Author: stack
Date: Sun Feb 6 05:43:52 2011
New Revision: 1067594
URL: http://svn.apache.org/viewvc?rev=1067594&view=rev
Log:
HBASE-3500 Documentation update for replication
Modified:
hbase/branches/0.90/CHANGES.txt
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java
hbase/branches/0.90/src/main/javadoc/org/apache/hadoop/hbase/replication/package.html
hbase/branches/0.90/src/site/xdoc/replication.xml
Modified: hbase/branches/0.90/CHANGES.txt
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/CHANGES.txt?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
--- hbase/branches/0.90/CHANGES.txt (original)
+++ hbase/branches/0.90/CHANGES.txt Sun Feb 6 05:43:52 2011
@@ -20,6 +20,7 @@ Release 0.90.1 - Unreleased
HBASE-3502 Can't open region because can't open .regioninfo because
AlreadyBeingCreatedException
HBASE-3501 Remove the deletion limit in LogCleaner
+ HBASE-3500 Documentation update for replication
IMPROVEMENTS
Modified:
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
---
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java
(original)
+++
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/HMaster.java
Sun Feb 6 05:43:52 2011
@@ -22,7 +22,6 @@ package org.apache.hadoop.hbase.master;
import java.io.IOException;
import java.lang.reflect.Constructor;
import java.lang.reflect.InvocationTargetException;
-import java.net.InetSocketAddress;
import java.net.UnknownHostException;
import java.util.ArrayList;
import java.util.Arrays;
@@ -55,8 +54,8 @@ import org.apache.hadoop.hbase.catalog.M
import org.apache.hadoop.hbase.client.HConnection;
import org.apache.hadoop.hbase.client.HConnectionManager;
import org.apache.hadoop.hbase.client.MetaScanner;
-import org.apache.hadoop.hbase.client.Result;
import org.apache.hadoop.hbase.client.MetaScanner.MetaScannerVisitor;
+import org.apache.hadoop.hbase.client.Result;
import org.apache.hadoop.hbase.executor.ExecutorService;
import org.apache.hadoop.hbase.executor.ExecutorService.ExecutorType;
import org.apache.hadoop.hbase.ipc.HBaseRPC;
@@ -599,23 +598,10 @@ implements HMasterInterface, HMasterRegi
public MapWritable regionServerStartup(final HServerInfo serverInfo,
final long serverCurrentTime)
throws IOException {
- // Set the ip into the passed in serverInfo. Its ip is more than likely
- // not the ip that the master sees here. See at end of this method where
- // we pass it back to the regionserver by setting
"hbase.regionserver.address"
- // Everafter, the HSI combination 'server name' is what uniquely identifies
- // the incoming RegionServer.
- InetSocketAddress address = new InetSocketAddress(
- HBaseServer.getRemoteIp().getHostName(),
- serverInfo.getServerAddress().getPort());
- serverInfo.setServerAddress(new HServerAddress(address));
-
// Register with server manager
this.serverManager.regionServerStartup(serverInfo, serverCurrentTime);
// Send back some config info
- MapWritable mw = createConfigurationSubset();
- mw.put(new Text("hbase.regionserver.address"),
- serverInfo.getServerAddress());
- return mw;
+ return createConfigurationSubset();
}
/**
Modified:
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
---
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
(original)
+++
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/master/ServerManager.java
Sun Feb 6 05:43:52 2011
@@ -123,11 +123,10 @@ public class ServerManager {
// is, reject the server and trigger its expiration. The next time it comes
// in, it should have been removed from serverAddressToServerInfo and
queued
// for processing by ProcessServerShutdown.
- HServerInfo info = new HServerInfo(serverInfo);
- checkIsDead(info.getServerName(), "STARTUP");
- checkAlreadySameHostPort(info);
- checkClockSkew(info, serverCurrentTime);
- recordNewServer(info, false, null);
+ checkIsDead(serverInfo.getServerName(), "STARTUP");
+ checkAlreadySameHostPort(serverInfo);
+ checkClockSkew(serverInfo, serverCurrentTime);
+ recordNewServer(serverInfo, false, null);
}
/**
@@ -240,24 +239,21 @@ public class ServerManager {
HMsg [] regionServerReport(final HServerInfo serverInfo,
final HMsg [] msgs, final HRegionInfo[] mostLoadedRegions)
throws IOException {
- // Be careful. This method does returns in the middle.
