We now use the methods described above to add partial support to the SimpleTxn example that we presented in Transactional Application. That is, in this section we will:
Enhance our command line options to accept information of interest to a replicated application.
Configure our environment handle to use replication and the Replication Manager.
Minimally configure the Replication Manager.
Start replication.
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).
Also, note that in the following code fragments, additions
and changes to the code are marked in bold
.
To begin, we make some significant changes to our
RepConfig
class because we will be
using it to maintain a lot more information that we needed
for our simple transactional example.
We begin by importing a few new
classes. java.util.Vector
is used to
organize a list of "other host" definitions (that is, the host and
port information for the other replication participants known to
this application). We also need a couple of classes used to manage
individual host and port information, as well as replication
startup policy information.
package db.repquote_gsg;
import java.util.Vector;
import com.sleepycat.db.ReplicationHostAddress;
import com.sleepycat.db.ReplicationManagerStartPolicy;
public class RepConfig
{
Next we add considerably to the constants and data members used by this class. All of this is used to manage information necessary for replication purposes. We also at this point change the program's name, since we will be doing that to the main class in our application a little later in this description.
// Constant values used in the RepQuote application. public static final String progname ="RepQuoteExampleGSG";
public static final int CACHESIZE = 10 * 1024 * 1024;public static final int SLEEPTIME = 5000;
// member variables containing configuration information // String specifying the home directory for rep files. public String home; // stores an optional set of "other" hosts.public Vector<ReplicationHostAddress> otherHosts; // priority within the replication group. public int priority; public ReplicationManagerStartPolicy startPolicy; // The host address to listen to. public ReplicationHostAddress thisHost; // Optional parameter specifying the # of sites in the // replication group. public int totalSites; // member variables used internally. private int currOtherHost; private boolean gotListenAddress;
Now we update our class constructor to initialize all of these new variables:
public RepConfig() {startPolicy = ReplicationManagerStartPolicy.REP_ELECTION;
home = "TESTDIR";gotListenAddress = false; totalSites = 0; priority = 100; currOtherHost = 0; thisHost = new ReplicationHostAddress(); otherHosts = new Vector()<ReplicationHostAddress>;
}
Finally, we finish updating this class by providing a series of new getter and setter methods. These are used primarily for setting a retrieving host information of interest to our replicated application:
public java.io.File getHome()
{
return new java.io.File(home);
}
public void setThisHost(String host, int port)
{
gotListenAddress = true;
thisHost.port = port;
thisHost.host = host;
}
public ReplicationHostAddress getThisHost()
{
if (!gotListenAddress) {
System.err.println("Warning: no host specified.");
System.err.println("Returning default.");
}
return thisHost;
}
public boolean gotListenAddress() {
return gotListenAddress;
}
public void addOtherHost(String host, int port, boolean peer)
{
ReplicationHostAddress newInfo =
new ReplicationHostAddress(host, port, peer);
otherHosts.add(newInfo);
}
public ReplicationHostAddress getFirstOtherHost()
{
currOtherHost = 0;
if (otherHosts.size() == 0)
return null;
return (ReplicationHostAddress)otherHosts.get(currOtherHost);
}
public ReplicationHostAddress getNextOtherHost()
{
currOtherHost++;
if (currOtherHost >= otherHosts.size())
return null;
return (ReplicationHostAddress)otherHosts.get(currOtherHost);
}
public ReplicationHostAddress getOtherHost(int i)
{
if (i >= otherHosts.size())
return null;
return (ReplicationHostAddress)otherHosts.get(i);
}
}
Having completed our update to the
RepConfig
class, we can now start making
changes to the main portion of our program. We begin by changing
the program's name. (This, of course, means that we copy our
SimpleTxn
code to a file named RepQuoteExampleGSG.java
.)
package db.repquote_gsg; import java.io.FileNotFoundException; import java.io.BufferedReader; import java.io.InputStreamReader; import java.io.IOException; import java.io.UnsupportedEncodingException;import java.lang.Thread; import java.lang.InterruptedException;
import com.sleepycat.db.Cursor; import com.sleepycat.db.Database; import com.sleepycat.db.DatabaseConfig; import com.sleepycat.db.DatabaseEntry; import com.sleepycat.db.DatabaseException; import com.sleepycat.db.DatabaseType; import com.sleepycat.db.Environment; import com.sleepycat.db.EnvironmentConfig; import com.sleepycat.db.LockMode; import com.sleepycat.db.OperationStatus; import db.repquote.RepConfig; public classRepQuoteExampleGSG
{ private RepConfig repConfig; private Environment dbenv;
Next we update our usage function. The application will continue to
accept the -h
parameter so that we can identify
the environment home directory used by this application. However,
we also add the
-l
parameter which allows us to
identify the host and port used by this application to
listen for replication messages.
-r
parameter which allows us to
specify other replicas.
-n
parameter which allows us to
identify the number of sites in this replication
group.
