hbase-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From st...@apache.org
Subject svn commit: r660364 - in /hadoop/hbase/branches/0.1: CHANGES.txt src/java/org/apache/hadoop/hbase/HRegionServer.java
Date Tue, 27 May 2008 04:09:55 GMT
Author: stack
Date: Mon May 26 21:09:51 2008
New Revision: 660364

URL: http://svn.apache.org/viewvc?rev=660364&view=rev
Log:
HBASE-644 DroppedSnapshotException but RegionServer doesn't restart

Modified:
    hadoop/hbase/branches/0.1/CHANGES.txt
    hadoop/hbase/branches/0.1/src/java/org/apache/hadoop/hbase/HRegionServer.java

Modified: hadoop/hbase/branches/0.1/CHANGES.txt
URL: http://svn.apache.org/viewvc/hadoop/hbase/branches/0.1/CHANGES.txt?rev=660364&r1=660363&r2=660364&view=diff
==============================================================================
--- hadoop/hbase/branches/0.1/CHANGES.txt (original)
+++ hadoop/hbase/branches/0.1/CHANGES.txt Mon May 26 21:09:51 2008
@@ -1,5 +1,8 @@
 HBase Change Log
 
+  BUG FIXES
+   HBASE-644   DroppedSnapshotException but RegionServer doesn't restart
+
 Release 0.1.2 - 05/13/2008
 
   BUG FIXES

Modified: hadoop/hbase/branches/0.1/src/java/org/apache/hadoop/hbase/HRegionServer.java
URL: http://svn.apache.org/viewvc/hadoop/hbase/branches/0.1/src/java/org/apache/hadoop/hbase/HRegionServer.java?rev=660364&r1=660363&r2=660364&view=diff
==============================================================================
--- hadoop/hbase/branches/0.1/src/java/org/apache/hadoop/hbase/HRegionServer.java (original)
+++ hadoop/hbase/branches/0.1/src/java/org/apache/hadoop/hbase/HRegionServer.java Mon May
26 21:09:51 2008
@@ -430,8 +430,7 @@
      * times.
      * 
      * @return true if the region was successfully flushed, false otherwise. If 
-     * false, there will be accompanying log messages explaining why the log was
-     * not flushed.
+     * false, we exit the Flusher thread.
      */
     private boolean flushRegion(HRegion region, boolean removeFromQueue) {
       synchronized (regionsInQueue) {
@@ -451,12 +450,10 @@
           // Cache flush can fail in a few places.  If it fails in a critical
           // section, we get a DroppedSnapshotException and a replay of hlog
           // is required. Currently the only way to do this is a restart of
-          // the server.
+          // the server.  Abort because hdfs is probably bad (HBASE-644 is
+          // a case where hdfs was bad but passed the hdfs check).
           LOG.fatal("Replay of hlog required. Forcing server restart", ex);
-          if (!checkFileSystem()) {
-            return false;
-          }
-          server.stop();
+          abort();
           return false;
         } catch (IOException ex) {
           LOG.error("Cache flush failed" +
@@ -771,8 +768,7 @@
                       serverInfo.setStartCode(System.currentTimeMillis());
                       log = setupHLog();
                     } catch (IOException e) {
-                      this.abortRequested = true;
-                      this.stopRequested.set(true);
+                      abort();
                       e = RemoteExceptionHandler.checkIOException(e); 
                       LOG.fatal("error restarting server", e);
                       break;
@@ -1047,12 +1043,14 @@
 
   /**
    * Sets a flag that will cause all the HRegionServer threads to shut down
-   * in an orderly fashion.  Used by unit tests and called by {@link Flusher}
-   * if it judges server needs to be restarted.
+   * in an orderly fashion.  Used by unit tests.
    */
-  synchronized void stop() {
+  void stop() {
     this.stopRequested.set(true);
-    notifyAll();                        // Wakes run() if it is sleeping
+    synchronized (this) {
+      // Wakes run() if it is sleeping
+      notifyAll();
+    }
   }
   
   /**
@@ -1061,7 +1059,7 @@
    * Used unit testing and on catastrophic events such as HDFS is yanked out
    * from under hbase or we OOME.
    */
-  synchronized void abort() {
+  void abort() {
     this.abortRequested = true;
     stop();
   }
@@ -1776,14 +1774,13 @@
         FSUtils.checkFileSystemAvailable(fs);
       } catch (IOException e) {
         LOG.fatal("Shutting down HRegionServer: file system not available", e);
-        this.abortRequested = true;
-        this.stopRequested.set(true);
+        abort();
         fsOk = false;
       }
     }
     return this.fsOk;
   }
- 
+
   /**
    * @return Returns list of non-closed regions hosted on this server.  If no
    * regions to check, returns an empty list.



Mime
View raw message