- HServerInfo info = new HServerInfo(serverInfo);
-
// Check if dead. If it is, it'll get a 'You Are Dead!' exception.
- checkIsDead(info.getServerName(), "REPORT");
+ checkIsDead(serverInfo.getServerName(), "REPORT");
// If we don't know this server, tell it shutdown.
- HServerInfo storedInfo = this.onlineServers.get(info.getServerName());
+ HServerInfo storedInfo =
this.onlineServers.get(serverInfo.getServerName());
if (storedInfo == null) {
// Maybe we already have this host+port combo and its just different
// start code?
- checkAlreadySameHostPort(info);
+ checkAlreadySameHostPort(serverInfo);
// Just let the server in. Presume master joining a running cluster.
// recordNewServer is what happens at the end of reportServerStartup.
// The only thing we are skipping is passing back to the regionserver
// the HServerInfo to use. Here we presume a master has already done
// that so we'll press on with whatever it gave us for HSI.
- recordNewServer(info, true, null);
+ recordNewServer(serverInfo, true, null);
// If msgs, put off their processing but this is not enough because
// its possible that the next time the server reports in, we'll still
// not be up and serving. For example, if a split, we'll need the
@@ -270,7 +266,7 @@ public class ServerManager {
}
// Check startcodes
- if (raceThatShouldNotHappenAnymore(storedInfo, info)) {
+ if (raceThatShouldNotHappenAnymore(storedInfo, serverInfo)) {
return HMsg.STOP_REGIONSERVER_ARRAY;
}
@@ -299,7 +295,7 @@ public class ServerManager {
reply = HMsg.STOP_REGIONSERVER_ARRAY;
}
}
- return processRegionServerAllsWell(info, mostLoadedRegions, reply);
+ return processRegionServerAllsWell(serverInfo, mostLoadedRegions, reply);
}
private boolean raceThatShouldNotHappenAnymore(final HServerInfo storedInfo,
Modified:
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
---
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java
(original)
+++
hbase/branches/0.90/src/main/java/org/apache/hadoop/hbase/regionserver/HRegionServer.java
Sun Feb 6 05:43:52 2011
@@ -50,7 +50,6 @@ import java.util.concurrent.atomic.Atomi
import java.util.concurrent.atomic.AtomicInteger;
import java.util.concurrent.locks.ReentrantReadWriteLock;
-import com.google.common.collect.Lists;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
@@ -76,7 +75,6 @@ import org.apache.hadoop.hbase.Stoppable
import org.apache.hadoop.hbase.UnknownRowLockException;
import org.apache.hadoop.hbase.UnknownScannerException;
import org.apache.hadoop.hbase.YouAreDeadException;
-import org.apache.hadoop.hbase.HConstants.OperationStatusCode;
import org.apache.hadoop.hbase.catalog.CatalogTracker;
import org.apache.hadoop.hbase.catalog.MetaEditor;
import org.apache.hadoop.hbase.catalog.RootLocationEditor;
@@ -136,6 +134,7 @@ import org.apache.hadoop.net.DNS;
import org.apache.zookeeper.KeeperException;
import com.google.common.base.Function;
+import com.google.common.collect.Lists;
/**
* HRegionServer makes a set of HRegions available to clients. It checks in
with
@@ -839,16 +838,7 @@ public class HRegionServer implements HR
protected void handleReportForDutyResponse(final MapWritable c) throws
IOException {
try {
for (Map.Entry<Writable, Writable> e : c.entrySet()) {
-
String key = e.getKey().toString();
- // Use the address the master passed us
- if (key.equals("hbase.regionserver.address")) {
- HServerAddress hsa = (HServerAddress) e.getValue();
- LOG.info("Master passed us address to use. Was="
- + this.serverInfo.getServerAddress() + ", Now=" + hsa.toString());
- this.serverInfo.setServerAddress(hsa);
- continue;
- }
String value = e.getValue().toString();
if (LOG.isDebugEnabled()) {
LOG.debug("Config from master: " + key + "=" + value);
Modified:
hbase/branches/0.90/src/main/javadoc/org/apache/hadoop/hbase/replication/package.html
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/src/main/javadoc/org/apache/hadoop/hbase/replication/package.html?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
---
hbase/branches/0.90/src/main/javadoc/org/apache/hadoop/hbase/replication/package.html
(original)
+++
hbase/branches/0.90/src/main/javadoc/org/apache/hadoop/hbase/replication/package.html
Sun Feb 6 05:43:52 2011
@@ -92,6 +92,7 @@ to another.
