HBASE-26552 Introduce retry to logroller to avoid abort

Author: sunhelly

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

@@ -60,13 +60,21 @@ public abstract class AbstractWALRoller<T extends Abortable> extends Thread
 
   protected static final String WAL_ROLL_PERIOD_KEY = "hbase.regionserver.logroll.period";
 
+  protected static final String WAL_ROLL_WAIT_TIMEOUT = "hbase.regionserver.logroll.wait.timeout.ms";
+
+  protected static final String WAL_ROLL_RETRIES = "hbase.regionserver.logroll.retries";
+
   protected final ConcurrentMap<WAL, RollController> wals = new ConcurrentHashMap<>();
   protected final T abortable;
   // Period to roll log.
   private final long rollPeriod;
   private final int threadWakeFrequency;
   // The interval to check low replication on hlog's pipeline
   private final long checkLowReplicationInterval;
+  // Wait period for roll log
+  private final long rollWaitTimeout;
+  // Max retry for roll log
+  private final int maxRollRetry;
 
   private volatile boolean running = true;
 
@@ -114,6 +122,8 @@ protected AbstractWALRoller(String name, Configuration conf, T abortable) {
     this.threadWakeFrequency = conf.getInt(HConstants.THREAD_WAKE_FREQUENCY, 10 * 1000);
     this.checkLowReplicationInterval =
       conf.getLong("hbase.regionserver.hlog.check.lowreplication.interval", 30 * 1000);
+    this.rollWaitTimeout = conf.getLong(WAL_ROLL_WAIT_TIMEOUT, 30000);
+    this.maxRollRetry = conf.getInt(WAL_ROLL_RETRIES, 2);
   }
 
   /**
@@ -184,18 +194,38 @@ public void run() {
           } else {
             continue;
           }
-          try {
-            // Force the roll if the logroll.period is elapsed or if a roll was requested.
-            // The returned value is an collection of actual region and family names.
-            Map<byte[], List<byte[]>> regionsToFlush = controller.rollWal(now);
-            if (regionsToFlush != null) {
-              for (Map.Entry<byte[], List<byte[]>> r : regionsToFlush.entrySet()) {
-                scheduleFlush(Bytes.toString(r.getKey()), r.getValue());
+          Map<byte[], List<byte[]>> regionsToFlush = null;
+          int nAttempts = 0;
+          long startWaiting = EnvironmentEdgeManager.currentTime();
+          do {
+            try {
+              // Force the roll if the logroll.period is elapsed or if a roll was requested.
+              // The returned value is an collection of actual region and family names.
+              regionsToFlush = controller.rollWal(EnvironmentEdgeManager.currentTime());
+              break;
+            } catch (IOException ioe) {
+              if (ioe instanceof WALClosedException) {
+                LOG.warn("WAL has been closed. Skipping rolling of writer and just remove it", ioe);
+                iter.remove();
+                break;
               }
+              long waitingTime = EnvironmentEdgeManager.currentTime() - startWaiting;
+              if (waitingTime < rollWaitTimeout && nAttempts < maxRollRetry) {
+                nAttempts++;
+                LOG.warn("Retry to roll log, nAttempts={}, waiting time={}ms, sleeping 1s to retry,"
+                    + " last excepiton= {}", nAttempts, waitingTime,
+                  ioe.getCause().getClass().getSimpleName());
+                sleep(1000);
+              } else {
+                LOG.error("Roll wal failed and waiting timeout, will not retry", ioe);
+                throw ioe;
+              }
+            }
+          } while (EnvironmentEdgeManager.currentTime() - startWaiting < rollWaitTimeout);
+          if (regionsToFlush != null) {
+            for (Map.Entry<byte[], List<byte[]>> r : regionsToFlush.entrySet()) {
+              scheduleFlush(Bytes.toString(r.getKey()), r.getValue());
             }
-          } catch (WALClosedException e) {
-            LOG.warn("WAL has been closed. Skipping rolling of writer and just remove it", e);
-            iter.remove();
           }
         }
       } catch (FailedLogCloseException | ConnectException e) {