|
[
Permlink
| « Hide
]
Doug Cutting added a comment - 20/Mar/08 10:54 PM
Here's a nearly-complete patch. TestCrcCorruption fails with it, though.
The problem with TestCrcCorruption was insufficient file descriptors. This fails for me with 1024 but succeeds with 2048. Is that new?
The "running out of fds" might be related to
New version that fixes a bug in MRCaching.java.
-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12378359/HADOOP-2902.patch against trunk revision 619744. @author +1. The patch does not contain any @author tags. tests included +1. The patch appears to include 84 new or modified tests. javadoc +1. The javadoc tool did not generate any warning messages. javac +1. The applied patch does not generate any new javac compiler warnings. release audit +1. The applied patch does not generate any new release audit warnings. findbugs +1. The patch does not introduce any new Findbugs warnings. core tests -1. The patch failed core unit tests. contrib tests +1. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2023/testReport/ This message is automatically generated. MRCaching bug caused Hudson failure.
+1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12378396/HADOOP-2902.patch against trunk revision 619744. @author +1. The patch does not contain any @author tags. tests included +1. The patch appears to include 84 new or modified tests. javadoc +1. The javadoc tool did not generate any warning messages. javac +1. The applied patch does not generate any new javac compiler warnings. release audit +1. The applied patch does not generate any new release audit warnings. findbugs +1. The patch does not introduce any new Findbugs warnings. core tests +1. The patch passed core unit tests. contrib tests +1. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2026/testReport/ This message is automatically generated. Can someone please review this before I commit it?
Patch looks fine. Applied. Tests won't compile for me (r640362). Two errors of same type in same file. Here is one:
[javac] /Users/stack/Documents/checkouts/hadoop-trunk/src/test/org/apache/hadoop/mapred/TestMRServerPorts.java:103: set(java.lang.String,java.lang.String) in org.apache.hadoop.conf.Configuration cannot be applied to (java.lang.String,java.net.URI) [javac] conf2.set("mapred.job.tracker", You need to toString it? Couple of minor items. Wouldn't put them in way of a commit. Regards: + if (namenode.startsWith("hdfs://")) ...isn't there a define for "hdfs://" you could use or a method you could pass a URI to under dfs package that would tell you if its a HDFS URI? (You do "hdfs://" multiple times in TestCopyFiles. Same for the string "file:///"). Are you toStringing a URI and then creating a URI in the following code from TestCopyFiles? + namenode = FileSystem.getDefaultUri(conf).toString(); + if (namenode.startsWith("hdfs://")) { + MyFile[] files = createFiles(URI.create(namenode), "/srcdat"); > Tests won't compile for me (r640362).
Yes, trunk has changed. I'll generate an updated patch. Update patch for changes to trunk.
> You do "hdfs://" multiple times in TestCopyFiles. Same for the string "file:///"
There were lots of those before too. I think there are actually fewer now, just in different places! I don't think the situation in TestCopyFiles is made any worse. > Are you toStringing a URI and then creating a URI in the following code from TestCopyFiles? Yes. The namenode variable is used frequently as a String in this file. Changing it to a URI would be a substantial rewrite. +1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12378572/HADOOP-2902.patch against trunk revision 619744. @author +1. The patch does not contain any @author tags. tests included +1. The patch appears to include 84 new or modified tests. javadoc +1. The javadoc tool did not generate any warning messages. javac +1. The applied patch does not generate any new javac compiler warnings. release audit +1. The applied patch does not generate any new release audit warnings. findbugs +1. The patch does not introduce any new Findbugs warnings. core tests +1. The patch passed core unit tests. contrib tests +1. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/2052/testReport/ This message is automatically generated. Integrated in Hadoop-trunk #443 (See http://hudson.zones.apache.org/hudson/job/Hadoop-trunk/443/
This patch has caused org.apache.hadoop.mapred.TestMiniMRLocalFS.testWithLocal to fail since it was committed. See
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||