HBASE-27632 Refactor WAL.Reader implementation so we can better support WAL splitting and replication (#5055)

Signed-off-by: GeorryHuang <[email protected]>

Author: Apache9

hbase-it/src/test/java/org/apache/hadoop/hbase/IntegrationTestIngestWithEncryption.java

@@ -26,11 +26,9 @@
 import org.apache.hadoop.hbase.client.TableDescriptor;
 import org.apache.hadoop.hbase.io.crypto.KeyProviderForTesting;
 import org.apache.hadoop.hbase.io.hfile.HFile;
-import org.apache.hadoop.hbase.regionserver.wal.SecureProtobufLogReader;
 import org.apache.hadoop.hbase.regionserver.wal.SecureProtobufLogWriter;
 import org.apache.hadoop.hbase.testclassification.IntegrationTests;
 import org.apache.hadoop.hbase.util.EncryptionTest;
-import org.apache.hadoop.hbase.wal.WAL.Reader;
 import org.apache.hadoop.hbase.wal.WALProvider.Writer;
 import org.apache.hadoop.util.ToolRunner;
 import org.junit.Before;
@@ -54,8 +52,6 @@ public void setUpCluster() throws Exception {
       conf.setInt(HFile.FORMAT_VERSION_KEY, 3);
       conf.set(HConstants.CRYPTO_KEYPROVIDER_CONF_KEY, KeyProviderForTesting.class.getName());
       conf.set(HConstants.CRYPTO_MASTERKEY_NAME_CONF_KEY, "hbase");
-      conf.setClass("hbase.regionserver.hlog.reader.impl", SecureProtobufLogReader.class,
-        Reader.class);
       conf.setClass("hbase.regionserver.hlog.writer.impl", SecureProtobufLogWriter.class,
         Writer.class);
       conf.setBoolean(HConstants.ENABLE_WAL_ENCRYPTION, true);

hbase-mapreduce/src/main/java/org/apache/hadoop/hbase/mapreduce/WALInputFormat.java

@@ -32,12 +32,14 @@
 import org.apache.hadoop.fs.LocatedFileStatus;
 import org.apache.hadoop.fs.Path;
 import org.apache.hadoop.fs.RemoteIterator;
+import org.apache.hadoop.hbase.util.LeaseNotRecoveredException;
 import org.apache.hadoop.hbase.wal.AbstractFSWALProvider;
 import org.apache.hadoop.hbase.wal.WAL;
 import org.apache.hadoop.hbase.wal.WAL.Entry;
-import org.apache.hadoop.hbase.wal.WAL.Reader;
 import org.apache.hadoop.hbase.wal.WALEdit;
+import org.apache.hadoop.hbase.wal.WALFactory;
 import org.apache.hadoop.hbase.wal.WALKey;
+import org.apache.hadoop.hbase.wal.WALStreamReader;
 import org.apache.hadoop.io.Writable;
 import org.apache.hadoop.mapreduce.InputFormat;
 import org.apache.hadoop.mapreduce.InputSplit;
@@ -135,7 +137,7 @@ public String toString() {
    * HLogInputFormat.
    */
   static abstract class WALRecordReader<K extends WALKey> extends RecordReader<K, WALEdit> {
-    private Reader reader = null;
+    private WALStreamReader reader = null;
     // visible until we can remove the deprecated HLogInputFormat
     Entry currentEntry = new Entry();
     private long startTime;
@@ -144,6 +146,47 @@ static abstract class WALRecordReader<K extends WALKey> extends RecordReader<K,
     private Path logFile;
     private long currentPos;
 
+    @edu.umd.cs.findbugs.annotations.SuppressWarnings(value = "DCN_NULLPOINTER_EXCEPTION",
+        justification = "HDFS-4380")
+    private WALStreamReader openReader(Path path, long startPosition) throws IOException {
+      long retryInterval = 2000; // 2 sec
+      int maxAttempts = 30;
+      int attempt = 0;
+      Exception ee = null;
+      WALStreamReader reader = null;
+      while (reader == null && attempt++ < maxAttempts) {
+        try {
+          // Detect if this is a new file, if so get a new reader else
+          // reset the current reader so that we see the new data
+          reader =
+            WALFactory.createStreamReader(path.getFileSystem(conf), path, conf, startPosition);
+          return reader;
+        } catch (LeaseNotRecoveredException lnre) {
+          // HBASE-15019 the WAL was not closed due to some hiccup.
+          LOG.warn("Try to recover the WAL lease " + path, lnre);
+          AbstractFSWALProvider.recoverLease(conf, path);
+          reader = null;
+          ee = lnre;
+        } catch (NullPointerException npe) {
+          // Workaround for race condition in HDFS-4380
+          // which throws a NPE if we open a file before any data node has the most recent block
+          // Just sleep and retry. Will require re-reading compressed WALs for compressionContext.
+          LOG.warn("Got NPE opening reader, will retry.");
+          reader = null;
+          ee = npe;
+        }
+        if (reader == null) {
+          // sleep before next attempt
+          try {
+            Thread.sleep(retryInterval);
+          } catch (InterruptedException e) {
+            Thread.currentThread().interrupt();
+          }
+        }
+      }
+      throw new IOException("Could not open reader", ee);
+    }
+
     @Override
     public void initialize(InputSplit split, TaskAttemptContext context)
       throws IOException, InterruptedException {
@@ -158,8 +201,7 @@ public void initialize(InputSplit split, TaskAttemptContext context)
 
