RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-04 02:02:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge PR #30626 into master

Browse Source 
* refs/pull/30626/head:
	doc: document recover_session= option in mount.ceph manpage
	doc: update documents on using kcephfs

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
This commit is contained in:
Patrick Donnelly 2019-10-02 05:15:02 -07:00
commit 98f0682b0e
No known key found for this signature in database
GPG Key ID: 3A2A7E25BEA8AADB
4 changed files with 54 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
doc/cephfs/fstab.rst
View File

@ -11,11 +11,11 @@ Kernel Driver
To mount CephFS in your file systems table as a kernel driver, add the
following to ``/etc/fstab``::
{ipaddress}:{port}:/ {mount}/{mountpoint} {file-system-name} [name=username,secret=secretkey|secretfile=/path/to/secretfile],[{mount.options}]
[{ipaddress}:{port}]:/ {mount}/{mountpoint} ceph [name=username,secret=secretkey|secretfile=/path/to/secretfile],[{mount.options}]
For example::
10.10.10.10:6789:/ /mnt/ceph ceph name=admin,noatime,_netdev 0 2
:/ /mnt/ceph ceph name=admin,noatime,_netdev 0 2
The default for the ``name=`` parameter is ``guest``. If the ``secret`` or
``secretfile`` options are not specified then the mount helper will attempt to

17
doc/cephfs/kernel-features.rst
View File

@ -1,13 +1,20 @@
Supported Features of Kernel Driver
Supported Features of the Kernel Driver
========================================
The kernel driver is developed separately from the core ceph code, and as
such it sometimes differs from the FUSE driver in feature implementation.
The following details the implementation status of various CephFS features
in the kernel driver.
Inline data
-----------
Inline data was introduced by the Firefly release. Linux kernel clients >= 3.19
can read inline data, can convert existing inline data to RADOS objects when
file data is modified. At present, Linux kernel clients do not store file data
as inline data.
Inline data was introduced by the Firefly release. This feature is being
deprecated in mainline CephFS, and may be removed from a future kernel
release.
Linux kernel clients >= 3.19 can read inline data and convert existing
inline data to RADOS objects when file data is modified. At present,
Linux kernel clients do not store file data as inline data.
See `Experimental Features`_ for more information.

48
doc/cephfs/kernel.rst
View File

@ -2,36 +2,52 @@
Mount CephFS with the Kernel Driver
====================================
To mount CephFS with the Kernel Driver you may use the ``mount`` command if you know the
monitor host IP address(es), or use the ``mount.ceph`` utility to resolve the
monitor host name(s) into IP address(es) for you. For example::
It is recommended to install the ``/sbin/mount.ceph`` kernel mount
helper if working with the kernel cephfs driver. To mount CephFS with
the kernel driver you may use the ``mount`` command::
sudo mkdir /mnt/mycephfs
sudo mount -t ceph 192.168.0.1:6789:/ /mnt/mycephfs
# mkdir /mnt/mycephfs
# mount -t ceph :/ /mnt/mycephfs
To mount the Ceph file system with ``cephx`` authentication enabled, the kernel
must authenticate with the cluster. The default ``name=`` option is ``guest``.
The mount.ceph helper will automatically attempt to find a secret key in the
keyring.
Omitting the monitor addresses will cue ``mount.ceph`` to look them up
in the local configuration file. If you know at least one of the monitor
addresses and ports you can specify them directly in a comma-separated
list, and the kernel will avoid looking them up::
The secret can also be specified manually with the ``secret=`` option. ::
# mount -t ceph 192.168.0.1:6789,192.168.0.2:6789:/ /mnt/mycephfs
sudo mount -t ceph 192.168.0.1:6789:/ /mnt/mycephfs -o name=admin,secret=AQATSKdNGBnwLhAAnNDKnH65FmVKpXZJVasUeQ==
To mount a subtree of the cephfs root, append the path to the device
string ::
The foregoing usage leaves the secret in the Bash history. A more secure
approach reads the secret from a file. For example::
# mount -t ceph :/subvolume/dir1/dir2 /mnt/mycephfs -o name=fs
sudo mount -t ceph 192.168.0.1:6789:/ /mnt/mycephfs -o name=admin,secretfile=/etc/ceph/admin.secret
To mount the Ceph file system with ``cephx`` authentication enabled, the
kernel must authenticate with the cluster. The default ``name=`` option
is ``guest``. The mount.ceph helper will automatically attempt to find
a secret key in a cephx keyring if it's not specified. For example, to
mount the filesystem as the cephx user ``fs``::
# mount -t ceph :/ /mnt/mycephfs -o name=fs
The secret can also be specified manually with the ``secret=`` option.::
# mount -t ceph :/ /mnt/mycephfs -o name=fs,secret=AQATSKdNGBnwLhAAnNDKnH65FmVKpXZJVasUeQ==
For legacy usage, mount.ceph can be told to read a lone secret from a
file. For example::
# mount -t ceph :/ /mnt/mycephfs -o name=fs,secretfile=/etc/ceph/admin.secret
See `User Management`_ for details on cephx.
If you have more than one file system, specify which one to mount using
the ``mds_namespace`` option, e.g.::
the ``mds_namespace`` option, e.g. ``-o mds_namespace=myfs``.
# mount -t ceph :/ -o name=fs,mds_namespace=myfs
To unmount the Ceph file system, you may use the ``umount`` command. For example::
sudo umount /mnt/mycephfs
# umount /mnt/mycephfs
.. tip:: Ensure that you are not within the file system directories before
executing this command.

8
doc/man/8/mount.ceph.rst
View File

@ -135,6 +135,14 @@ Options
for autodiscovery of monitor addresses and auth secrets. The default is
to use the standard search path for ceph.conf files.
:command:`recover_session=<no|clean>`
Set auto reconnect mode in the case where the client is blacklisted. The
available modes are ``no`` and ``clean``. The default is ``no``.
- ``no``: never attempt to reconnect when client detects that it has been blacklisted. Blacklisted clients will not attempt to reconnect and their operations will fail too.
- ``clean``: client reconnects to the ceph cluster automatically when it detects that it has been blacklisted. During reconnect, client drops dirty data/metadata, invalidates page caches and writable file handles. After reconnect, file locks become stale because the MDS loses track of them. If an inode contains any stale file locks, read/write on the inode is not allowed until applications release all stale file locks.
Mount Secrets
=============
If the `secret` and `secretfile` options are not specified on the command-line