adversarial client cannot be prevented
  78from writing as much data as it needs.
  80Mount Syntax
  83The basic mount syntax is::
  85 # mount -t ceph user@fsid.fs_name=/[subdir] mnt -o mon_addr=monip1[:port][/monip2[:port]]
  87You only need to specify a single monitor, as the client will get the
  88full list when it connects.  (However, if the monitor you specify
  89happens to be down, the mount won't succeed.)  The port can be left
  90off if the monitor is using the default.  So if the monitor is at
  93 # mount -t ceph cephuser@07fe3187-00d9-42a3-814b-72a4d5e7d5be.cephfs=/ /mnt/ceph -o mon_addr=
  95is sufficient.  If /sbin/mount.ceph is installed, a hostname can be
  96used instead of an IP address and the cluster FSID can be left out
  97(as the mount helper will fill it in by reading the ceph configuration
 100  # mount -t ceph cephuser@cephfs=/ /mnt/ceph -o mon_addr=mon-addr
 102Multiple monitor addresses can be passed by separating each address with a slash (`/`)::
 104  # mount -t ceph cephuser@cephfs=/ /mnt/ceph -o mon_addr=
 106When using the mount helper, monitor address can be read from ceph
 107configuration file if available. Note that, the cluster FSID (passed as part
 108of the device string) is validated by checking it with the FSID reported by
 109the monitor.
 111Mount Options
 114  mon_addr=ip_address[:port][/ip_address[:port]]
 115        Monitor address to the cluster. This is used to bootstrap the
 116        connection to the cluster. Once connection is established, the
 117        monitor addresses in the monitor map are followed.
 119  fsid=cluster-id
 120        FSID of the cluster (from `ceph fsid` command).
 122  ip=A.B.C.D[:N]
 123        Specify the IP and/or port the client should bind to locally.
 124        There is normally not much reason to do this.  If the IP is not
 125        specified, the client's IP address is determined by looking at the
 126        address its connection to the monitor originates from.
 128  wsize=X
 129        Specify the maximum write size in bytes.  Default: 64 MB.
 131  rsize=X
 132        Specify the maximum read size in bytes.  Default: 64 MB.
 134  rasize=X
 135        Specify the maximum readahead size in bytes.  Default: 8 MB.
 137  mount_timeout=X
 138        Specify the timeout value for mount (in seconds), in the case
 139        of a non-responsive Ceph file system.  The default is 60
 140        seconds.
 142  caps_max=X
 143        Specify the maximum number of caps to hold. Unused caps are released
 144        when number of caps exceeds the limit. The default is 0 (no limit)
 146  rbytes
 147        When stat() is called on a directory, set st_size to 'rbytes',
 148        the summation of file sizes over all files nested beneath that
 149        directory.  This is the default.
 151  norbytes
 152        When stat() is called on a directory, set st_size to the
 153        number of entries in that directory.
 155  nocrc
 156        Disable CRC32C calculation for data writes.  If set, the storage node
 157        must rely on TCP's error correction to detect data corruption
 158        in the data payload.
 160  dcache
 161        Use the dcache contents to perform negative lookups and
 162        readdir when the client has the entire directory contents in
 163        its cache.  (This does not change correctness; the client uses
 164        cached metadata only when a lease or capability ensures it is
 165        valid.)
 167  nodcache
 168        Do not use the dcache as above.  This avoids a significant amount of
 169        complex code, sacrificing performance without affecting correctness,
 170        and is useful for tracking down bugs.
 172  noasyncreaddir
 173        Do not use the dcache as above for readdir.
 175  noquotadf
 176        Report overall filesystem usage in statfs instead of using the root
 177        directory quota.
 179  nocopyfrom
 180        Don't use the RADOS 'copy-from' operation to perform remote object
 181        copies.  Currently, it's only used in copy_file_range, which will revert
 182        to the default VFS implementation if this option is used.
 184  recover_session=<no|clean>
 185        Set auto reconnect mode in the case where the client is blocklisted. The
 186        available modes are "no" and "clean". The default is "no".
 188        * no: never attempt to reconnect when client detects that it has been
 189          blocklisted. Operations will generally fail after being blocklisted.
 191        * clean: client reconnects to the ceph cluster automatically when it
 192          detects that it has been blocklisted. During reconnect, client drops
 193          dirty data/metadata, invalidates page caches and writable file handles.
 194          After reconnect, file locks become stale because the MDS loses track
 195          of them. If an inode contains any stale file locks, read/write on the
 196          inode is not allowed until applications release all stale file locks.
 198More Information
 201For more information on Ceph, see the home page at
 204The Linux kernel client source tree is available at
 205        -
 206        - git://
 208and the source for the full system is at