greg/libdb
1
0
Fork 0
mirror of https://github.com/berkeleydb/libdb.git synced 2024-11-17 01:26:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
libdb/docs/gsg_db_rep/CXX/repmgr_init_example_c.html

514 lines
18 KiB
HTML
Raw Normal View History

Release 5.2.28 on 6/10/2011
2011-09-13 17:44:24 +00:00
<?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>Adding the Replication Manager to RepMgr</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="repapp.html" title="Chapter 3. The DB Replication Manager" />
<link rel="prev" href="rep_init_code.html" title="Starting and Stopping Replication" />
<link rel="next" href="fwrkpermmessage.html" title="Permanent Message Handling" />
</head>
<body>
<div xmlns="" class="navheader">
<div class="libver">
<p>Library Version 11.2.5.2</p>
</div>
<table width="100%" summary="Navigation header">
<tr>
<th colspan="3" align="center">Adding the Replication Manager to
<span xmlns="http://www.w3.org/1999/xhtml">RepMgr</span>
</th>
</tr>
<tr>
<td width="20%" align="left"><a accesskey="p" href="rep_init_code.html">Prev</a> </td>
<th width="60%" align="center">Chapter 3. The DB Replication Manager</th>
<td width="20%" align="right"> <a accesskey="n" href="fwrkpermmessage.html">Next</a></td>
</tr>
</table>
<hr />
</div>
<div class="sect1" lang="en" xml:lang="en">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a id="repmgr_init_example_c"></a>Adding the Replication Manager to
<span>RepMgr</span>
</h2>
</div>
</div>
</div>
<p>
We now use the methods described above to add partial
support to the
<span>RepMgr</span>
example that we presented in
<a class="xref" href="txnapp.html" title="Chapter 2. Transactional Application">Transactional Application</a>.
That is, in this section we will:
</p>
<div class="itemizedlist">
<ul type="disc">
<li>
<p>
Enhance our command line options to accept information
of interest to a replicated application.
</p>
</li>
<li>
<p>
Configure our environment handle to use replication and
the Replication Manager.
</p>
</li>
<li>
<p>
Minimally configure the Replication Manager.
</p>
</li>
<li>
<p>
Start replication.
</p>
</li>
</ul>
</div>
<p>
Note that when we are done with this section, we will be
only partially ready to run the application. Some critical
pieces will be missing; specifically, we will not yet be
handling the differences between a master and a
replica. (We do that in the next chapter).
</p>
<p>
Also, note that in the following code fragments, additions
and changes to the code are marked in <strong class="userinput"><code>bold</code></strong>.
</p>
<p>
To begin, we copy the
<span>SimpleTxn</span>
code to a new file called
<code class="literal">RepMgrGSG.cpp</code>.
<span>
Having done that, we must make some significant
changes to our
<code class="classname">RepConfigInfo</code>
class because now we will be using it to
maintain a lot more information.
</span>
</p>
<p>
First, we create a new structure,
<code class="literal">RepHostInfoObj</code>, which we use to store
host and port information for all "other" servers
identified to the application via the
<code class="literal">-o</code> command line option. This structure
is chain-able, which makes cleaning up at program shutdown
time easier.
</p>
<pre class="programlisting">#include &lt;db_cxx.h&gt;
#include &lt;iostream&gt;
<strong class="userinput"><code>// Chain-able struct used to store host information.
typedef struct RepHostInfoObj{
char* host;
u_int16_t port;
RepHostInfoObj* next; // used for chaining multiple "other" hosts.
bool creator; // whether this site is the group creator.
} REP_HOST_INFO; </code></strong></pre>
<p>
Next, we update our <code class="classname">RepConfigInfo</code> class
definition to manage a lot more information and a new method.
</p>
<pre class="programlisting">class RepConfigInfo {
public:
RepConfigInfo();
virtual ~RepConfigInfo();
<strong class="userinput"><code>void addOtherHost(char* host, int port);</code></strong>
public:
<strong class="userinput"><code>u_int32_t start_policy;</code></strong>
char* home;
<strong class="userinput"><code>bool got_listen_address;
REP_HOST_INFO this_host;
int nrsites; // number of remote sites
int priority;
// used to store a set of optional other hosts.
