libdb/docs/gsg_db_rep/JAVA/repapp.html
2012-11-14 16:35:20 -05:00

683 lines
28 KiB
HTML
Raw Permalink Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<title>Chapter 3. The DB Replication Manager</title>
<link rel="stylesheet" href="gettingStarted.css" type="text/css" />
<meta name="generator" content="DocBook XSL Stylesheets V1.73.2" />
<link rel="start" href="index.html" title="Getting Started with Replicated Berkeley DB Applications" />
<link rel="up" href="index.html" title="Getting Started with Replicated Berkeley DB Applications" />
<link rel="prev" href="simpleprogramlisting.html" title="Program Listing" />
<link rel="next" href="repmgr_init_example_c.html" title="Adding the Replication Manager to SimpleTxn" />
</head>
<body>
<div xmlns="" class="navheader">
<div class="libver">
<p>Library Version 11.2.5.3</p>
</div>
<table width="100%" summary="Navigation header">
<tr>
<th colspan="3" align="center">Chapter 3. The DB Replication Manager</th>
</tr>
<tr>
<td width="20%" align="left"><a accesskey="p" href="simpleprogramlisting.html">Prev</a> </td>
<th width="60%" align="center"> </th>
<td width="20%" align="right"> <a accesskey="n" href="repmgr_init_example_c.html">Next</a></td>
</tr>
</table>
<hr />
</div>
<div class="chapter" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h2 class="title"><a id="repapp"></a>Chapter 3. The DB Replication Manager</h2>
</div>
</div>
</div>
<div class="toc">
<p>
<b>Table of Contents</b>
</p>
<dl>
<dt>
<span class="sect1">
<a href="repapp.html#rep_init_code">
Starting and Stopping Replication
</a>
</span>
</dt>
<dd>
<dl>
<dt>
<span class="sect2">
<a href="repapp.html#election_flags">Managing Election Policies</a>
</span>
</dt>
<dt>
<span class="sect2">
<a href="repapp.html#thread_count">Selecting the Number of Threads</a>
</span>
</dt>
</dl>
</dd>
<dt>
<span class="sect1">
<a href="repmgr_init_example_c.html">Adding the Replication Manager to
<span>SimpleTxn</span>
</a>
</span>
</dt>
<dt>
<span class="sect1">
<a href="fwrkpermmessage.html">Permanent Message Handling</a>
</span>
</dt>
<dd>
<dl>
<dt>
<span class="sect2">
<a href="fwrkpermmessage.html#fmwrkpermpolicy">Identifying Permanent Message Policies</a>
</span>
</dt>
<dt>
<span class="sect2">
<a href="fwrkpermmessage.html#fmwrkpermtimeout">Setting the Permanent Message Timeout</a>
</span>
</dt>
<dt>
<span class="sect2">
<a href="fwrkpermmessage.html#perm2fmwrkexample">Adding a Permanent Message Policy to
<span>RepQuoteExampleGSG</span>
</a>
</span>
</dt>
</dl>
</dd>
<dt>
<span class="sect1">
<a href="electiontimes.html">Managing Election Times</a>
</span>
</dt>
<dd>
<dl>
<dt>
<span class="sect2">
<a href="electiontimes.html#electiontimeout">Managing Election Timeouts</a>
</span>
</dt>
<dt>
<span class="sect2">
<a href="electiontimes.html#electretrytime">Managing Election Retry Times</a>
</span>
</dt>
</dl>
</dd>
<dt>
<span class="sect1">
<a href="fmwrkconnectretry.html">Managing Connection Retries</a>
</span>
</dt>
<dt>
<span class="sect1">
<a href="heartbeats.html">Managing Heartbeats</a>
</span>
</dt>
</dl>
</div>
<p>
The easiest way to add replication to your transactional
application is to use the Replication Manager. The Replication Manager provides a comprehensive
communications layer that enables replication. For a brief listing
of the Replication Manager's feature set, see
<a class="xref" href="apioverview.html#repframeworkoverview" title="Replication Manager Overview">Replication Manager Overview</a>.
</p>
<p>
To use the Replication Manager, you make use of special methods off the
<code class="classname">Environment</code> and
<code class="classname">EnvironmentConfig</code> classes, as well as
the <code class="classname">ReplicationManagerSiteConfig</code> class.
