HDFS-5778. Add rolling upgrade user document.

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/HDFS-5535@1571508 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Tsz-wo Sze 2014-02-25 01:14:12 +00:00
parent 1c6b5d2b58
commit 8c4d8058d8
3 changed files with 273 additions and 0 deletions

View File

@ -83,3 +83,5 @@ HDFS-5535 subtasks:
Arpit Agarwal)
HDFS-5583. Make DN send an OOB Ack on shutdown before restarting. (kihwal)
HDFS-5778. Add rolling upgrade user document. (szetszwo)

View File

@ -0,0 +1,270 @@
<?xml version="1.0" encoding="UTF-8"?>
<!--
Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to You under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
<document xmlns="http://maven.apache.org/XDOC/2.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/XDOC/2.0 http://maven.apache.org/xsd/xdoc-2.0.xsd">
<properties>
<title>HDFS Rolling Upgrade</title>
</properties>
<body>
<h1>HDFS Rolling Upgrade</h1>
<macro name="toc">
<param name="section" value="0"/>
<param name="fromDepth" value="0"/>
<param name="toDepth" value="4"/>
</macro>
<section name="Introduction" id="Introduction">
<p>
<em>HDFS rolling upgrade</em> allows upgrading individual HDFS daemons.
For examples, the datanodes can be upgraded independent of the namenodes.
A namenode can be upgraded independent of the other namenodes.
The namenodes can be upgraded independent of datanods and journal nodes.
</p>
</section>
<section name="Upgrade" id="Upgrade">
<p>
In Hadoop v2, HDFS supports highly-available (HA) namenode services and wire compatibility.
These two capabilities make it feasible to upgrade HDFS without incurring HDFS downtime.
In order to upgrade a HDFS cluster without downtime, the cluster must be setup with HA.
</p>
<subsection name="Upgrade without Downtime" id="UpgradeWithoutDowntime">
<p>
In a HA cluster, there are two or more <em>NameNodes (NNs)</em>, many <em>DataNodes (DNs)</em>,
a few <em>JournalNodes (JNs)</em> and a few <em>ZooKeeperNodes (ZKNs)</em>.
<em>JNs</em> is relatively stable and does not require upgrade when upgrading HDFS in most of the cases.
In the rolling upgrade procedure described here,
only <em>NNs</em> and <em>DNs</em> are considered but <em>JNs</em> and <em>ZKNs</em> are not.
Upgrading <em>JNs</em> and <em>ZKNs</em> may incur cluster downtime.
</p>
<h4>Upgrading Non-Federated Clusters</h4>
<p>
Suppose there are two namenodes <em>NN1</em> and <em>NN2</em>,
where <em>NN1</em> and <em>NN2</em> are respectively in active and standby states.
The following are the steps for upgrading a HA cluster:
</p>
<ol>
<li>Prepare Rolling Upgrade<ul>
<li>Run "<code><a href="#dfsadmin_-rollingUpgrade">hdfs dfsadmin -rollingUpgrade prepare</a></code>"
to create a fsimage for rollback.
</li>
<li>Run "<code><a href="#dfsadmin_-rollingUpgrade">hdfs dfsadmin -rollingUpgrade query</a></code>"
to check the status of the rollback image.
Wait and re-run the command until the "Proceed with rolling upgrade" message is shown.
</li>
</ul></li>
<li>Upgrade Active and Standby <em>NNs</em><ol>
<li>Shutdown, upgrade and restart <em>NN2</em> as standby.</li>
<li>Failover from <em>NN1</em> to <em>NN2</em>
so that <em>NN2</em> becomes active and <em>NN1</em> becomes standby.</li>
<li>Shutdown, upgrade and restart <em>NN1</em> as standby.</li>
</ol></li>
<li>Upgrade <em>DNs</em><ol>
<li>Choose a small subset of datanodes (e.g. all datanodes under a particular rack).</li>
<ol>
<li>Run "<code><a href="#dfsadmin_-shutdownDatanode">hdfs dfsadmin -shutdownDatanode &lt;DATANODE_HOST:IPC_PORT&gt; upgrade</a></code>"
to shutdown one of the chosen datanodes.</li>
<li>Run "<code><a href="#dfsadmin_-getDatanodeInfo">hdfs dfsadmin -getDatanodeInfo &lt;DATANODE_HOST:IPC_PORT&gt;</a></code>"