REP_HOST_INFO *other_hosts;</code></strong>
}; </pre>
<p>
Then, we update our constructor to initialize our new variables.
</p>
<pre class="programlisting">RepConfigInfo::RepConfigInfo()
{
<strong class="userinput"><code>start_policy = DB_REP_ELECTION;</code></strong>
home = "TESTDIR";
<strong class="userinput"><code>got_listen_address = false;
nrsites = 0;
priority = 100;
other_hosts = NULL;</code></strong>
} </pre>
<p>
Next, we implement our new method, <code class="methodname">RepConfigInfo::addOtherHost</code>,
which is used to create <code class="literal">RepHostInfoObj</code> instances and add them to
the chain of "other" hosts.
</p>
<pre class="programlisting">RepConfigInfo::addOtherHost(char* host, int port)
{
REP_HOST_INFO *newinfo;
newinfo = (REP_HOST_INFO*)malloc(sizeof(REP_HOST_INFO));
newinfo-&gt;host = host;
newinfo-&gt;port = port;
if (other_hosts == NULL) {
other_hosts = newinfo;
newinfo-&gt;next = NULL;
} else {
newinfo-&gt;next = other_hosts;
other_hosts = newinfo;
}
nrsites++;
}</pre>
<p>
Having done that, we update our class destructor to release the <code class="literal">RepHostInfoObj</code>
chain of objects at class destruction time.
</p>
<pre class="programlisting">RepConfigInfo::~RepConfigInfo()
{
<strong class="userinput"><code>// release any other_hosts structs.
if (other_hosts != NULL) {
REP_HOST_INFO *CurItem = other_hosts;
while (CurItem-&gt;next != NULL)
{
REP_HOST_INFO *TmpItem = CurItem;
free(CurItem);
CurItem = TmpItem;
}
free(CurItem);
}
other_hosts = NULL;</code></strong>
} </pre>
<p>
Having completed our update to the
<code class="classname">RepConfigInfo</code>
class, we can now start making
changes to the main portion of our program. We begin by changing
the program's name.
</p>
<pre class="programlisting">using std::cout;
using std::cin;
using std::cerr;
using std::endl;
using std::flush;
#define CACHESIZE (10 * 1024 * 1024)
#define DATABASE "quote.db"
<strong class="userinput"><code>const char *progname = "RepMgrGSG";</code></strong> </pre>
<p>
Next we update our usage function. The application will continue to
accept the <code class="literal">-h</code> parameter so that we can identify
the environment home directory used by this application. However,
we also add the:
</p>
<div class="itemizedlist">
<ul type="disc">
<li>
<p>
<code class="literal">-l</code> parameter which allows us to identify the
host and port used by this application to listen for
replication messages. This parameter is required unless the -L
parameter is specified.
</p>
</li>
<li>
<p>
<code class="literal">-L</code> parameter, which allows us to identify
the local site as the group creator.
</p>
</li>
<li>
<p>
<code class="literal">-r</code> parameter which allows us to specify
other replicas.
</p>
</li>
<li>
<p>
<code class="literal">-p</code> option, which is used to identify this
replica's priority (recall that the priority is used as a tie
breaker for elections)
</p>
</li>
</ul>
</div>
<pre class="programlisting">class RepMgrGSG
{
public:
// Constructor.
RepMgrGSG();
// Initialization method. Creates and opens our environment handle.
int init(RepConfigInfo* config);
// The doloop is where all the work is performed.
int doloop();
// terminate() provides our shutdown code.
int terminate();
private:
// disable copy constructor.
RepMgrGSG(const RepMgrGSG &amp;);
void operator = (const RepMgrGSG &amp;);
// internal data members.
RepConfigInfo *app_config;
DbEnv dbenv;
// private methods.
