hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Loughran (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-10036) RetryInvocationHandler should recognise that there is no point retrying to auth failures
Date Wed, 09 Oct 2013 15:36:43 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-10036?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13790524#comment-13790524
] 

Steve Loughran commented on HADOOP-10036:
-----------------------------------------

... better: have {{SaslRpcClient.getServerPrincipal()}} raise a specific exception for for
an invalid (and irreconcilable) auth configuration problem which retrying will have no effect
whatsoever -there are four possible causes of this.

> RetryInvocationHandler should recognise that there is no point retrying to auth failures
> ----------------------------------------------------------------------------------------
>
>                 Key: HADOOP-10036
>                 URL: https://issues.apache.org/jira/browse/HADOOP-10036
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: ipc
>    Affects Versions: 2.2.1
>            Reporter: Steve Loughran
>            Priority: Minor
>
> The {{RetryInvocationHandler}} tries to retry connections, so as to handle transient
failures. 
> However, auth failures aren't treated as special, so it spins even though the operation
will not succeed with the current configuration



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message