to check and wait for the datanode to shutdown.</li>
<li>Upgrade and restart the datanode.</li>
<li>Repeat the above steps for all the chosen datanodes in the subset.</li>
</ol>
<li>Repeat the above steps until all datanodes in the cluster are upgraded.</li>
</ol></li>
<li>Finalize Rolling Upgrade<ul>
<li>Run "<code><a href="#dfsadmin_-rollingUpgrade">hdfs dfsadmin -rollingUpgrade finalize</a></code>"
to finalize the rolling upgrade.</li>
</ul></li>
</ol>
<h4>Upgrading Federated Clusters</h4>
<p>
In a federated cluster, there are multiple namespaces
and a pair of active and standby <em>NNs</em> for each namespace.
The procedure for upgrading a federated cluster is similar to upgrading a non-federated cluster
except that Step 1 and Step 4 are performed on each namespace
and Step 2 is performed on each pair of active and standby <em>NNs</em>, i.e.
</p>
<ol>
<li>Prepare Rolling Upgrade for Each Namespace</li>
<li>Upgrade Active and Standby <em>NN</em> pairs for Each Namespace</li>
<li>Upgrade <em>DNs</em></li>
<li>Finalize Rolling Upgrade for Each Namespace</li>
</ol>
</subsection>
<subsection name="Upgrade with Downtime" id="UpgradeWithDowntime">
<p>
For non-HA clusters,
it is impossible to upgrade HDFS without downtime since it requires restarting the namenodes.
However, datanodes can still be upgraded in a rolling manner.
</p>
<h4>Upgrading Non-HA Clusters</h4>
<p>
In a non-HA cluster, there are a <em>NameNode (NN)</em>, a <em>SecondaryNameNode (SNN)</em>
and many <em>DataNodes (DNs)</em>.
The procedure for upgrading a non-HA cluster is similar to upgrading a HA cluster
except that Step 2 "Upgrade Active and Standby <em>NNs</em>" is changed to below:
</p>
<ul>
<li>Upgrade <em>NN</em> and <em>SNN</em><ol>
<li>Shutdown <em>SNN</em></li>
<li>Shutdown, upgrade and restart <em>NN</em></li>
<li>Upgrade and restart <em>SNN</em></li>
</ol></li>
</ul>
</subsection>
</section>
<section name="Downgrade and Rollback" id="DowngradeAndRollback">
<p>
When the upgraded release is undesirable
or, in some unlikely case, the upgrade fails (due to bugs in the newer release),
administrators may choose to downgrade HDFS back to the pre-upgrade release
or rollback HDFS to the pre-upgrade release and the pre-upgrade state.
Both downgrade and rollback require cluster downtime and are not done in a rolling fashion.
</p>
<p>
Note that downgrade and rollback are possible only after a rolling upgrade is started and
before the upgrade is terminated.
An upgrade can be terminated by either finalize, downgrade or rollback.
Therefore, it is impossible to run rollback after finalize or downgrade,
or to run downgrade after finalize.
</p>
<subsection name="Downgrade" id="Downgrade">
<p>
<em>Downgrade</em> restores the software back to the pre-upgrade release
and preserves the user data.
Suppose time <em>T</em> is the rolling upgrade start time and the upgrade is terminated by downgrade.
Then, the files created before or after <em>T</em> remain available in HDFS.
The files deleted before or after <em>T</em> remain deleted in HDFS.
</p>
<p>
A newer release is downgradable to the pre-upgrade release
only if both the namenode layout version and the datenode layout version
are not changed between these two releases.
Below are the steps for downgrade:
</p>
<ul>
<li>Downgrade HDFS<ol>
<li>Shutdown all <em>NNs</em> and <em>DNs</em>.</li>
<li>Restore the pre-upgrade release in all machines.</li>
<li>Start <em>NNs</em> with the
"<a href="#namenode_-rollingUpgrade"><code>-rollingUpgrade downgrade</code></a>" option.</li>
<li>Start <em>DNs</em> normally.</li>
</ol></li>
</ul>
</subsection>
<subsection name="Rollback" id="Rollback">