<name>hbase.replication</name>
<value>true</value>
</property></pre>
+ deploy the files, and then restart HBase if it was running.
</li>
<li>Run the following command in the master's shell while it's running
<pre>add_peer</pre>
@@ -100,6 +101,18 @@ to another.
to use a different <b>zookeeper.znode.parent</b> since they can't
write in the same folder.
</li>
+ <li>
+ Once you have a peer, you need to enable replication on your column
families.
+ One way to do it is to alter the table and to set the scope like this:
+ <pre>
+ disable 'your_table'
+ alter 'your_table', {NAME => 'family_name', REPLICATION_SCOPE => '1'}
+ enable 'your_table'
+ </pre>
+ Currently, a scope of 0 (default) means that it won't be replicated and a
+ scope of 1 means it's going to be. In the future, different scope can be
+ used for routing policies.
+ </li>
</ol>
You can confirm that your setup works by looking at any region server's log
Modified: hbase/branches/0.90/src/site/xdoc/replication.xml
URL:
http://svn.apache.org/viewvc/hbase/branches/0.90/src/site/xdoc/replication.xml?rev=1067594&r1=1067593&r2=1067594&view=diff
==============================================================================
--- hbase/branches/0.90/src/site/xdoc/replication.xml (original)
+++ hbase/branches/0.90/src/site/xdoc/replication.xml Sun Feb 6 05:43:52 2011
@@ -43,7 +43,10 @@
other well known solutions like MySQL master/slave replication where
thereâs only one bin log to keep track of. One master cluster can
replicate to any number of slave clusters, and each region server will
- participate to replicate their own stream of edits.
+ participate to replicate their own stream of edits. For more
information
+ on the different properties of master/slave replication and other types
+ of replication, please consult <a
href="http://highscalability.com/blog/2009/8/24/how-google-serves-data-from-multiple-datacenters.html";>
+ How Google Serves Data From Multiple Datacenters</a>.
</p>
<p>
The replication is done asynchronously, meaning that the clusters can
@@ -73,6 +76,17 @@
</p>
<img src="images/replication_overview.png"/>
</section>
+ <section name="Enabling replication">
+ <p>
+ The guide on enabling and using cluster replication is contained
+ in the API documentation shipped with your HBase distribution.
+ </p>
+ <p>
+ The most up-to-date documentation is
+ <a
href="apidocs/org/apache/hadoop/hbase/replication/package-summary.html#requirements">
+ available at this address</a>.
+ </p>
+ </section>
<section name="Life of a log edit">
<p>
The following sections describe the life of a single edit going from a
@@ -350,15 +364,6 @@
</section>
</section>
<section name="FAQ">
- <section name="Why do all clusters need to be in the same timezone?">
- <p>
- Suppose an edit to cell X happens in a EST cluster, then 2 minutes
- later a new edits happens to the same cell in a PST cluster and that
- both clusters are in a master-master replication. The second edit is
- considered younger, so the first will always hide it while in fact
the
- second is older.
- </p>
- </section>
<section name="GLOBAL means replicate? Any provision to replicate only
to cluster X and not to cluster Y? or is that for later?">
<p>
Yes, this is for much later.