You also use a series of related classes to perform your
implementation. For example, in order to detect whether your
code is running as a master or a replica, you must implement
<code class="classname">com.sleepycat.db.EventHandler</code>.
(see <a class="xref" href="fwrkmasterreplica.html#determinestate" title="Determining State">Determining State</a>).
That is:
</p>
<div class="orderedlist">
<ol type="1">
<li>
<p>
Create an environment handle as normal.
</p>
</li>
<li>
<p>
Configure your environment handle as needed (e.g. set the
error file and error prefix values, if desired).
</p>
</li>
<li>
<p>
Use the Replication Manager replication classes and methods to configure
the Replication Manager. Using these classes and methods causes DB
to know that you are using the Replication Manager.
</p>
<p>
Configuring the Replication Manager entails setting the replication
environment's priority, setting the TCP/IP address that
this replication environment will use for incoming
replication messages, identifying TCP/IP addresses of other
replication environments, setting the number of replication
environments in the replication group, and so forth. These
actions are discussed throughout the remainder of this
chapter.
</p>
</li>
<li>
<p>
Open your environment handle. When you do this, be sure to
specify
<span>
<code class="methodname">EnvironmentConfig.setInitializeReplication()</code>
when you configure your environment handle. This is in
addition to the configuration that you would normally
use for a transactional application. This causes
replication to be initialized for the application.
</span>
</p>
</li>
<li>
<p>
Start replication by calling
<span><code class="methodname">Environment.replicationManagerStart()</code>.</span>
</p>
</li>
<li>
<p>
Open your databases as needed. Masters must open their databases
for read and write activity. Replicas can open their databases
for read-only activity, but doing so means they must
re-open the databases if the replica ever becomes a master.
Either way, replicas should never attempt to write to the
database(s) directly.
</p>
</li>
</ol>
</div>
<div class="note" style="margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>
The Replication Manager allows you to only use one environment handle per
process.
</p>
</div>
<p>
When you are ready to shut down your application:
</p>
<div class="orderedlist">
<ol type="1">
<li>
<p>
Close your databases
</p>
</li>
<li>
<p>
Close your environment. This causes replication to stop as
well.
</p>
</li>
</ol>
</div>
<div class="note" style="margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>
Before you can use the Replication Manager, you may have to enable it
in your DB library. This is
<span class="emphasis"><em>not</em></span> a requirement for Microsoft
Windows systems, or Unix systems that use pthread
mutexes by default. Other systems, notably BSD and
BSD-derived systems (such as Mac OS X), must enable the
Replication Manager when you configure the DB build.
</p>
<p>
You do this by <span class="emphasis"><em>not</em></span> disabling
replication and by configuring the library with POSIX
threads support. In other words, replication must be
turned on in the build (it is by default), and POSIX
thread support must be enabled if it is not already by
default. To do this, use the
<code class="literal">--enable-pthread_api</code> switch on the
configure script.
</p>
<p>
For example:
</p>
<pre class="programlisting">../dist/configure --enable-pthread-api</pre>
</div>
<div class="sect1" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a id="rep_init_code"></a>
Starting and Stopping Replication
</h2>
</div>
</div>
</div>
<div class="toc">
<dl>
<dt>
<span class="sect2">
<a href="repapp.html#election_flags">Managing Election Policies</a>
</span>
</dt>
<dt>
<span class="sect2">
<a href="repapp.html#thread_count">Selecting the Number of Threads</a>
</span>
</dt>
</dl>
</div>
<p>
As described above, you introduce replication to an application by
starting with a transactional application, performing some basic
replication configuration, and then starting replication using
<span><code class="methodname">Environment.replicationManagerStart()</code>.</span>
</p>
<p>
You stop replication by closing your environment
cleanly in the same way you would for any DB application.
</p>
<p>
For example, the following code fragment initializes, then
stops and starts replication. Note that other replication
activities are omitted for brevity.
</p>
<div class="note" style="margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>
Note that the following code fragment would be part of a larger
class that must implement
<code class="classname">com.sleepycat.db.EventHandler</code>. This
class is used to track state changes between master and
replica. We put off that implementation for the moment, but the
point remains that the following code fragment would be
contained in a method or two that you would include in your
<code class="classname">com.sleepycat.db.EventHandler</code>
implementation.
