1
0
mirror of https://github.com/samba-team/samba.git synced 2024-12-23 17:34:34 +03:00
samba-mirror/docs-xml/manpages/vfs_glusterfs.8.xml

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

214 lines
6.4 KiB
XML
Raw Normal View History

<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE refentry PUBLIC "-//Samba-Team//DTD DocBook V4.2-Based Variant V1.0//EN" "http://www.samba.org/samba/DTD/samba-doc">
<refentry id="vfs_glusterfs.8">
<refmeta>
<refentrytitle>vfs_glusterfs</refentrytitle>
<manvolnum>8</manvolnum>
<refmiscinfo class="source">Samba</refmiscinfo>
<refmiscinfo class="manual">System Administration tools</refmiscinfo>
<refmiscinfo class="version">&doc.version;</refmiscinfo>
</refmeta>
<refnamediv>
<refname>vfs_glusterfs</refname>
<refpurpose>
Utilize features provided by GlusterFS
</refpurpose>
</refnamediv>
<refsynopsisdiv>
<cmdsynopsis>
<command>vfs objects = glusterfs</command>
</cmdsynopsis>
</refsynopsisdiv>
<refsect1>
<title>DESCRIPTION</title>
<para>This VFS module is part of the
<citerefentry><refentrytitle>samba</refentrytitle>
<manvolnum>8</manvolnum></citerefentry> suite.</para>
<para>
GlusterFS
(<ulink url="http://www.gluster.org/">http://www.gluster.org</ulink>)
is an Open Source clustered file system capable of scaling to
several peta-bytes. With its FUSE based native client,
GlusterFS is available as a POSIX compliant file system and can
hence be shared by Samba without additional steps.
</para>
<para>
The <command>vfs_glusterfs</command> VFS module provides an
alternative, and superior way to access a Gluster filesystem
from Samba for sharing. It does not require a Gluster FUSE mount
but directly accesses the GlusterFS daemon through its library
<command>libgfapi</command>, thereby omitting the expensive
kernel-userspace context switches and taking advantage of some
of the more advanced features of GlusterFS.
</para>
<para>
This module can be combined with other modules, but it
should be the last module in the <command>vfs objects</command>
list. Modules added to this list to the right of the glusterfs
entry may not have any effect at all.
</para>
</refsect1>
<refsect1>
<title>CONFIGURATION</title>
<para>
A basic configuration looks like this.
</para>
<programlisting>
<smbconfsection name="[share]"/>
<smbconfoption name="vfs objects">glusterfs</smbconfoption>
<smbconfoption name="path">/relative/base/path</smbconfoption>
<smbconfoption name="glusterfs:volume">gv0</smbconfoption>
<smbconfoption name="kernel share modes">no</smbconfoption>
</programlisting>
<para>
Note that since <command>vfs_glusterfs</command> does not
require a Gluster mount, the share <command>path</command> is
treated differently than for other shares: It is interpreted as
the base path of the share relative to the gluster volume used.
Because this is usually not at the same time a system path, in a
ctdb cluster setup where ctdb manages Samba, you need to set
<command>CTDB_SAMBA_SKIP_SHARE_CHECK=yes</command> in ctdb's
configuration file. Otherwise ctdb will not get healthy.
</para>
<para>
Note that currently kernel share modes have to be disabled
in a share running with the glusterfs vfs module for file
serving to work properly.
</para>
</refsect1>
<refsect1>
<title>OPTIONS</title>
<variablelist>
<varlistentry>
<term>glusterfs:logfile = path</term>
<listitem>
<para>
Defines whether and where to store a vfs_glusterfs specific
logfile. Client variable substitution is supported (i.e.
%M, %m, %I), hence per client log file can be specified.
</para>
<para>
Example: glusterfs:logfile =
/var/log/samba/glusterfs-vol2.%M.log
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>glusterfs:loglevel = 0-9</term>
<listitem>
<para>
Defines the level of logging, with higher numbers corresponding to more verbosity.
0 - No logs; 9 - Trace log level; 7 being the info log level is preferred.
</para>
<para>
If this option is not defined with an explicit loglevel,
the glusterfs default is used (currently loglevel 7).
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>glusterfs:volfile_server = servername</term>
<listitem>
<para>
Defines which volfile server to use, defaults to