// print_stocks() is used to display the contents of our database.
static int print_stocks(Db *dbp);
};
static void usage()
{
cerr &lt;&lt; "usage: " &lt;&lt; progname &lt;&lt; endl
&lt;&lt; "-h home<strong class="userinput"><code>-l|L host:port [-r host:port]</code></strong>"
<strong class="userinput"><code>&lt;&lt; "[-p priority]" &lt;&lt; endl;</code></strong>
cerr &lt;&lt; "\t -h home directory" &lt;&lt; endl
<strong class="userinput"><code>&lt;&lt; "\t -l host:port (required unless -L is specified;"
&lt;&lt; "\t l stands for local)" &lt;&lt; endl
&lt;&lt; "\t -L host:port (optional, L means group "
&lt;&lt; " creator)" &lt;&lt; endl
&lt;&lt; "\t -r host:port (optional; r stands for replica; any "
&lt;&lt; "number of these may be specified)" &lt;&lt; endl
&lt;&lt; "\t -p priority (optional: defaults to 100)" &lt;&lt; endl;</code></strong>
exit(EXIT_FAILURE);
} </pre>
<p>
Now we can begin working on our <code class="literal">main()</code> function.
We begin by adding a couple of variables that we will use to
collect TCP/IP host and port information.
</p>
<pre class="programlisting">int main(int argc, char **argv)
{
RepConfigInfo config;
char ch<strong class="userinput"><code>, *portstr, *tmphost;
int tmpport;</code></strong>
int ret; </pre>
<p>
Now we collect our command line arguments. As we do so, we will
configure host and port information as required, and we will
configure the application's election priority if necessary.
</p>
<pre class="programlisting"> // Extract the command line parameters
while ((ch = getopt(argc, argv, "h:<strong class="userinput"><code>l:n:p:r:</code></strong>")) != EOF) {
switch (ch) {
case 'h':
config.home = optarg;
break;
<strong class="userinput"><code>case 'L':
config.this_host.creator = true; // FALLTHROUGH
case 'l':
config.this_host.host = strtok(optarg, ":");
if ((portstr = strtok(NULL, ":")) == NULL) {
cerr &lt;&lt; "Bad host specification." &lt;&lt; endl;
usage();
}
config.this_host.port = (unsigned short)atoi(portstr);
config.got_listen_address = true;
break;
case 'p':
config.priority = atoi(optarg);
break;
case 'r':
tmphost = strtok(optarg, ":");
if ((portstr = strtok(NULL, ":")) == NULL) {
cerr &lt;&lt; "Bad host specification." &lt;&lt; endl;
usage();
}
tmpport = (unsigned short)atoi(portstr);
config.addOtherHost(tmphost, tmpport);
break;</code></strong>
case '?':
default:
usage();
}
}
// Error check command line.
if ((!config.got_listen_address) || config.home == NULL)
usage(); </pre>
<p>
Having done that, the remainder of our <code class="function">main()</code>
function is left unchanged:
</p>
<pre class="programlisting"> RepMgrGSG runner;
try {
if((ret = runner.init(&amp;config)) != 0)
goto err;
if((ret = runner.doloop()) != 0)
goto err;
} catch (DbException dbe) {
cerr &lt;&lt; "Caught an exception during initialization or"
&lt;&lt; " processing: " &lt;&lt; dbe.what() &lt;&lt; endl;
}
err:
runner.terminate();
return 0;
} </pre>
<p>
Now we need to update our
<code class="methodname">RepMgrGSG::init()</code>
method. Our updates are at first related to configuring
replication. First, we need to update the method so that we can
identify the local site to the environment handle (that is, the site identified by the
<code class="literal">-l</code> command line option):
</p>
<pre class="programlisting">RepMgrGSG::RepMgrGSG() : app_config(0), dbenv(0)
{
}
int RepMgrGSG::init(RepConfigInfo *config)
{
int ret = 0;
app_config = config;
dbenv.set_errfile(stderr);
dbenv.set_errpfx(progname);
<strong class="userinput"><code>DbSite *dbsite;
dbenv.repmgr_site(app_config-&gt;this_host.host,
app_config-&gt;this_host.port, &amp;dbsite, 0);
dbsite-&gt;set_config(DB_LOCAL_SITE, 1);
if (app_config-&gt;this_host.creator)
dbsite-&gt;set_config(DB_GROUP_CREATOR, 1);
dbsite-&gt;close();</code></strong> </pre>
<p>
And we also add code to allow us to identify "other" sites to the
environment handle (that is, the sites that we identify using the
<code class="literal">-o</code> command line option). To do this, we iterate over
each of the "other" sites provided to us using the
<code class="literal">-o</code> command line option, and we add each one
individually in turn:
</p>
<p>
We also add code here to set the environment's priority.
