HBASE-2868 Do some small cleanups in org.apache.hadoop.hbase.regionserver.wal

git-svn-id: https://svn.apache.org/repos/asf/hbase/trunk@984433 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael Stack 2010-08-11 15:03:21 +00:00
parent 8b5d7dbb1c
commit 92ffa8580c
3 changed files with 143 additions and 97 deletions

View File

@ -833,6 +833,8 @@ Release 0.21.0 - Unreleased
(Chongxin Li via Stack) (Chongxin Li via Stack)
HBASE-2844 Capping the number of regions (Pranav Khaitan via Stack) HBASE-2844 Capping the number of regions (Pranav Khaitan via Stack)
HBASE-2870 Add Backup CLI Option to HMaster (Nicolas Spiegelberg via Stack) HBASE-2870 Add Backup CLI Option to HMaster (Nicolas Spiegelberg via Stack)
HBASE-2868 Do some small cleanups in org.apache.hadoop.hbase.regionserver.wal
(Alex Newman via Stack)
NEW FEATURES NEW FEATURES
HBASE-1961 HBase EC2 scripts HBASE-1961 HBase EC2 scripts

View File

@ -28,6 +28,7 @@ import java.io.FileNotFoundException;
import java.io.IOException; import java.io.IOException;
import java.io.OutputStream; import java.io.OutputStream;
import java.io.UnsupportedEncodingException; import java.io.UnsupportedEncodingException;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method; import java.lang.reflect.Method;
import java.net.URLEncoder; import java.net.URLEncoder;
import java.util.ArrayList; import java.util.ArrayList;
@ -1001,10 +1002,13 @@ public class HLog implements Syncable {
* If the pipeline isn't started yet or is empty, you will get the default * If the pipeline isn't started yet or is empty, you will get the default
* replication factor. Therefore, if this function returns 0, it means you * replication factor. Therefore, if this function returns 0, it means you
* are not properly running with the HDFS-826 patch. * are not properly running with the HDFS-826 patch.
* @throws InvocationTargetException
* @throws IllegalAccessException
* @throws IllegalArgumentException
* *
* @throws Exception * @throws Exception
*/ */
int getLogReplication() throws Exception { int getLogReplication() throws IllegalArgumentException, IllegalAccessException, InvocationTargetException {
if(this.getNumCurrentReplicas != null && this.hdfs_out != null) { if(this.getNumCurrentReplicas != null && this.hdfs_out != null) {
Object repl = this.getNumCurrentReplicas.invoke(this.hdfs_out, NO_ARGS); Object repl = this.getNumCurrentReplicas.invoke(this.hdfs_out, NO_ARGS);
if (repl instanceof Integer) { if (repl instanceof Integer) {

View File

@ -19,7 +19,9 @@
*/ */
package org.apache.hadoop.hbase.regionserver.wal; package org.apache.hadoop.hbase.regionserver.wal;
import java.io.IOException;
import java.io.OutputStream; import java.io.OutputStream;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method; import java.lang.reflect.Method;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.List; import java.util.List;
@ -27,10 +29,13 @@ import java.util.List;
import org.apache.commons.logging.Log; import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory; import org.apache.commons.logging.LogFactory;
import org.apache.commons.logging.impl.Log4JLogger; import org.apache.commons.logging.impl.Log4JLogger;
import org.apache.hadoop.hbase.HBaseClusterTestCase;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.hbase.HBaseTestingUtility;
import org.apache.hadoop.hbase.HColumnDescriptor; import org.apache.hadoop.hbase.HColumnDescriptor;
import org.apache.hadoop.hbase.HConstants; import org.apache.hadoop.hbase.HConstants;
import org.apache.hadoop.hbase.HTableDescriptor; import org.apache.hadoop.hbase.HTableDescriptor;
import org.apache.hadoop.hbase.MiniHBaseCluster;
import org.apache.hadoop.hbase.regionserver.HRegionServer; import org.apache.hadoop.hbase.regionserver.HRegionServer;
import org.apache.hadoop.hbase.regionserver.HRegion; import org.apache.hadoop.hbase.regionserver.HRegion;
import org.apache.hadoop.hbase.client.HBaseAdmin; import org.apache.hadoop.hbase.client.HBaseAdmin;
@ -39,22 +44,33 @@ import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.util.Bytes; import org.apache.hadoop.hbase.util.Bytes;
import org.apache.hadoop.hbase.util.FSUtils; import org.apache.hadoop.hbase.util.FSUtils;