support volfile fetch from multiple glusterd nodes glusterfs:volfile_server option can be used in smb.conf to define where to fetch the volfile from. Currently it supports only a single IP or a hostname. The default is 'localhost'. glfs_set_volfile_server() has been enhanced in gfapi to support multiple invocations. A list is maintained in libgfapi which gets appended on every invocation. When glfs_init is performed, libgfapi would first try to fetch the volfile from glusterd on that node. However, on failure to fetch the volfile, it would proceed to contact glusterd on every node in the list until it gets the volfile or exhausts the list. This enhacement was done in Gluster commit [2]. This commit is available in 3.6, 3.7, 3.8 versions of Gluster. As we cannot have multiple lines having the same key of glusterfs:volfile_server in a share definition in smb.conf, we propose a scheme like this: where value of glusterfs:volfile_server could be list of white space seperated elements where each element could be unix+/path/to/socket/file or [tcp+]IP|hostname|\[IPv6\][:port]. Note the restriction on naming a IPv6 host, it follows the same restriction that is based on IPv6 naming in URL as per RFC 2732[1]. [1] http://www.ietf.org/rfc/rfc2732.txt [2] 0c1d78f5c52c69268ec3a1d8d5fcb1a1bf15f243 Signed-off-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Michael Adam <obnox@samba.org> Reviewed-by: Guenther Deschner <gd@samba.org> Autobuild-User(master): Günther Deschner <gd@samba.org> Autobuild-Date(master): Fri Oct 14 17:09:24 CEST 2016 on sn-devel-144
2016-08-25 13:33:55 +03:00
localhost. It could be list of white space
separated elements where each element could be
support volfile fetch from multiple glusterd nodes glusterfs:volfile_server option can be used in smb.conf to define where to fetch the volfile from. Currently it supports only a single IP or a hostname. The default is 'localhost'. glfs_set_volfile_server() has been enhanced in gfapi to support multiple invocations. A list is maintained in libgfapi which gets appended on every invocation. When glfs_init is performed, libgfapi would first try to fetch the volfile from glusterd on that node. However, on failure to fetch the volfile, it would proceed to contact glusterd on every node in the list until it gets the volfile or exhausts the list. This enhacement was done in Gluster commit [2]. This commit is available in 3.6, 3.7, 3.8 versions of Gluster. As we cannot have multiple lines having the same key of glusterfs:volfile_server in a share definition in smb.conf, we propose a scheme like this: where value of glusterfs:volfile_server could be list of white space seperated elements where each element could be unix+/path/to/socket/file or [tcp+]IP|hostname|\[IPv6\][:port]. Note the restriction on naming a IPv6 host, it follows the same restriction that is based on IPv6 naming in URL as per RFC 2732[1]. [1] http://www.ietf.org/rfc/rfc2732.txt [2] 0c1d78f5c52c69268ec3a1d8d5fcb1a1bf15f243 Signed-off-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Jeremy Allison <jra@samba.org> Reviewed-by: Michael Adam <obnox@samba.org> Reviewed-by: Guenther Deschner <gd@samba.org> Autobuild-User(master): Günther Deschner <gd@samba.org> Autobuild-Date(master): Fri Oct 14 17:09:24 CEST 2016 on sn-devel-144
2016-08-25 13:33:55 +03:00
</para>
<para>
1. unix+/path/to/socket/file
</para>
<para>
2. [tcp+]IP|hostname|\[IPv6\][:port]
</para>
<para>
Note the restriction on naming a IPv6 host, it follows
the same restriction that is based on IPv6 naming in
URL as per RFC 2732.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>glusterfs:volume = volumename</term>
<listitem>
<para>
Defines the glusterfs volumename to use for this share.
</para>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>CAVEATS</title>
<para>
The GlusterFS write-behind performance translator, when used
with Samba, could be a source of data corruption. The
translator, while processing a write call, immediately returns
success but continues writing the data to the server in the
background. This can cause data corruption when two clients
relying on Samba to provide data consistency are operating on
the same file.
</para>
<para>
The write-behind translator is enabled by default on GlusterFS.
The vfs_glusterfs plugin will check for the presence of the
translator and refuse to connect if detected.
Please disable the write-behind translator for the GlusterFS
volume to allow the plugin to connect to the volume.
The write-behind translator can easily be disabled via calling
<programlisting>
gluster volume set &lt;volumename&gt; performance.write-behind off
</programlisting> on the commandline.
</para>
<para>
With GlusterFS versions >= 9, we silently bypass write-behind
translator during initial connect and failure is avoided.
</para>
</refsect1>
<refsect1>
<title>VERSION</title>
<para>
This man page is part of version &doc.version; of the Samba suite.
</para>
</refsect1>
<refsect1>
<title>AUTHOR</title>
<para>The original Samba software and related utilities
were created by Andrew Tridgell. Samba is now developed
by the Samba Team as an Open Source project similar
to the way the Linux kernel is developed.</para>
</refsect1>
</refentry>