</p>
<pre class="programlisting"> <strong class="userinput"><code>int i = 1;
for ( REP_HOST_INFO *cur = app_config-&gt;other_hosts;
cur != NULL &amp;&amp; i &lt;= app_config-&gt;nrsites;
cur = cur-&gt;next, i++) {
dbenv.repmgr_site(cur-&gt;host, cur-&gt;port, &amp;dbsite, 0);
dbsite-&gt;set_config(DB_BOOTSTRAP_HELPER, 1);
dbsite-&gt;close();
}
dbenv.rep_set_priority(app_config-&gt;priority); </code></strong> </pre>
<p>
<span>We can now open our environment. Note that the flags</span>
we use to open the environment are slightly different for a
replicated application than they are for a non-replicated
application. Namely, replication requires the
<span>
<code class="literal">DB_INIT_REP</code> flag.
</span>
</p>
<p>
Also, because we are using the Replication Manager, we must prepare
our environment for threaded usage. For this reason, we also
need the <code class="literal">DB_THREAD</code> flag.
</p>
<pre class="programlisting"> dbenv.set_cachesize(0, CACHESIZE, 0);
dbenv.set_flags(DB_TXN_NOSYNC, 1);
try {
dbenv.open(app_config-&gt;home,
DB_CREATE |
DB_INIT_LOCK |
DB_INIT_LOG |
DB_INIT_MPOOL |
<strong class="userinput"><code>DB_INIT_REP |</code></strong>
DB_INIT_TXN |
DB_RECOVER <strong class="userinput"><code>|
DB_THREAD;</code></strong>
0);
} catch(DbException dbe) {
cerr &lt;&lt; "Caught an exception during DB environment open." &lt;&lt; endl
&lt;&lt; "Ensure that the home directory is created prior to "
&lt;&lt; "starting the application." &lt;&lt; endl;
ret = ENOENT;
goto err;
}</pre>
<p>
Finally, we start replication before we exit this method.
Immediately after exiting this method, our application will go into
the
<code class="methodname">RepMgrGSG::doloop()</code>
method, which is where
the bulk of our application's work is performed. We update that
method in the next chapter.
</p>
<pre class="programlisting"> if ((ret = dbenv.repmgr_start(3, app_config-&gt;start_policy)) != 0)
goto err;
err:
return ret;
} </pre>
<p>
This completes our replication updates for the moment. We are not as
yet ready to actually run this program; there remains a few
critical pieces left to add to it. However, the work that we
performed in this section represents a solid foundation for the
remainder of our replication work.
</p>
</div>
<div class="navfooter">
<hr />
<table width="100%" summary="Navigation footer">
<tr>
<td width="40%" align="left"><a accesskey="p" href="rep_init_code.html">Prev</a> </td>
<td width="20%" align="center">
<a accesskey="u" href="repapp.html">Up</a>
</td>
<td width="40%" align="right"> <a accesskey="n" href="fwrkpermmessage.html">Next</a></td>
</tr>
<tr>
<td width="40%" align="left" valign="top">
Starting and Stopping Replication
 </td>
<td width="20%" align="center">
<a accesskey="h" href="index.html">Home</a>
</td>
<td width="40%" align="right" valign="top"> Permanent Message Handling</td>
</tr>
</table>
</div>
</body>
</html>
Reference in a new issue Copy permalink