[
https://issues.apache.org/jira/browse/HDFS-1568?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13036976#comment-13036976
]
Joey Echeverria commented on HDFS-1568:
---------------------------------------
> It is better to log before ERROR.write(out), otherwise, ERROR.write(out) may
> throw another exception.
Good catch, I'll fix that.
> Since the exception e is re-thrown, are you sure that it won't be printed
> twice?
You're correct. Should I just include e.toString() or e.getMessage() there?
I think there are a couple of other places where the exception might get
printed more than once. Should those be fixed as part of this jira, or filed
separately?
Finally, do you know why some stack traces are printed out by using
stringifyException() and others are printed by just passing the exception to
the LOG.error() function?
> Improve DataXceiver error logging
> ---------------------------------
>
> Key: HDFS-1568
> URL: https://issues.apache.org/jira/browse/HDFS-1568
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: data-node
> Affects Versions: 0.23.0
> Reporter: Todd Lipcon
> Assignee: Joey Echeverria
> Priority: Minor
> Labels: newbie
> Attachments: HDFS-1568-1.patch, HDFS-1568-3.patch,
> HDFS-1568-output-changes.patch
>
>
> In supporting customers we often see things like SocketTimeoutExceptions or
> EOFExceptions coming from DataXceiver, but the logging isn't very good. For
> example, if we get an IOE while setting up a connection to the downstream
> mirror in writeBlock, the IP of the downstream mirror isn't logged on the DN
> side.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