import org.apache.hadoop.hdfs.DFSClient; import org.apache.hadoop.hdfs.DFSClient;
import org.apache.hadoop.hdfs.MiniDFSCluster.DataNodeProperties; import org.apache.hadoop.hdfs.MiniDFSCluster;
import org.apache.hadoop.hdfs.protocol.DatanodeInfo; import org.apache.hadoop.hdfs.protocol.DatanodeInfo;
import org.apache.hadoop.hdfs.server.datanode.DataNode; import org.apache.hadoop.hdfs.server.datanode.DataNode;
import org.apache.hadoop.hdfs.server.namenode.FSNamesystem; import org.apache.hadoop.hdfs.server.namenode.FSNamesystem;
import org.apache.hadoop.hdfs.server.namenode.LeaseManager; import org.apache.hadoop.hdfs.server.namenode.LeaseManager;
import org.apache.log4j.Level; import org.apache.log4j.Level;
import org.junit.AfterClass;
import org.junit.BeforeClass;
import org.junit.Test;
import static org.junit.Assert.assertTrue;
/** /**
* Test log deletion as logs are rolled. * Test log deletion as logs are rolled.
*/ */
public class TestLogRolling extends HBaseClusterTestCase { public class TestLogRolling {
private static final Log LOG = LogFactory.getLog(TestLogRolling.class); private static final Log LOG = LogFactory.getLog(TestLogRolling.class);
private HRegionServer server; private HRegionServer server;
private HLog log; private HLog log;
private String tableName; private String tableName;
private byte[] value; private byte[] value;
private static FileSystem fs;
private static MiniDFSCluster dfsCluster;
private static HBaseAdmin admin;
private static MiniHBaseCluster cluster;
private final static HBaseTestingUtility TEST_UTIL = new HBaseTestingUtility();
// verbose logging on classes that are touched in these tests // verbose logging on classes that are touched in these tests
{ {
@ -71,10 +87,9 @@ public class TestLogRolling extends HBaseClusterTestCase {
* constructor * constructor
* @throws Exception * @throws Exception
*/ */
public TestLogRolling() throws Exception { public TestLogRolling() {
// start one regionserver and a minidfs. // start one regionserver and a minidfs.
super(); super();
try {
this.server = null; this.server = null;
this.log = null; this.log = null;
this.tableName = null; this.tableName = null;
@ -86,63 +101,67 @@ public class TestLogRolling extends HBaseClusterTestCase {
v.append(className); v.append(className);
} }
value = Bytes.toBytes(v.toString()); value = Bytes.toBytes(v.toString());
} catch (Exception e) {
LOG.fatal("error in constructor", e);
throw e;
}
} }
// Need to override this setup so we can edit the config before it gets sent // Need to override this setup so we can edit the config before it gets sent
// to the HDFS & HBase cluster startup. // to the HDFS & HBase cluster startup.
@Override @BeforeClass
protected void setUp() throws Exception { public static void setUpBeforeClass() throws Exception {
/**** configuration for testLogRolling ****/ /**** configuration for testLogRolling ****/
// Force a region split after every 768KB // Force a region split after every 768KB
conf.setLong("hbase.hregion.max.filesize", 768L * 1024L); TEST_UTIL.getConfiguration().setLong("hbase.hregion.max.filesize", 768L * 1024L);
// We roll the log after every 32 writes // We roll the log after every 32 writes
conf.setInt("hbase.regionserver.maxlogentries", 32); TEST_UTIL.getConfiguration().setInt("hbase.regionserver.maxlogentries", 32);
// For less frequently updated regions flush after every 2 flushes // For less frequently updated regions flush after every 2 flushes
conf.setInt("hbase.hregion.memstore.optionalflushcount", 2); TEST_UTIL.getConfiguration().setInt("hbase.hregion.memstore.optionalflushcount", 2);
// We flush the cache after every 8192 bytes // We flush the cache after every 8192 bytes
conf.setInt("hbase.hregion.memstore.flush.size", 8192); TEST_UTIL.getConfiguration().setInt("hbase.hregion.memstore.flush.size", 8192);
// Increase the amount of time between client retries // Increase the amount of time between client retries
conf.setLong("hbase.client.pause", 15 * 1000); TEST_UTIL.getConfiguration().setLong("hbase.client.pause", 15 * 1000);
// Reduce thread wake frequency so that other threads can get // Reduce thread wake frequency so that other threads can get
// a chance to run. // a chance to run.