</p>
</div>
<pre class="programlisting">
import com.sleepycat.db.Environment;
import com.sleepycat.db.EnvironmentConfig;
import com.sleepycat.db.ReplicationHostAddress;
import com.sleepycat.db.ReplicationManagerStartPolicy;
import com.sleepycat.db.ReplicationManagerSiteConfig;
...
String progname = "example_replication";
String envHome = "TESTDIR";
int cachesize = 10 * 1024 * 1024;
Environment dbenv;
String listenHost = "mymachine.sleepycat.com";
String otherHost = "anothermachine.sleepycat.com";
int listenPort = 8080;
int otherPort = 8081;
try {
// Configure the environment handle
EnvironmentConfig envConfig = new EnvironmentConfig();
envConfig.setErrorStream(System.err);
envConfig.setErrorPrefix(progname);
envConfig.setCacheSize(cachesize);
envConfig.setTxnNoSync(true);
// Identify the local replication site. This is the local hostname
// and port that this replication participant will use to receive
// incoming replication messages. Note that this can be
// performed only once for the application. It is required.
ReplicationManagerSiteConfig thisHostConfig =
new ReplicationManagerSiteConfig(listenHost, listenPort);
thisHostConfig.setGroupCreator(creator);
thisHostConfig.setLocalSite(true)
envConfig.addReplicationManagerSite(thisHostConfig);
// Set this application's priority. This is used for elections.
//
// Set this number to a positive integer, or 0 if you do not want
// this site to be able to become a master.
envConfig.setReplicationPriority(100);
// Add a site to the list of replication environments known to
// this application.
ReplicationManagerSiteConfig remoteSiteConfig =
new ReplicationManagerSiteConfig(otherHost, otherPort);
remoteSiteConfig.setBootstrapHelper(true);
envConfig.addReplicationManagerSite(remoteSiteConfig);
// Configure the environment's subsystems. Note that we initialize
// replication. This is required.
envConfig.setAllowCreate(true);
envConfig.setRunRecovery(true);
envConfig.setThreaded(true);
envConfig.setInitializeReplication(true);
envConfig.setInitializeLocking(true);
envConfig.setInitializeLogging(true);
envConfig.setInitializeCache(true);
envConfig.setTransactional(true);
// Missing from this is where we set the event handle and the
// acknowledgement policy. We discuss these things later in this
// book.
// Open our environment handle.
try {
dbenv = new Environment(envHome, envConfig);
} catch(FileNotFoundException e) {
System.err.println("FileNotFound exception: " + e.toString());
System.err.println(
"Ensure that the environment directory is pre-created.");
}
// Start the replication manager such that it has three threads.
dbenv.replicationManagerStart(3,
ReplicationManagerStartPolicy.REP_ELECTION);
//////////////////////////////////////////////////
// All other application code goes here, including
// database opens.
//////////////////////////////////////////////////
} catch (DatabaseException dbe) {
// Error handling goes here
}
// Close out your application here.
try {
// Make sure all your databases are closed.
// Closing your environment stops replication.
dbenv.close();
} catch (DatabaseException dbe) {
// Error handling here.
}
// All done. </pre>
<div class="sect2" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a id="election_flags"></a>Managing Election Policies</h3>
</div>
</div>
</div>
<p>
Before continuing, it is worth taking a look at the
<span>
startup election options that you can set for
replication. You set these using the
<code class="classname">ReplicationManagerStartPolicy</code>
class that you pass to the
<code class="methodname">Environment.replicationManagerStart()</code>
method.
</span>
</p>
<p>
In the previous example, we specified
<code class="literal">ReplicationManagerStartPolicy.REP_ELECTION</code>
when we started replication. This causes the
application to try to find a master upon startup. If it
cannot, it calls for an election. In the event an
election is held, the environment receiving the most number of
votes will become the master.
</p>
<p>
There's some important points to make here:
</p>
<div class="itemizedlist">
<ul type="disc">
<li>
<p>
This
<span>option</span>
only requires that other
environments in the replication group
participate in the vote. There is no
requirement that
<span class="emphasis"><em>all</em></span> such
environments participate. In other
words, if an environment
starts up, it can call for an
election, and select a master, even
if all other environment have not yet
joined the replication group.
</p>
</li>
<li>
<p>
It only requires a simple majority of
participating environments to elect a master.
This is always true of elections held using the Replication Manager.
</p>
</li>
<li>
<p>
As always, the environment participating in the election with the most
up-to-date log files is selected as
master. If an environment with more recent log files
has not yet joined the replication
group, it may not become the master.
