SOLR-1145: Add capability to specify an infoStream log file for the underlying Lucene IndexWriter in solrconfig.xml

git-svn-id: https://svn.apache.org/repos/asf/lucene/solr/trunk@791578 13f79535-47bb-0310-9956-ffa450edef68

CHANGES.txt

@@ -243,6 +243,10 @@ New Features
 
 62. SOLR-1216 :  disambiguate the replication command names. 'snappull' becomes 'fetchindex' 'abortsnappull' becomes 'abortfetch' (noble) 
  
+63. SOLR-1145: Add capability to specify an infoStream log file for the underlying Lucene IndexWriter in solrconfig.xml.
+    This is an advanced debug log file that can be used to aid developers in fixing IndexWriter bugs. See the commented
+    out example in the example solrconfig.xml under the indexDefaults section. 
+	(Chris Harris, Mark Miller)
 Optimizations
 ----------------------
  1. SOLR-374: Use IndexReader.reopen to save resources by re-using parts of the

example/solr/conf/solrconfig.xml

@@ -81,6 +81,12 @@
      -->
     <!--<mergeScheduler>org.apache.lucene.index.ConcurrentMergeScheduler</mergeScheduler>-->
 
+    <!--  
+	 To aid in advanced debugging, you may turn on IndexWriter debug logging. Uncommenting this and setting to true
+	 will set the file that the underlying Lucene IndexWriter will write its debug infostream to.
+	-->
+	<!-- <infoStream file="/path/file">false</infoStream> -->
+	  
     <!--
       This option specifies which Lucene LockFactory implementation to use.
       

src/java/org/apache/solr/update/SolrIndexConfig.java

@@ -20,6 +20,8 @@ package org.apache.solr.update;
 import org.apache.solr.core.SolrConfig;
 import org.apache.lucene.index.LogByteSizeMergePolicy;
 import org.apache.lucene.index.ConcurrentMergeScheduler;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 //
 // For performance reasons, we don't want to re-read
@@ -30,6 +32,8 @@ import org.apache.lucene.index.ConcurrentMergeScheduler;
  * @version $Id$
  */
 public class SolrIndexConfig {
+  public static final Logger log = LoggerFactory.getLogger(SolrIndexConfig.class);
+  
   public static final String defaultsName ="indexDefaults";
   public static final String DEFAULT_MERGE_POLICY_CLASSNAME = LogByteSizeMergePolicy.class.getName();
   public static final String DEFAULT_MERGE_SCHEDULER_CLASSNAME = ConcurrentMergeScheduler.class.getName();
@@ -66,6 +70,8 @@ public class SolrIndexConfig {
   public final String mergeSchedulerClassname;
   public final boolean luceneAutoCommit;
   
+  public String infoStreamFile = null;
+
   public SolrIndexConfig(SolrConfig solrConfig, String prefix, SolrIndexConfig def)  {
     if (prefix == null)
       prefix = defaultsName;
@@ -84,5 +90,12 @@ public class SolrIndexConfig {
     mergePolicyClassName = solrConfig.get(prefix + "/mergePolicy", def.mergePolicyClassName);
     mergeSchedulerClassname = solrConfig.get(prefix + "/mergeScheduler", def.mergeSchedulerClassname);
     luceneAutoCommit = solrConfig.getBool(prefix + "/luceneAutoCommit", def.luceneAutoCommit);
+    
+    boolean infoStreamEnabled = solrConfig.getBool(prefix + "/infoStream", false);
+    if(infoStreamEnabled) {
+      infoStreamFile= solrConfig.get(prefix + "/infoStream/@file", null);
+      log.info("IndexWriter infoStream debug log is enabled: " + infoStreamFile);
+    }
+
   }
 }

src/java/org/apache/solr/update/SolrIndexWriter.java

@@ -27,7 +27,14 @@ import org.apache.solr.schema.IndexSchema;
 
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
+
+import java.io.File;
+import java.io.FileOutputStream;
 import java.io.IOException;
+import java.io.OutputStream;
+import java.io.PrintStream;
+import java.text.DateFormat;
+import java.util.Date;
 
 /**
  * An IndexWriter that is configured via Solr config mechanisms.
@@ -43,6 +50,8 @@ public class SolrIndexWriter extends IndexWriter {
   String name;
   IndexSchema schema;
 
+  private PrintStream infoStream;
+
   private void init(String name, IndexSchema schema, SolrIndexConfig config) throws IOException {
     log.debug("Opened Writer " + name);
     this.name = name;
@@ -73,6 +82,14 @@ public class SolrIndexWriter extends IndexWriter {
         setMergeScheduler(scheduler);
       }
 
+      String infoStreamFile = config.infoStreamFile;
+      if (infoStreamFile != null) {
+        File f = new File(infoStreamFile);
+        f.getParentFile().mkdirs();
+        FileOutputStream fos = new FileOutputStream(f, true);
+        infoStream = new TimeLoggingPrintStream(fos, true);
+        setInfoStream(infoStream);
+      }
       //if (config.commitLockTimeout != -1) setWriteLockTimeout(config.commitLockTimeout);
     }
 
@@ -196,6 +213,9 @@ public class SolrIndexWriter extends IndexWriter {
   public void close() throws IOException {
     log.debug("Closing Writer " + name);
     super.close();
+    if(infoStream != null) {
+      infoStream.close();
+    }
   }
 
   @Override
@@ -208,4 +228,22 @@ public class SolrIndexWriter extends IndexWriter {
     
   }
   
+  // Helper class for adding timestamps to infoStream logging
+  class TimeLoggingPrintStream extends PrintStream {
+    private DateFormat dateFormat;
+    public TimeLoggingPrintStream(OutputStream underlyingOutputStream,
+        boolean autoFlush) {
+      super(underlyingOutputStream, autoFlush);
+      this.dateFormat = DateFormat.getDateTimeInstance();
+    }
+
+    // We might ideally want to override print(String) as well, but
+    // looking through the code that writes to infoStream, it appears
+    // that all the classes except CheckIndex just use println.
+    public void println(String x) {
+      print(dateFormat.format(new Date()) + " ");
+      super.println(x);
+    }
+  }
+
 }