conf.setInt(HConstants.THREAD_WAKE_FREQUENCY, 2 * 1000); TEST_UTIL.getConfiguration().setInt(HConstants.THREAD_WAKE_FREQUENCY, 2 * 1000);
/**** configuration for testLogRollOnDatanodeDeath ****/ /**** configuration for testLogRollOnDatanodeDeath ****/
// make sure log.hflush() calls syncFs() to open a pipeline // make sure log.hflush() calls syncFs() to open a pipeline
conf.setBoolean("dfs.support.append", true); TEST_UTIL.getConfiguration().setBoolean("dfs.support.append", true);
// lower the namenode & datanode heartbeat so the namenode // lower the namenode & datanode heartbeat so the namenode
// quickly detects datanode failures // quickly detects datanode failures
conf.setInt("heartbeat.recheck.interval", 5000); TEST_UTIL.getConfiguration().setInt("heartbeat.recheck.interval", 5000);
conf.setInt("dfs.heartbeat.interval", 1); TEST_UTIL.getConfiguration().setInt("dfs.heartbeat.interval", 1);
// the namenode might still try to choose the recently-dead datanode // the namenode might still try to choose the recently-dead datanode
// for a pipeline, so try to a new pipeline multiple times // for a pipeline, so try to a new pipeline multiple times
conf.setInt("dfs.client.block.write.retries", 30); TEST_UTIL.getConfiguration().setInt("dfs.client.block.write.retries", 30);
TEST_UTIL.startMiniCluster(2);
super.setUp();
cluster = TEST_UTIL.getHBaseCluster();
dfsCluster = TEST_UTIL.getDFSCluster();
fs = TEST_UTIL.getTestFileSystem();
admin = TEST_UTIL.getHBaseAdmin();
} }
private void startAndWriteData() throws Exception { @AfterClass
public static void tearDownAfterClass() throws IOException {
TEST_UTIL.cleanupTestDir();
TEST_UTIL.shutdownMiniCluster();
}
private void startAndWriteData() throws IOException {
// When the META table can be opened, the region servers are running // When the META table can be opened, the region servers are running
new HTable(conf, HConstants.META_TABLE_NAME); new HTable(TEST_UTIL.getConfiguration(), HConstants.META_TABLE_NAME);
this.server = cluster.getRegionServerThreads().get(0).getRegionServer(); this.server = cluster.getRegionServerThreads().get(0).getRegionServer();
this.log = server.getLog(); this.log = server.getLog();
// Create the test table and open it // Create the test table and open it
HTableDescriptor desc = new HTableDescriptor(tableName); HTableDescriptor desc = new HTableDescriptor(tableName);
desc.addFamily(new HColumnDescriptor(HConstants.CATALOG_FAMILY)); desc.addFamily(new HColumnDescriptor(HConstants.CATALOG_FAMILY));
HBaseAdmin admin = new HBaseAdmin(conf);
admin.createTable(desc); admin.createTable(desc);
HTable table = new HTable(conf, tableName); HTable table = new HTable(TEST_UTIL.getConfiguration(), tableName);
for (int i = 1; i <= 256; i++) { // 256 writes should cause 8 log rolls for (int i = 1; i <= 256; i++) { // 256 writes should cause 8 log rolls
Put put = new Put(Bytes.toBytes("row" + String.format("%1$04d", i))); Put put = new Put(Bytes.toBytes("row" + String.format("%1$04d", i)));
put.add(HConstants.CATALOG_FAMILY, null, value); put.add(HConstants.CATALOG_FAMILY, null, value);
@ -160,12 +179,12 @@ public class TestLogRolling extends HBaseClusterTestCase {
/** /**
* Tests that logs are deleted * Tests that logs are deleted
* * @throws IOException
* @throws Exception * @throws FailedLogCloseException
*/ */
public void testLogRolling() throws Exception { @Test
public void testLogRolling() throws FailedLogCloseException, IOException {
this.tableName = getName(); this.tableName = getName();
try {
startAndWriteData(); startAndWriteData();
