hbase-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From st...@apache.org
Subject svn commit: r735783 - in /hadoop/hbase/trunk: CHANGES.txt src/java/overview.html
Date Mon, 19 Jan 2009 19:46:50 GMT
Author: stack
Date: Mon Jan 19 11:46:50 2009
New Revision: 735783

URL: http://svn.apache.org/viewvc?rev=735783&view=rev
HBASE-1137 Add not on xceivers count to overview documentation


Modified: hadoop/hbase/trunk/CHANGES.txt
URL: http://svn.apache.org/viewvc/hadoop/hbase/trunk/CHANGES.txt?rev=735783&r1=735782&r2=735783&view=diff
--- hadoop/hbase/trunk/CHANGES.txt (original)
+++ hadoop/hbase/trunk/CHANGES.txt Mon Jan 19 11:46:50 2009
@@ -226,6 +226,7 @@
                (Michael Gottesman via Stack)
    HBASE-1082  Administrative functions for table/region maintenance
    HBASE-1090  Atomic Check And Save in HTable (Michael Gottesman via Stack)
+   HBASE-1137  Add not on xceivers count to overview documentation
    HBASE-875   Use MurmurHash instead of JenkinsHash [in bloomfilters]

Modified: hadoop/hbase/trunk/src/java/overview.html
URL: http://svn.apache.org/viewvc/hadoop/hbase/trunk/src/java/overview.html?rev=735783&r1=735782&r2=735783&view=diff
--- hadoop/hbase/trunk/src/java/overview.html (original)
+++ hadoop/hbase/trunk/src/java/overview.html Mon Jan 19 11:46:50 2009
@@ -39,7 +39,11 @@
   1024 on *nix systems is insufficient if you are loading any significant
   amount of data into regionservers.  See the
   <a href="http://wiki.apache.org/hadoop/Hbase/FAQ#6">FAQ: Why do I see "java.io.IOException...(Too
many open files)" in my logs?</a>
-  for how to up the limit.</li>
+  for how to up the limit.  Also, as of 0.18.x hadoop, datanodes have an upper-bound
+      on the number of threads they will support (<code>dfs.datanode.max.xcievers</code>).
+      Default is 256.  If loading lots of data into hbase, up this limit on your
+      hadoop cluster.  Also consider upping the number of datanode handlers from
+      the default of 3. See <code>dfs.datanode.handler.count</code>.</li>
       <li>The clocks on cluster members should be in basic alignments.  Some skew is
tolerable but
       wild skew can generate odd behaviors.  Run <a href="http://en.wikipedia.org/wiki/Network_Time_Protocol">NTP</a>
       on your cluster, or an equivalent.</li>

View raw message