--- /dev/null
+<?xml version='1.0' encoding='UTF-8'?><chapter xmlns="http://docbook.org/ns/docbook" xmlns:xl="http://www.w3.org/1999/xlink" version="5.0" xml:lang="en-US" xml:id="zfssnapshots">
+ <title xml:id="zfssnapshots.title">Lustre ZFS Snapshots</title>
+ <para>This chapter describes the ZFS Snapshot feature support in Lustre and
+ contains following sections:</para>
+ <itemizedlist>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotIntro"/></para>
+ </listitem>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotConfig"/></para>
+ </listitem>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotOps"/></para>
+ </listitem>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotBarrier"/></para>
+ </listitem>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotLogs"/></para>
+ </listitem>
+ <listitem>
+ <para><xref linkend="dbdoclet.zfssnapshotLustreLogs"/></para>
+ </listitem>
+ </itemizedlist>
+ <section xml:id="dbdoclet.zfssnapshotIntro">
+ <title><indexterm><primary>Introduction</primary>
+ </indexterm>Introduction</title>
+ <para>Snapshots provide fast recovery of files from a previously created
+ checkpoint without recourse to an offline backup or remote replica.
+ Snapshots also provide a means to version-control storage, and can be used
+ to recover lost files or previous versions of files.</para>
+ <para>Filesystem snapshots are intended to be mounted on user-accessible
+ nodes, such as login nodes, so that users can restore files (e.g. after
+ accidental delete or overwrite) without administrator intervention. It
+ would be possible to mount the snapshot filesystem(s) via automount when
+ users access them, rather than mounting all snapshots, to reduce overhead
+ on login nodes when the snapshots are not in use.</para>
+ <para>Recovery of lost files from a snapshot is usually considerably
+ faster than from any offline backup or remote replica. However, note that
+ snapshots do not improve storage reliability and are just as exposed to
+ hardware failure as any other storage volume.</para>
+ <section xml:id="dbdoclet.zfssnapshotsReq">
+ <title><indexterm><primary>Introduction</primary>
+ <secondary>Requirements</secondary></indexterm>Requirements
+ </title>
+ <para>All Lustre server targets must be ZFS file systems running
+ Lustre version 2.10 or later. In addition, the MGS must be able to
+ communicate via ssh or another remote access protocol, without
+ password authentication, to all other servers.</para>
+ <para>The feature is enabled by default and cannot be disabled. The
+ management of snapshots is done through <literal>lctl</literal>
+ commands on the MGS.</para>
+ <para>Lustre snapshot is based on Copy-On-Write; the snapshot and file
+ system may share a single copy of the data until a file is changed on
+ the file system. The snapshot will prevent the space of deleted or
+ overwritten files from being released until the snapshot(s)
+ referencing those files is deleted. The file system administrator
+ needs to establish a snapshot create/backup/remove policy according to
+ their system’s actual size and usage.</para>
+ </section>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotConfig">
+ <title><indexterm><primary>feature overview</primary>
+ <secondary>configuration</secondary></indexterm>Configuration
+ </title>
+ <para>The snapshot tool loads system configuration from the
+ <literal>/etc/ldev.conf</literal> file on the MGS and calls related
+ ZFS commands to maintian the Lustre snapshot pieces on all targets
+ (MGS/MDT/OST). Please note that the <literal>/etc/ldev.conf</literal>
+ file is used for other purposes as well.</para>
+ <para>The format of the file is:</para>
+ <screen><host> foreign/- <label> <device> [journal-path]/- [raidtab]</screen>
+ <para>The format of <literal><label></literal> is:</para>
+ <screen>fsname-<role><index> or <role><index></screen>
+ <para>The format of <device> is:</para>
+ <screen>[md|zfs:][pool_dir/]<pool>/<filesystem></screen>
+ <para>Snapshot only uses the fields <host>, <label> and
+ <device>.</para>
+ <para>Example:</para>
+ <screen>mgs# cat /etc/ldev.conf
+host-mdt1 - myfs-MDT0000 zfs:/tmp/myfs-mdt1/mdt1
+host-mdt2 - myfs-MDT0001 zfs:myfs-mdt2/mdt2
+host-ost1 - OST0000 zfs:/tmp/myfs-ost1/ost1
+host-ost2 - OST0001 zfs:myfs-ost2/ost2</screen>
+ <para>The configuration file is edited manually.</para>
+ <para> Once the configuration file is updated to reflect the current
+ file system setup, you are ready to create a file system snapshot.