LOG.info("after writing there are " + log.getNumLogFiles() + " log files"); LOG.info("after writing there are " + log.getNumLogFiles() + " log files");
@ -184,13 +203,13 @@ public class TestLogRolling extends HBaseClusterTestCase {
LOG.info("after flushing all regions and rolling logs there are " + LOG.info("after flushing all regions and rolling logs there are " +
log.getNumLogFiles() + " log files"); log.getNumLogFiles() + " log files");
assertTrue(("actual count: " + count), count <= 2); assertTrue(("actual count: " + count), count <= 2);
} catch (Exception e) {
LOG.fatal("unexpected exception", e);
throw e;
}
} }
void writeData(HTable table, int rownum) throws Exception { private static String getName() {
return "TestLogRolling";
}
void writeData(HTable table, int rownum) throws IOException {
Put put = new Put(Bytes.toBytes("row" + String.format("%1$04d", rownum))); Put put = new Put(Bytes.toBytes("row" + String.format("%1$04d", rownum)));
put.add(HConstants.CATALOG_FAMILY, null, value); put.add(HConstants.CATALOG_FAMILY, null, value);
table.put(put); table.put(put);
@ -202,80 +221,101 @@ public class TestLogRolling extends HBaseClusterTestCase {
// continue // continue
} }
} }
/** /**
* Tests that logs are rolled upon detecting datanode death * Give me the HDFS pipeline for this log file
* Requires an HDFS jar with HDFS-826 & syncFs() support (HDFS-200)
*
* @throws Exception
*/ */
public void testLogRollOnDatanodeDeath() throws Exception { @SuppressWarnings("null")
assertTrue("This test requires HLog file replication.", DatanodeInfo[] getPipeline(HLog log) throws IllegalArgumentException,
fs.getDefaultReplication() > 1); IllegalAccessException, InvocationTargetException {
// When the META table can be opened, the region servers are running
new HTable(conf, HConstants.META_TABLE_NAME);
this.server = cluster.getRegionServer(0);
this.log = server.getLog();
assertTrue("Need HDFS-826 for this test", log.canGetCurReplicas());
// don't run this test without append support (HDFS-200 & HDFS-142)
assertTrue("Need append support for this test", FSUtils.isAppendSupported(conf));
// add up the datanode count, to ensure proper replication when we kill 1
dfsCluster.startDataNodes(conf, 1, true, null, null);
dfsCluster.waitActive();
assertTrue(dfsCluster.getDataNodes().size() >=
fs.getDefaultReplication() + 1);
// Create the test table and open it
String tableName = getName();
HTableDescriptor desc = new HTableDescriptor(tableName);
desc.addFamily(new HColumnDescriptor(HConstants.CATALOG_FAMILY));
HBaseAdmin admin = new HBaseAdmin(conf);
admin.createTable(desc);
HTable table = new HTable(conf, tableName);
table.setAutoFlush(true);
long curTime = System.currentTimeMillis();
long oldFilenum = log.getFilenum();
assertTrue("Log should have a timestamp older than now",
curTime > oldFilenum && oldFilenum != -1);
// normal write
writeData(table, 1);
assertTrue("The log shouldn't have rolled yet",
oldFilenum == log.getFilenum());
// kill a datanode in the pipeline to force a log roll on the next sync() // kill a datanode in the pipeline to force a log roll on the next sync()
OutputStream stm = log.getOutputStream(); OutputStream stm = log.getOutputStream();
Method getPipeline = null; Method getPipeline = null;
for (Method m : stm.getClass().getDeclaredMethods()) { for (Method m : stm.getClass().getDeclaredMethods()) {
if(m.getName().endsWith("getPipeline")) { if (m.getName().endsWith("getPipeline")) {
getPipeline = m; getPipeline = m;
getPipeline.setAccessible(true); getPipeline.setAccessible(true);
break; break;
} }
} }
assertTrue("Need DFSOutputStream.getPipeline() for this test",
getPipeline != null);