</p>
</li>
</ul>
</div>
<p>
Any one of these points may be enough to cause a
less-than-optimum environment to be selected as master.
Therefore, to give you a better degree of control over
which environment becomes a master at application startup,
the Replication Manager offers the following start-up
<span>options:</span>
</p>
<div class="informaltable">
<table border="1" width="80%">
<colgroup>
<col />
<col />
</colgroup>
<thead>
<tr>
<th>Option</th>
<th>Description</th>
</tr>
</thead>
<tbody>
<tr>
<td>
<code class="literal">ReplicationManagerStartPolicy. REP_MASTER</code>
</td>
<td>
<p>
The application starts up and declares the environment to be a master
without calling for an election. It is an error for more
than one environment to start up using this flag, or for
an environment
to use this flag when a master already exists.
</p>
<p>
Note that no replication group should
<span class="emphasis"><em>ever</em></span> operate with more than
one master.
</p>
<p>
In the event that a environment attempts to become a
master when a master already exists, the
replication code will resolve the problem by
holding an election. Note, however, that there
is always a possibility of data loss in the face
of duplicate masters, because once a master is
selected, the environment that loses the election will
have to roll back any transactions committed
until it is in sync with the "real" master.
</p>
</td>
</tr>
<tr>
<td>
<code class="literal">ReplicationManagerStartPolicy. REP_CLIENT</code>
</td>
<td>
<p>
The application starts up and declares
the environment to be a replica without calling for
an election. Note that the environment
can still become a master if a subsequent
application starts up, calls for an
election, and this environment is elected
master.
</p>
</td>
</tr>
<tr>
<td>
<code class="literal">ReplicationManagerStartPolicy. REP_ELECTION</code>
</td>
<td>
<p>
As described above, the application starts up,
looks for a master, and if one is not found calls
for an election.
</p>
</td>
</tr>
</tbody>
</table>
</div>
</div>
<div class="sect2" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a id="thread_count"></a>Selecting the Number of Threads</h3>
</div>
</div>
</div>
<p>
Under the hood, the Replication Manager is threaded and you can
control the number of threads used to process messages received from
other replicas. The threads that the Replication Manager uses are:
</p>
<div class="itemizedlist">
<ul type="disc">
<li>
<p>
Incoming message thread. This thread
receives messages from the site's
socket and passes those messages to
message processing threads (see below)
for handling.
</p>
</li>
<li>
<p>
Outgoing message thread. Outgoing
messages are sent from whatever thread
performed a write to the database(s).
That is, the thread that called, for
example,
<code class="methodname">Database.put()</code>
is the thread that writes replication messages
about that fact to the socket.
</p>
<p>
Note that if this write activity would
cause the thread to be blocked due to
some condition on the socket, the Replication Manager
will hand the outgoing message to the
incoming message thread, and it will
then write the message to the socket.
This prevents your database write
threads from blocking due to abnormal
network I/O conditions.
</p>
</li>
<li>
<p>
Message processing threads are
responsible for parsing and then
responding to incoming replication
messages. Typically, a response will
include write activity to your
database(s), so these threads can be
busy performing disk I/O.
</p>
</li>
</ul>
</div>
<p>
Of these threads, the only ones that you have any
configuration control over are the message processing
threads. In this case, you can determine how many
of these threads you want to run.
</p>
<p>
It is always a bit of an art to decide on a thread count,
but the short answer is you probably do not need more
than three threads here, and it is likely that one will
suffice. That said, the best thing to do is set your
thread count to a fairly low number and then increase
it if it appears that your application will benefit
from the additional threads.
</p>
</div>
</div>
</div>
<div class="navfooter">
<hr />
<table width="100%" summary="Navigation footer">
<tr>
<td width="40%" align="left"><a accesskey="p" href="simpleprogramlisting.html">Prev</a> </td>
<td width="20%" align="center"> </td>
<td width="40%" align="right"> <a accesskey="n" href="repmgr_init_example_c.html">Next</a></td>
</tr>
<tr>
<td width="40%" align="left" valign="top">Program Listing </td>
<td width="20%" align="center">
<a accesskey="h" href="index.html">Home</a>
</td>
<td width="40%" align="right" valign="top"> Adding the Replication Manager to
<span>SimpleTxn</span>
</td>
</tr>
</table>
</div>
</body>
</html>