     private void openReader(Path path) throws IOException {
       closeReader();
-      reader = AbstractFSWALProvider.openReader(path, conf);
-      seek();
+      reader = openReader(path, currentPos > 0 ? currentPos : -1);
       setCurrentPath(path);
     }
 
@@ -174,12 +216,6 @@ private void closeReader() throws IOException {
       }
     }
 
-    private void seek() throws IOException {
-      if (currentPos != 0) {
-        reader.seek(currentPos);
-      }
-    }
-
     @Override
     public boolean nextKeyValue() throws IOException, InterruptedException {
       if (reader == null) {

hbase-server/src/main/java/org/apache/hadoop/hbase/master/region/MasterRegion.java

@@ -271,6 +271,9 @@ private static HRegion open(Configuration conf, TableDescriptor td, RegionInfo r
     WAL wal = createWAL(walFactory, walRoller, serverName, walFs, walRootDir, regionInfo);
     conf.set(HRegion.SPECIAL_RECOVERED_EDITS_DIR,
       replayEditsDir.makeQualified(walFs.getUri(), walFs.getWorkingDirectory()).toString());
+    // we do not do WAL splitting here so it is possible to have uncleanly closed WAL files, so we
+    // need to ignore EOFException.
+    conf.setBoolean(HRegion.RECOVERED_EDITS_IGNORE_EOF, true);
     return HRegion.openHRegionFromTableDir(conf, fs, tableDir, regionInfo, td, wal, null, null);
   }
 

hbase-server/src/main/java/org/apache/hadoop/hbase/procedure2/store/region/WALProcedurePrettyPrinter.java

@@ -37,6 +37,7 @@
 import org.apache.hadoop.hbase.wal.WALFactory;
 import org.apache.hadoop.hbase.wal.WALKey;
 import org.apache.hadoop.hbase.wal.WALPrettyPrinter;
+import org.apache.hadoop.hbase.wal.WALStreamReader;
 import org.apache.yetus.audience.InterfaceAudience;
 import org.apache.yetus.audience.InterfaceStability;
 
@@ -88,7 +89,7 @@ protected void processOptions(CommandLine cmd) {
   protected int doWork() throws Exception {
     Path path = new Path(file);
     FileSystem fs = path.getFileSystem(conf);
-    try (WAL.Reader reader = WALFactory.createReader(fs, path, conf)) {
+    try (WALStreamReader reader = WALFactory.createStreamReader(fs, path, conf)) {
       for (;;) {
         WAL.Entry entry = reader.next();
         if (entry == null) {

hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java

@@ -182,6 +182,7 @@
 import org.apache.hadoop.hbase.wal.WALKeyImpl;
 import org.apache.hadoop.hbase.wal.WALSplitUtil;
 import org.apache.hadoop.hbase.wal.WALSplitUtil.MutationReplay;
+import org.apache.hadoop.hbase.wal.WALStreamReader;
 import org.apache.hadoop.util.StringUtils;
 import org.apache.yetus.audience.InterfaceAudience;
 import org.slf4j.Logger;
@@ -268,6 +269,14 @@ public class HRegion implements HeapSize, PropagatingConfigurationObserver, Regi
   public static final String SPECIAL_RECOVERED_EDITS_DIR =
     "hbase.hregion.special.recovered.edits.dir";
 
+  /**
+   * Mainly used for master local region, where we will replay the WAL file directly without
+   * splitting, so it is possible to have WAL files which are not closed cleanly, in this way,
+   * hitting EOF is expected so should not consider it as a critical problem.
+   */
+  public static final String RECOVERED_EDITS_IGNORE_EOF =
+    "hbase.hregion.recovered.edits.ignore.eof";
+
   /**
    * Whether to use {@link MetaCellComparator} even if we are not meta region. Used when creating
    * master local region.
@@ -5533,9 +5542,7 @@ private long replayRecoveredEdits(final Path edits, Map<byte[], Long> maxSeqIdIn
     MonitoredTask status = TaskMonitor.get().createStatus(msg);
 
