|
|
@@ -2,7 +2,7 @@
|
|
|
<html>
|
|
|
<head>
|
|
|
<META http-equiv="Content-Type" content="text/html; charset=UTF-8">
|
|
|
-<title>Hadoop 1.0.3 Release Notes</title>
|
|
|
+<title>Hadoop 1.0.4 Release Notes</title>
|
|
|
<STYLE type="text/css">
|
|
|
H1 {font-family: sans-serif}
|
|
|
H2 {font-family: sans-serif; margin-left: 7mm}
|
|
|
@@ -10,11 +10,41 @@
|
|
|
</STYLE>
|
|
|
</head>
|
|
|
<body>
|
|
|
-<h1>Hadoop 1.0.3 Release Notes</h1>
|
|
|
+<h1>Hadoop 1.0.4 Release Notes</h1>
|
|
|
These release notes include new developer and user-facing incompatibilities, features, and major improvements.
|
|
|
|
|
|
<a name="changes"/>
|
|
|
|
|
|
+<h2>Changes since Hadoop 1.0.3</h2>
|
|
|
+
|
|
|
+<h3>Jiras with Release Notes (describe major or incompatible changes)</h3>
|
|
|
+<ul>
|
|
|
+
|
|
|
+</ul>
|
|
|
+
|
|
|
+
|
|
|
+<h3>Other Jiras (describe bug fixes and minor changes)</h3>
|
|
|
+<ul>
|
|
|
+
|
|
|
+<li> <a href="https://issues.apache.org/jira/browse/HADOOP-7154">HADOOP-7154</a>.
|
|
|
+ Minor improvement reported by tlipcon and fixed by tlipcon (scripts)<br>
|
|
|
+ <b>Should set MALLOC_ARENA_MAX in hadoop-config.sh</b><br>
|
|
|
+ <blockquote>New versions of glibc present in RHEL6 include a new arena allocator design. In several clusters we've seen this new allocator cause huge amounts of virtual memory to be used, since when multiple threads perform allocations, they each get their own memory arena. On a 64-bit system, these arenas are 64M mappings, and the maximum number of arenas is 8 times the number of cores. We've observed a DN process using 14GB of vmem for only 300M of resident set. This causes all kinds of nasty issues fo...</blockquote></li>
|
|
|
+
|
|
|
+<li> <a href="https://issues.apache.org/jira/browse/HDFS-3652">HDFS-3652</a>.
|
|
|
+ Blocker bug reported by tlipcon and fixed by tlipcon (name-node)<br>
|
|
|
+ <b>1.x: FSEditLog failure removes the wrong edit stream when storage dirs have same name</b><br>
|
|
|
+ <blockquote>In {{FSEditLog.removeEditsForStorageDir}}, we iterate over the edits streams trying to find the stream corresponding to a given dir. To check equality, we currently use the following condition:<br>{code}<br> File parentDir = getStorageDirForStream(idx);<br> if (parentDir.getName().equals(sd.getRoot().getName())) {<br>{code}<br>... which is horribly incorrect. If two or more storage dirs happen to have the same terminal path component (eg /data/1/nn and /data/2/nn) then it will pick the wrong strea...</blockquote></li>
|
|
|
+
|
|
|
+<li> <a href="https://issues.apache.org/jira/browse/MAPREDUCE-4399">MAPREDUCE-4399</a>.
|
|
|
+ Major bug reported by vicaya and fixed by vicaya (performance, tasktracker)<br>
|
|
|
+ <b>Fix performance regression in shuffle </b><br>
|
|
|
+ <blockquote>There is a significant (up to 3x) performance regression in shuffle (vs 0.20.2) in the Hadoop 1.x series. Most noticeable with high-end switches.</blockquote></li>
|
|
|
+
|
|
|
+
|
|
|
+</ul>
|
|
|
+
|
|
|
+
|
|
|
<h2>Changes since Hadoop 1.0.2</h2>
|
|
|
|
|
|
<h3>Jiras with Release Notes (describe major or incompatible changes)</h3>
|
Matthew Foley