+ </para>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotOps">
+ <title><indexterm><primary>operations</primary>
+ </indexterm>Snapshot Operations</title>
+ <section xml:id="dbdoclet.zfssnapshotCreate">
+ <title><indexterm><primary>operations</primary>
+ <secondary>create</secondary></indexterm>Creating a Snapshot
+ </title>
+ <para>To create a snapshot of an existing Lustre file system, run the
+ following <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl snapshot_create [-b | --barrier [on | off]] [-c | --comment
+comment] -F | --fsname fsname> [-h | --help] -n | --name ssname>
+[-r | --rsh remote_shell][-t | --timeout timeout]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis></para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-b</literal></para>
+ </entry>
+ <entry>
+ <para>set write barrier before creating snapshot. The
+ default value is 'on'.</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-c</literal></para>
+ </entry>
+ <entry>
+ <para>a description for the purpose of the snapshot
+ </para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the name of the snapshot</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-t</literal></para>
+ </entry>
+ <entry>
+ <para>the lifetime (seconds) for write barrier. The
+ default value is 30 seconds</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotDelete">
+ <title><indexterm><primary>operations</primary>
+ <secondary>delete</secondary></indexterm>Delete a Snapshot
+ </title>
+ <para>To delete an existing snapshot, run the following
+ <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl snapshot_destroy [-f | --force] <-F | --fsname fsname>
+<-n | --name ssname> [-r | --rsh remote_shell]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-f</literal></para>
+ </entry>
+ <entry>
+ <para>destroy the snapshot by force</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the name of the snapshot</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotMount">
+ <title><indexterm><primary>operations</primary>
+ <secondary>mount</secondary></indexterm>Mounting a Snapshot
+ </title>
+ <para>Snapshots are treated as separate file systems and can be mounted on
+ Lustre clients. The snapshot file system must be mounted as a
+ read-only file system with the <literal>-o ro</literal> option.
+ If the <literal>mount</literal> command does not include the read-only
+ option, the mount will fail.</para>
+ <note><para>Before a snapshot can be mounted on the client, the snapshot
+ must first be mounted on the servers using the <literal>lctl</literal>
+ utility.</para></note>
+ <para>To mount a snapshot on the server, run the following lctl command
+ on the MGS:</para>
+ <screen>lctl snapshot_mount <-F | --fsname fsname> [-h | --help]
+<-n | --name ssname> [-r | --rsh remote_shell]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the name of the snapshot</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ <para>After the successful mounting of the snapshot on the server, clients
+ can now mount the snapshot as a read-only filesystem. For example, to
+ mount a snapshot named <replaceable>snapshot_20170602</replaceable> for a
+ filesystem named <replaceable>myfs</replaceable>, the following mount
+ command would be used:</para>
+ <screen>mgs# lctl snapshot_mount -F myfs -n snapshot_20170602</screen>
+ <para>After mounting on the server, use
+ <literal>lctl snapshot_list</literal> to get the fsname for the snapshot
+ itself as follows:</para>
+ <screen>ss_fsname=$(lctl snapshot_list -F myfs -n snapshot_20170602 |
+ awk '/^snapshot_fsname/ { print $2 }')</screen>
+ <para>Finally, mount the snapshot on the client:</para>
+ <screen>mount -t lustre -o ro $MGS_nid:/$ss_fsname $local_mount_point</screen>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotUnmount">
+ <title><indexterm><primary>operations</primary>
+ <secondary>unmount</secondary></indexterm>Unmounting a Snapshot
+ </title>
+ <para>To unmount a snapshot from the servers, first unmount the snapshot