     status.setStatus("Opening recovered edits");
-    WAL.Reader reader = null;
-    try {
-      reader = WALFactory.createReader(fs, edits, conf);
+    try (WALStreamReader reader = WALFactory.createStreamReader(fs, edits, conf)) {
       long currentEditSeqId = -1;
       long currentReplaySeqId = -1;
       long firstSeqIdInLog = -1;
@@ -5689,12 +5696,17 @@ private long replayRecoveredEdits(final Path edits, Map<byte[], Long> maxSeqIdIn
           coprocessorHost.postReplayWALs(this.getRegionInfo(), edits);
         }
       } catch (EOFException eof) {
-        Path p = WALSplitUtil.moveAsideBadEditsFile(walFS, edits);
-        msg = "EnLongAddered EOF. Most likely due to Master failure during "
-          + "wal splitting, so we have this data in another edit. Continuing, but renaming " + edits
-          + " as " + p + " for region " + this;
-        LOG.warn(msg, eof);
-        status.abort(msg);
+        if (!conf.getBoolean(RECOVERED_EDITS_IGNORE_EOF, false)) {
+          Path p = WALSplitUtil.moveAsideBadEditsFile(walFS, edits);
+          msg = "EnLongAddered EOF. Most likely due to Master failure during "
+            + "wal splitting, so we have this data in another edit. Continuing, but renaming "
+            + edits + " as " + p + " for region " + this;
+          LOG.warn(msg, eof);
+          status.abort(msg);
+        } else {
+          LOG.warn("EOF while replaying recover edits and config '{}' is true so "
+            + "we will ignore it and continue", RECOVERED_EDITS_IGNORE_EOF, eof);
+        }
       } catch (IOException ioe) {
         // If the IOE resulted from bad file format,
         // then this problem is idempotent and retrying won't help
@@ -5721,9 +5733,6 @@ private long replayRecoveredEdits(final Path edits, Map<byte[], Long> maxSeqIdIn
       return currentEditSeqId;
     } finally {
       status.cleanup();
-      if (reader != null) {
-        reader.close();
-      }
     }
   }
 

hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/wal/AbstractFSWAL.java

@@ -110,8 +110,9 @@
  * <code>F</code> when all of the edits in <code>F</code> have a log-sequence-id that's older
  * (smaller) than the most-recent flush.
  * <p>
- * To read an WAL, call
- * {@link WALFactory#createReader(org.apache.hadoop.fs.FileSystem, org.apache.hadoop.fs.Path)}. *
+ * To read an WAL, call {@link WALFactory#createStreamReader(FileSystem, Path)} for one way read,
+ * call {@link WALFactory#createTailingReader(FileSystem, Path, Configuration, long)} for
+ * replication where we may want to tail the active WAL file.
  * <h2>Failure Semantic</h2> If an exception on append or sync, roll the WAL because the current WAL
  * is now a lame duck; any more appends or syncs will fail also with the same original exception. If
  * we have made successful appends to the WAL and we then are unable to sync them, our current

hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/wal/AbstractProtobufLogWriter.java

@@ -17,8 +17,10 @@
  */
 package org.apache.hadoop.hbase.regionserver.wal;
 
-import static org.apache.hadoop.hbase.regionserver.wal.ProtobufLogReader.DEFAULT_WAL_TRAILER_WARN_SIZE;
-import static org.apache.hadoop.hbase.regionserver.wal.ProtobufLogReader.WAL_TRAILER_WARN_SIZE;
+import static org.apache.hadoop.hbase.regionserver.wal.AbstractProtobufWALReader.DEFAULT_WAL_TRAILER_WARN_SIZE;
+import static org.apache.hadoop.hbase.regionserver.wal.AbstractProtobufWALReader.PB_WAL_COMPLETE_MAGIC;
+import static org.apache.hadoop.hbase.regionserver.wal.AbstractProtobufWALReader.PB_WAL_MAGIC;
+import static org.apache.hadoop.hbase.regionserver.wal.AbstractProtobufWALReader.WAL_TRAILER_WARN_SIZE;
 
 import java.io.IOException;
 import java.io.OutputStream;
@@ -185,8 +187,7 @@ public void init(FileSystem fs, Path path, Configuration conf, boolean overwrita
         headerBuilder.setValueCompressionAlgorithm(
           CompressionContext.getValueCompressionAlgorithm(conf).ordinal());
       }
-      length.set(writeMagicAndWALHeader(ProtobufLogReader.PB_WAL_MAGIC,
-        buildWALHeader(conf, headerBuilder)));
+      length.set(writeMagicAndWALHeader(PB_WAL_MAGIC, buildWALHeader(conf, headerBuilder)));
 
       initAfterHeader(doCompress);
 
@@ -257,7 +258,7 @@ protected void writeWALTrailer() {
         LOG.warn("Please investigate WALTrailer usage. Trailer size > maximum size : " + trailerSize
           + " > " + this.trailerWarnSize);
       }
-      length.set(writeWALTrailerAndMagic(trailer, ProtobufLogReader.PB_WAL_COMPLETE_MAGIC));
+      length.set(writeWALTrailerAndMagic(trailer, PB_WAL_COMPLETE_MAGIC));
       this.trailerWritten = true;
     } catch (IOException ioe) {
       LOG.warn("Failed to write trailer, non-fatal, continuing...", ioe);