[
https://issues.apache.org/jira/browse/HADOOP-3530?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Konstantin Shvachko updated HADOOP-3530:
----------------------------------------
Priority: Critical (was: Blocker)
Fix Version/s: (was: 0.18.0)
0.19.0
I agree when initialize() fails and then consequently close() throws a runtime
exception we will get 2 dfiferent messages.
This is the case when the system cannot neither start nor exit.
But if close succeeds doubling the message is confusing.
I am looking at the close code in general and it looks like we will need to
clean up handling of exceptions in general.
So this is not a one line fix, I am reassigning it to 0.19.
> Redundant error logging in FSNamesystem.
> ----------------------------------------
>
> Key: HADOOP-3530
> URL: https://issues.apache.org/jira/browse/HADOOP-3530
> Project: Hadoop Core
> Issue Type: Bug
> Components: dfs
> Affects Versions: 0.18.0
> Reporter: Konstantin Shvachko
> Assignee: Tsz Wo (Nicholas), SZE
> Priority: Critical
> Fix For: 0.19.0
>
>
> HADOOP-3509 introduced a new error message which is logged if FSNamesystem
> fails to initialize.
> This message turned out to be redundant, because it should and is logged
> further up the stack say in NameNode.main().
> Right now I see the exception logged twice when I run a stand-alone name-node.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