-p
option, which is used to identify
this replica's priority (recall that the priority is
used as a tie breaker for elections)
publicRepQuoteExampleGSG()
throws DatabaseException { repConfig = null; dbenv = null; } public static void usage() { System.err.println("usage: " + repConfig.progname); System.err.println("-h home[-r host:port][-l host:port]" + "[-n nsites][-p priority]
"); System.err.println("\t -l host:port (required; l stands for local)\n" + "\t -r host:port (optional; r stands for replica; any " + "number of these may be specified)\n" +
"\t -h home directory\n" +"\t -n nsites (optional; number of sites in replication " + "group; defaults to 0\n" + "\t in which case we try to dynamically compute the " + "number of sites in\n" + "\t the replication group)\n" + "\t -p priority (optional: defaults to 100)\n");
System.exit(1); }
Now we can begin working on our main()
function.
We begin by adding a couple of variables that we will use to
collect TCP/IP host and port information.
public static void main(String[] argv)
throws Exception
{
RepConfig config = new RepConfig();
String tmpHost;
int tmpPort = 0;
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.
// Extract the command line parameters for (int i = 0; i < argv.length; i++) { if (argv[i].compareTo("-h") == 0) { // home - a string arg. i++; config.home = argv[i];} else if (argv[i].compareTo("-l") == 0) { // "me" should be host:port i++; String[] words = argv[i].split(":"); if (words.length != 2) { System.err.println( "Invalid host specification host:port needed."); usage(); } try { tmpPort = Integer.parseInt(words[1]); } catch (NumberFormatException nfe) { System.err.println("Invalid host specification, " + "could not parse port number."); usage(); } config.setThisHost(words[0], tmpPort); } else if (argv[i].compareTo("-n") == 0) { i++; config.totalSites = Integer.parseInt(argv[i]); } else if (argv[i].compareTo("-p") == 0) { i++; config.priority = Integer.parseInt(argv[i]); } else if (argv[i].compareTo("-r") == 0) { i++; String[] words = argv[i].split(":"); if (words.length != 2) { System.err.println( "Invalid host specification host:port needed."); usage(); } try { tmpPort = Integer.parseInt(words[1]); } catch (NumberFormatException nfe) { System.err.println("Invalid host specification, " + "could not parse port number."); usage(); } config.addOtherHost(words[0], tmpPort, isPeer); }
else { System.err.println("Unrecognized option: " + argv[i]); usage(); } } // Error check command line. if ((!config.gotListenAddress())
|| config.home.length() == 0) usage();
Having done that, the remainder of our main()
function is left unchanged, with the exception of a few name changes required by the
new class name:
RepQuoteExampleGSG
runner = null; try { runner = newRepQuoteExampleGSG();
runner.init(config); runner.doloop(); runner.terminate(); } catch (DatabaseException dbe) { System.err.println("Caught an exception during " + "initialization or processing: " + dbe.toString()); if (runner != null) runner.terminate(); } System.exit(0); } // end main
Now we need to update our
RepQuoteExampleGSG.init()
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
-l
command line option):
public int init(RepConfig config)
throws DatabaseException
{
int ret = 0;
appConfig = config;
EnvironmentConfig envConfig = new EnvironmentConfig();
envConfig.setErrorStream(System.err);
envConfig.setErrorPrefix(RepConfig.progname);
envConfig.setReplicationManagerLocalSite(appConfig.getThisHost());
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 -o
command line
option). To do this, we iterate over each of the "other" sites provided to
us using the -o
command line option, and we add each one
individually in turn:
for (ReplicationHostAddress host = appConfig.getFirstOtherHost();
host != null; host = appConfig.getNextOtherHost())
{
envConfig.replicationManagerAddRemoteSite(host);
}
And then we need code to allow us to identify the total number of sites in this replication group, and to set the environment's priority.
if (appConfig.totalSites > 0)
envConfig.setReplicationNumSites(appConfig.totalSites);
envConfig.setReplicationPriority(appConfig.priority);
We can now open our environment. Note that the options
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
EnvironmentConfig.setInitializeReplication()
option.
Also, because we are using the Replication Manager, we must prepare
our environment for threaded usage. For this reason, we also
need the DB_THREAD
flag.
envConfig.setCacheSize(RepConfig.CACHESIZE);
envConfig.setTxnNoSync(true);
envConfig.setAllowCreate(true);
envConfig.setRunRecovery(true);
envConfig.setInitializeReplication(true);
envConfig.setInitializeLocking(true);
envConfig.setInitializeLogging(true);
envConfig.setInitializeCache(true);
envConfig.setTransactional(true);
try {
dbenv = new Environment(appConfig.getHome(), envConfig);
} catch(FileNotFoundException e) {
System.err.println("FileNotFound exception: " + e.toString());
System.err.println(
"Ensure that the environment directory is pre-created.");
ret = 1;
}
Finally, we start replication before we exit this method.
Immediately after exiting this method, our application will go into
the
RepQuoteExampleGSG.doloop()
method, which is where
the bulk of our application's work is performed. We update that
method in the next chapter.
// start Replication Manager dbenv.replicationManagerStart(3, appConfig.startPolicy); return ret; }
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.