[jira] Commented: (HADOOP-6876) Path.suffix(...) on a path url with no "path" results in NullPointerException
[ https://issues.apache.org/jira/browse/HADOOP-6876?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12891390#action_12891390 ] Jordan Sissel commented on HADOOP-6876: --- Updated my own code and confirmed that my suggested workaround does work. > Path.suffix(...) on a path url with no "path" results in NullPointerException > - > > Key: HADOOP-6876 > URL: https://issues.apache.org/jira/browse/HADOOP-6876 > Project: Hadoop Common > Issue Type: Bug >Affects Versions: 0.20.2 >Reporter: Jordan Sissel > > I can most briefly demo this using the hbase jruby shell (as it has hadoop in > the classpath: > {noformat} > hbase(main):001:0> path = org.apache.hadoop.fs.Path.new("hdfs://testing/") > => # @java_object=#> > hbase(main):002:0> path.suffix("Test") > NativeException: java.lang.NullPointerException: null > {noformat} > I expected instead that path.suffix("Test") would simply generate a new path > with 'Test' appended. For above, this would be the path "hdfs://testing/Test" > Workaround: This should work - instead of using path.suffix(), use this > (java): > {noformat} >Path newpath = new Path(oldpath.toString() + "mysuffix"); > {noformat} -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Created: (HADOOP-6876) Path.suffix(...) on a path url with no "path" results in NullPointerException
Path.suffix(...) on a path url with no "path" results in NullPointerException - Key: HADOOP-6876 URL: https://issues.apache.org/jira/browse/HADOOP-6876 Project: Hadoop Common Issue Type: Bug Affects Versions: 0.20.2 Reporter: Jordan Sissel I can most briefly demo this using the hbase jruby shell (as it has hadoop in the classpath: {noformat} hbase(main):001:0> path = org.apache.hadoop.fs.Path.new("hdfs://testing/") => #> hbase(main):002:0> path.suffix("Test") NativeException: java.lang.NullPointerException: null {noformat} I expected instead that path.suffix("Test") would simply generate a new path with 'Test' appended. For above, this would be the path "hdfs://testing/Test" Workaround: This should work - instead of using path.suffix(), use this (java): {noformat} Path newpath = new Path(oldpath.toString() + "mysuffix"); {noformat} -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[jira] Created: (HADOOP-6867) Using socket address for datanode registry breaks multihoming
Using socket address for datanode registry breaks multihoming - Key: HADOOP-6867 URL: https://issues.apache.org/jira/browse/HADOOP-6867 Project: Hadoop Common Issue Type: Bug Affects Versions: 0.20.2 Environment: hadoop-0.20-0.20.2+228-1, centos 5, distcp Reporter: Jordan Sissel Related: * https://issues.apache.org/jira/browse/HADOOP-985 * https://issues.apache.org/jira/secure/attachment/12350813/HADOOP-985-1.patch * http://old.nabble.com/public-IP-for-datanode-on-EC2-td19336240.html * http://www.cloudera.com/blog/2008/12/securing-a-hadoop-cluster-through-a-gateway/ Datanodes register using their dns name (even configurable with dfs.datanode.dns.interface). However, the Namenode only really uses the source address that the registration came from when sharing it to clients wanting to write to HDFS. Specific environment that causes this problem: * Datanode and Namenode multihomed on two networks. * Datanode registers to namenode using dns name on network #1 * Client (distcp) connects to namenode on network #2 (*) and is told to write to datanodes on network #1, which doesn't work for us. (*) Allowing contact to the namenode on multiple networks was achieved with a socat proxy hack that tunnels network#2 to network#1 port 8020. This is unrelated to the issue at hand. The cloudera link above recommends proxying for other reasons than multihoming, but it would work, but it doesn't sound like it would well (bandwidth, multiplicity, multitenant, etc). Our specific scenario is wanting to distcp over a different network interface than the datanodes register themselves on, but it would be nice if both (all) interfaces worked. We are internally going to patch hadoop to roll back parts of the patch mentioned above so that we rely the datanode name rather than the socket address it uses to talk to the namenode. The alternate option is to push config changes to all nodes that force them to listen/register on one specific interface only. This helps us work around our specific problem, but doesn't really help with multihoming. I would propose that datanodes register all interface addresses during the registration/heartbeat/whatever process does this and hdfs clients would be given all addresses for a specific node to perform operations against and they could select accordingly (or 'whichever worked first') just like round-robin dns does. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.