[ 
https://issues.apache.org/jira/browse/HADOOP-4659?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12648600#action_12648600
 ] 

Steve Loughran commented on HADOOP-4659:
----------------------------------------

The two patches in here do slightly different things, and need to be merged. 

-my one left the class of an exception alone (and, for ConnectExceptions, 
inserted the host and port into the text, as with HADOOP-3844.
-Hairong's stopped some exceptions getting swallowed during setupIOstreams

I think both are needed, so will apply Hairong's to my code and generate a 
combined patch. 

I can test this by deploying an orphan task tracker, but in that situation, 
once the code is fixed, the TaskTracker will spin forever. If a timeout on the 
retries could be provided, we could add a test that verified the tracker ran 
for 20-30s before timing out and relaying the exception. In production you'd 
set the timeout to a number of hours or forever, obviously.

> Root cause of connection failure is being lost to code that uses it for 
> delaying startup
> ----------------------------------------------------------------------------------------
>
>                 Key: HADOOP-4659
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4659
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: ipc
>    Affects Versions: 0.18.3
>            Reporter: Steve Loughran
>            Assignee: Steve Loughran
>            Priority: Blocker
>             Fix For: 0.18.3
>
>         Attachments: connectRetry.patch, hadoop-4659.patch
>
>
> ipc.Client the root cause of a connection failure is being lost as the 
> exception is wrapped, hence the outside code, the one that looks for that 
> root cause, isn't working as expected. The results is you can't bring up a 
> task tracker before job tracker, and probably the same for a datanode before 
> a  namenode. The change that triggered this is not yet located, I had thought 
> it was HADOOP-3844 but I no longer believe this is the case.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to