Object repl = getPipeline.invoke(stm, new Object []{} /*NO_ARGS*/);
DatanodeInfo[] pipeline = (DatanodeInfo[]) repl;
assertTrue(pipeline.length == fs.getDefaultReplication());
DataNodeProperties dnprop = dfsCluster.stopDataNode(pipeline[0].getName());
assertTrue(dnprop != null);
assertTrue("Need DFSOutputStream.getPipeline() for this test",
null != getPipeline);
Object repl = getPipeline.invoke(stm, new Object[] {} /* NO_ARGS */);
return (DatanodeInfo[]) repl;
}
/**
* Tests that logs are rolled upon detecting datanode death
* Requires an HDFS jar with HDFS-826 & syncFs() support (HDFS-200)
* @throws IOException
* @throws InterruptedException
* @throws InvocationTargetException
* @throws IllegalAccessException
* @throws IllegalArgumentException
*/
@Test
public void testLogRollOnDatanodeDeath() throws IOException,
InterruptedException, IllegalArgumentException, IllegalAccessException,
InvocationTargetException {
assertTrue("This test requires HLog file replication.", fs
.getDefaultReplication() > 1);
// When the META table can be opened, the region servers are running
new HTable(TEST_UTIL.getConfiguration(), HConstants.META_TABLE_NAME);
this.server = cluster.getRegionServer(0);
this.log = server.getLog();
assertTrue("Need HDFS-826 for this test", log.canGetCurReplicas());
// don't run this test without append support (HDFS-200 & HDFS-142)
assertTrue("Need append support for this test", FSUtils
.isAppendSupported(TEST_UTIL.getConfiguration()));
// add up the datanode count, to ensure proper replication when we kill 1
dfsCluster
.startDataNodes(TEST_UTIL.getConfiguration(), 1, true, null, null);
dfsCluster.waitActive();
assertTrue(dfsCluster.getDataNodes().size() >= fs.getDefaultReplication() + 1);
// Create the test table and open it
String tableName = getName();
HTableDescriptor desc = new HTableDescriptor(tableName);
desc.addFamily(new HColumnDescriptor(HConstants.CATALOG_FAMILY));
if (admin.tableExists(tableName)) {
admin.disableTable(tableName);
admin.deleteTable(tableName);
}
admin.createTable(desc);
HTable table = new HTable(TEST_UTIL.getConfiguration(), tableName);
writeData(table, 2);
table.setAutoFlush(true);
long curTime = System.currentTimeMillis();
long oldFilenum = log.getFilenum();
assertTrue("Log should have a timestamp older than now",
curTime > oldFilenum && oldFilenum != -1);
assertTrue("The log shouldn't have rolled yet", oldFilenum == log
.getFilenum());
DatanodeInfo[] pipeline = getPipeline(log);
assertTrue(pipeline.length == fs.getDefaultReplication());
assertTrue(dfsCluster.stopDataNode(pipeline[0].getName()) != null);
Thread.sleep(10000);
// this write should succeed, but trigger a log roll // this write should succeed, but trigger a log roll
writeData(table, 2); writeData(table, 2);
long newFilenum = log.getFilenum(); long newFilenum = log.getFilenum();
assertTrue("Missing datanode should've triggered a log roll",
newFilenum > oldFilenum && newFilenum > curTime); assertTrue("Missing datanode should've triggered a log roll",
newFilenum > oldFilenum && newFilenum > curTime);
// write some more log data (this should use a new hdfs_out) // write some more log data (this should use a new hdfs_out)
writeData(table, 3); writeData(table, 3);
assertTrue("The log should not roll again.", assertTrue("The log should not roll again.", log.getFilenum() == newFilenum);
log.getFilenum() == newFilenum); assertTrue("New log file should have the default replication", log
assertTrue("New log file should have the default replication", .getLogReplication() == fs.getDefaultReplication());
log.getLogReplication() == fs.getDefaultReplication());
} }
} }