+ file system from all clients, using the standard <literal>umount</literal>
+ command on each client. For example, to unmount the snapshot file system
+ named <replaceable>snapshot_20170602</replaceable> run the following
+ command on each client that has it mounted:</para>
+ <screen>client# umount $local_mount_point</screen>
+ <para>After all clients have unmounted the snapshot file system, run the
+ following <literal>lctl</literal>command on a server node where the
+ snapshot is mounted:</para>
+ <screen>lctl snapshot_umount [-F | --fsname fsname] [-h | --help]
+<-n | -- name ssname> [-r | --rsh remote_shell]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the name of the snapshot</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ <para>For example:</para>
+ <screen>lctl snapshot_umount -F myfs -n snapshot_20170602</screen>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotList">
+ <title><indexterm><primary>operations</primary>
+ <secondary>list</secondary></indexterm>List Snapshots
+ </title>
+ <para>To list the available snapshots for a given file system, use the
+ following <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl snapshot_list [-d | --detail] <-F | --fsname fsname>
+[-h | -- help] [-n | --name ssname] [-r | --rsh remote_shell]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-d</literal></para>
+ </entry>
+ <entry>
+ <para>list every piece for the specified snapshot
+ </para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the snapshot's name. If the snapshot name is not
+ supplied, all snapshots for this file system will be
+ displayed</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotModify">
+ <title><indexterm><primary>operations</primary>
+ <secondary>modify</secondary></indexterm>Modify Snapshot Attributes
+ </title>
+ <para>Currently, Lustre snapshot has five user visible attributes;
+ snapshot name, snapshot comment, create time, modification time, and
+ snapshot file system name. Among them, the former two attributes can be
+ modified. Renaming follows the general ZFS snapshot name rules, such as
+ the maximum length is 256 bytes, cannot conflict with the reserved names,
+ and so on.</para>
+ <para>To modify a snapshot’s attributes, use the following
+ <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl snapshot_modify [-c | --comment comment]
+<-F | --fsname fsname> [-h | --help] <-n | --name ssname>
+[-N | --new new_ssname] [-r | --rsh remote_shell]</screen>
+ <informaltable frame="all">
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Option</emphasis></para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Description</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>-c</literal></para>
+ </entry>
+ <entry>
+ <para>update the snapshot's comment</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-F</literal></para>
+ </entry>
+ <entry>
+ <para>the filesystem name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-h</literal></para>
+ </entry>
+ <entry>
+ <para>help information</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-n</literal></para>
+ </entry>
+ <entry>
+ <para>the snapshot's name</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-N</literal></para>
+ </entry>
+ <entry>
+ <para>rename the snapshot's name as
+ <replaceable>new_ssname</replaceable></para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>-r</literal></para>
+ </entry>
+ <entry>
+ <para>the remote shell used for communication with
+ remote target. The default value is 'ssh'</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </informaltable>
+ </section>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotBarrier">
+ <title><indexterm><primary>barrier</primary>
+ </indexterm>Global Write Barriers</title>
+ <para>Snapshots are non-atomic across multiple MDTs and OSTs, which means
+ that if there is activity on the file system while a snapshot is being
+ taken, there may be user-visible namespace inconsistencies with files
+ created or destroyed in the interval between the MDT and OST snapshots.