<p>
<em>Rollback</em> restores the software back to the pre-upgrade release
but also reverts the user data back to the pre-upgrade state.
Suppose time <em>T</em> is the rolling upgrade start time and the upgrade is terminated by rollback.
The files created before <em>T</em> remain available in HDFS but the files created after <em>T</em> become unavailable.
The files deleted before <em>T</em> remain deleted in HDFS but the files deleted after <em>T</em> are restored.
</p>
<p>
Rollback from a newer release to the pre-upgrade release is always supported.
Below are the steps for rollback:
</p>
<ul>
<li>Rollback HDFS<ol>
<li>Shutdown all <em>NNs</em> and <em>DNs</em>.</li>
<li>Restore the pre-upgrade release in all machines.</li>
<li>Start <em>NNs</em> with the
"<a href="#namenode_-rollingUpgrade"><code>-rollingUpgrade rollback</code></a>" option.</li>
<li>Start <em>DNs</em> normally.</li>
</ol></li>
</ul>
</subsection>
</section>
<section name="Commands and Startup Options for Rolling Upgrade" id="dfsadminCommands">
<subsection name="DFSAdmin Commands" id="dfsadminCommands">
<h4><code>dfsadmin -rollingUpgrade</code></h4>
<source>hdfs dfsadmin -rollingUpgrade &lt;query|start|finalize&gt;</source>
<p>
Execute a rolling upgrade action.
<ul><li>Options:<table>
<tr><td><code>query</code></td><td>Query the current rolling upgrade status.</td></tr>
<tr><td><code>prepare</code></td><td>Prepare a new rolling upgrade.</td></tr>
<tr><td><code>finalize</code></td><td>Finalize the current rolling upgrade.</td></tr>
</table></li></ul>
</p>
<h4><code>dfsadmin -getDatanodeInfo</code></h4>
<source>hdfs dfsadmin -getDatanodeInfo &lt;DATANODE_HOST:IPC_PORT&gt;</source>
<p>
Get the information about the given datanode.
This command can be used for checking if a datanode is alive
like the Unix <code>ping</code> command.
</p>
<h4><code>dfsadmin -shutdownDatanode</code></h4>
<source>hdfs dfsadmin -shutdownDatanode &lt;DATANODE_HOST:IPC_PORT&gt; [upgrade]</source>
<p>
Submit a shutdown request for the given datanode.
If the optional <code>upgrade</code> argument is specified,
clients accessing the datanode will be advised to wait for it to restart
and the fast start-up mode will be enabled.
When the restart does not happen in time, clients will timeout and ignore the datanode.
In such case, the fast start-up mode will also be disabled.
</p>
<p>
Note that the command does not wait for the datanode shutdown to complete.
The "<a href="#dfsadmin_-getDatanodeInfo">dfsadmin -getDatanodeInfo</a>"
command can be used for checking if the datanode shutdown is complete.
</p>
</subsection>
<subsection name="NameNode Startup Options" id="dfsadminCommands">
<h4><code>namenode -rollingUpgrade</code></h4>
<source>hdfs namenode -rollingUpgrade &lt;downgrade|rollback&gt;</source>
<p>
Downgrade or rollback an ongoing rolling upgrade.
</p>
<ul><li>Options:<table>
<tr><td><code>downgrade</code></td>
<td>Restores the namenode back to the pre-upgrade release
and preserves the user data.</td>
</tr>
<tr><td><code>rollback</code></td>
<td>Restores the namenode back to the pre-upgrade release
but also reverts the user data back to the pre-upgrade state.</td>
</tr>
</table></li></ul>
</subsection>
</section>
</body>
</document>

View File

@ -83,6 +83,7 @@
href="hadoop-project-dist/hadoop-hdfs/ShortCircuitLocalReads.html"/>
<item name="Centralized Cache Management" href="hadoop-project-dist/hadoop-hdfs/CentralizedCacheManagement.html"/>
<item name="HDFS NFS Gateway" href="hadoop-project-dist/hadoop-hdfs/HdfsNfsGateway.html"/>
<item name="HDFS Rolling Upgrade" href="hadoop-project-dist/hadoop-hdfs/HdfsRollingUpgrade.html"/>
</menu>
<menu name="MapReduce" inherit="top">