You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Vinod K V (JIRA)" <ji...@apache.org> on 2009/05/19 10:13:45 UTC
[jira] Created: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
MiniMRCluster should point TaskTracker's http URL to the loopback address.
--------------------------------------------------------------------------
Key: HADOOP-5871
URL: https://issues.apache.org/jira/browse/HADOOP-5871
Project: Hadoop Core
Issue Type: Bug
Components: mapred
Reporter: Vinod K V
It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
Posted by "Todd Lipcon (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-5871?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Todd Lipcon resolved HADOOP-5871.
---------------------------------
Resolution: Duplicate
> MiniMRCluster should point TaskTracker's http URL to the loopback address.
> --------------------------------------------------------------------------
>
> Key: HADOOP-5871
> URL: https://issues.apache.org/jira/browse/HADOOP-5871
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Vinod K V
> Attachments: HADOOP-5871.txt
>
>
> It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
Posted by "Vinod K V (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-5871?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Vinod K V updated HADOOP-5871:
------------------------------
Attachment: HADOOP-5871.txt
Attaching the simple straight-forward patch. Though I don't want to believe it myself, it reduced test time of TestClusterMapReduce by a whopping 40% on my local box. Need to verify it on some other machine.
> MiniMRCluster should point TaskTracker's http URL to the loopback address.
> --------------------------------------------------------------------------
>
> Key: HADOOP-5871
> URL: https://issues.apache.org/jira/browse/HADOOP-5871
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Vinod K V
> Attachments: HADOOP-5871.txt
>
>
> It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
Posted by "Vinod K V (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-5871?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12710674#action_12710674 ]
Vinod K V commented on HADOOP-5871:
-----------------------------------
The large differences seem to be because of ipv6 enabled on my machine.
> MiniMRCluster should point TaskTracker's http URL to the loopback address.
> --------------------------------------------------------------------------
>
> Key: HADOOP-5871
> URL: https://issues.apache.org/jira/browse/HADOOP-5871
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Vinod K V
> Attachments: HADOOP-5871.txt
>
>
> It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
Posted by "Todd Lipcon (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-5871?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12712830#action_12712830 ]
Todd Lipcon commented on HADOOP-5871:
-------------------------------------
Looks like this is subsumed by HADOOP-5888 (fixes this as well as several other instances of 0.0.0.0)
> MiniMRCluster should point TaskTracker's http URL to the loopback address.
> --------------------------------------------------------------------------
>
> Key: HADOOP-5871
> URL: https://issues.apache.org/jira/browse/HADOOP-5871
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Vinod K V
> Attachments: HADOOP-5871.txt
>
>
> It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-5871) MiniMRCluster should point
TaskTracker's http URL to the loopback address.
Posted by "Steve Loughran (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-5871?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12710700#action_12710700 ]
Steve Loughran commented on HADOOP-5871:
----------------------------------------
I am +1 to this, because you can't turn ipv6 off the latest ubuntu release.
http://www.1060.org/blogxter/entry?publicid=DEDC82E94310CA8318C5E4E83DE3C70D
> MiniMRCluster should point TaskTracker's http URL to the loopback address.
> --------------------------------------------------------------------------
>
> Key: HADOOP-5871
> URL: https://issues.apache.org/jira/browse/HADOOP-5871
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Vinod K V
> Attachments: HADOOP-5871.txt
>
>
> It currently sets mapred.task.tracker.http.address to 0.0.0.0:0 and so every TT startup in MiniMRCluster makes a trip to DNS. Changing it to 127.0.0.1:0 decreases 5 seconds for every TT start-up on my local box.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.