+ In order to create a consistent snapshot of the file system, we are able
+ to set a global write barrier, or “freeze” the system. Once set, all
+ metadata modifications will be blocked until the write barrier is actively
+ removed (“thawed”) or expired. The user can set a timeout parameter on a
+ global barrier or the barrier can be explicitly removed. The default
+ timeout period is 30 seconds.</para>
+ <para>It is important to note that snapshots are usable without the global
+ barrier. Only files that are currently being modified by clients (write,
+ create, unlink) may be inconsistent as noted above if the barrier is not
+ used. Other files not curently being modified would be usable even
+ without the barrier.</para>
+ <para>The snapshot create command will call the write barrier internally
+ when requested using the <literal>-b</literal> option to
+ <literal>lctl snapshot_create</literal>. So, explicit use of the barrier
+ is not required when using snapshots but included here as an option to
+ quiet the file system before a snapshot is created.</para>
+ <section xml:id="dbdoclet.zfssnapshotBarrierImpose">
+ <title><indexterm><primary>barrier</primary>
+ <secondary>impose</secondary></indexterm>Impose Barrier
+ </title>
+ <para>To impose a global write barrier, run the
+ <literal>lctl barrier_freeze</literal> command on the MGS:</para>
+ <screen>lctl barrier_freeze <fsname> [timeout (in seconds)]
+where timeout default is 30.</screen>
+ <para>For example, to freeze the filesystem
+ <replaceable>testfs</replaceable> for <literal>15</literal> seconds:
+ </para>
+ <screen>mgs# lctl barrier_freeze testfs 15</screen>
+ <para>If the command is successful, there will be no output from
+ the command. Otherwise, an error message will be printed.</para>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotBarrierRemove">
+ <title><indexterm><primary>barrier</primary>
+ <secondary>remove</secondary></indexterm>Remove Barrier
+ </title>
+ <para>To remove a global write barrier, run the
+ <literal>lctl barrier_thaw</literal> command on the MGS:</para>
+ <screen>lctl barrier_thaw <fsname></screen>
+ <para>For example, to thaw the write barrier for the filesystem
+ <replaceable>testfs</replaceable>:
+ </para>
+ <screen>mgs# lctl barrier_thaw testfs</screen>
+ <para>If the command is successful, there will be no output from
+ the command. Otherwise, an error message will be printed.</para>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotBarrierQuery">
+ <title><indexterm><primary>barrier</primary>
+ <secondary>query</secondary></indexterm>Query Barrier
+ </title>
+ <para>To see how much time is left on a global write barrier, run the
+ <literal>lctl barrier_stat</literal> command on the MGS:</para>
+ <screen># lctl barrier_stat <fsname></screen>
+ <para>For example, to stat the write barrier for the filesystem
+ <replaceable>testfs</replaceable>:
+ </para>
+ <screen>mgs# lctl barrier_stat testfs
+The barrier for testfs is in 'frozen'
+The barrier will be expired after 7 seconds</screen>
+ <para>If the command is successful, a status from the table below
+ will be printed. Otherwise, an error message will be printed.</para>
+ <para>The possible status and related meanings for the write barrier
+ are as follows:</para>
+ <table frame="all" xml:id="writebarrierstatus.tab1">
+ <title>Write Barrier Status</title>
+ <tgroup cols="2">
+ <colspec colname="c1" colwidth="50*"/>
+ <colspec colname="c2" colwidth="50*"/>
+ <thead>
+ <row>
+ <entry>
+ <para><emphasis role="bold">Status</emphasis>
+ </para>
+ </entry>
+ <entry>
+ <para><emphasis role="bold">Meaning</emphasis>
+ </para>
+ </entry>
+ </row>
+ </thead>
+ <tbody>
+ <row>
+ <entry>
+ <para> <literal>init</literal></para>
+ </entry>
+ <entry>
+ <para>barrier has never been set on the system
+ </para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>freezing_p1</literal></para>
+ </entry>
+ <entry>
+ <para>In the first stage of setting the write
+ barrier</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>freezing_p2</literal></para>
+ </entry>
+ <entry>
+ <para> the second stage of setting the write
+ barrier</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>frozen</literal></para>
+ </entry>
+ <entry>
+ <para>the write barrier has been set successfully
+ </para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>thawing</literal></para>
+ </entry>
+ <entry>
+ <para>In thawing the write barrier</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>thawed</literal></para>
+ </entry>
+ <entry>
+ <para>The write barrier has been thawed</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>failed</literal></para>
+ </entry>
+ <entry>
+ <para>Failed to set write barrier</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>expired</literal></para>
+ </entry>
+ <entry>
+ <para>The write barrier is expired</para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>rescan</literal></para>
+ </entry>
+ <entry>
+ <para>In scanning the MDTs status, see the command
+ <literal>barrier_rescan</literal></para>
+ </entry>
+ </row>
+ <row>
+ <entry>
+ <para> <literal>unknown</literal></para>
+ </entry>
+ <entry>
+ <para>Other cases</para>
+ </entry>
+ </row>
+ </tbody>
+ </tgroup>
+ </table>
+ <para>If the barrier is in ’freezing_p1’, ’freezing_p2’ or ’frozen’
+ status, then the remaining lifetime will be returned also.</para>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotBarrierRescan">
+ <title><indexterm><primary>barrier</primary>
+ <secondary>rescan</secondary></indexterm>Rescan Barrier
+ </title>
+ <para> To rescan a global write barrier to check which MDTs are
+ active, run the <literal>lctl barrier_rescan</literal> command on the
+ MGS:</para>
+ <screen>lctl barrier_rescan <fsname> [timeout (in seconds)],
+where the default timeout is 30 seconds.</screen>
+ <para>For example, to rescan the barrier for filesystem
+ <replaceable>testfs</replaceable>:</para>
+ <screen>mgs# lctl barrier_rescan testfs
+1 of 4 MDT(s) in the filesystem testfs are inactive</screen>
+ <para>If the command is successful, the number of MDTs that are
+ unavailable against the total MDTs will be reported. Otherwise, an
+ error message will be printed.</para>
+ </section>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotLogs">
+ <title><indexterm><primary>logs</primary>
+ </indexterm>Snapshot Logs</title>
+ <para>A log of all snapshot activity can be found in the following file:
+ <literal>/var/log/lsnapshot.log</literal>. This file contains information
+ on when a snapshot was created, an attribute was changed, when it was
+ mounted, and other snapshot information.</para>
+ <para>The following is a sample <literal>/var/log/lsnapshot</literal>
+ file:</para>
+ <screen>Mon Mar 21 19:43:06 2016
+(15826:jt_snapshot_create:1138:scratch:ssh): Create snapshot lss_0_0
+successfully with comment <(null)>, barrier <enable>, timeout <30>
+Mon Mar 21 19:43:11 2016(13030:jt_snapshot_create:1138:scratch:ssh):
+Create snapshot lss_0_1 successfully with comment <(null)>, barrier
+<disable>, timeout <-1>
+Mon Mar 21 19:44:38 2016 (17161:jt_snapshot_mount:2013:scratch:ssh):
+The snapshot lss_1a_0 is mounted
+Mon Mar 21 19:44:46 2016
+(17662:jt_snapshot_umount:2167:scratch:ssh): the snapshot lss_1a_0
+have been umounted
+Mon Mar 21 19:47:12 2016
+(20897:jt_snapshot_destroy:1312:scratch:ssh): Destroy snapshot
+lss_2_0 successfully with force <disable></screen>
+ </section>
+ <section xml:id="dbdoclet.zfssnapshotLustreLogs">
+ <title><indexterm><primary>configlogs</primary>
+ </indexterm>Lustre Configuration Logs</title>
+ <para>A snapshot is independent from the original file system that it is
+ derived from and is treated as a new file system name that can be mounted
+ by Lustre client nodes. The file system name is part of the configuration
+ log names and exists in configuration log entries. Two commands exist to
+ manipulate configuration logs: <literal>lctl fork_lcfg</literal> and
+ <literal>lctl erase_lcfg</literal>.</para>
+ <para>The snapshot commands will use configuration log functionality
+ internally when needed. So, use of the barrier is not required to use
+ snapshots but included here as an option. The following configuration log
+ commands are independent of snapshots and can be used independent of
+ snapshot use.</para>
+ <para>To fork a configuration log, run the following
+ <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl fork_lcfg</screen>
+ <para>Usage: fork_lcfg <fsname> <newname></para>
+ <para>To erase a configuration log, run the following
+ <literal>lctl</literal> command on the MGS:</para>
+ <screen>lctl erase_lcfg</screen>
+ <para>Usage: erase_lcfg <fsname></para>
+ </section>
+</chapter>
git://git.whamcloud.com / doc / manual.git / commitdiff