www-infrastructure-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yongjun Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (INFRA-8097) Extra newline in /etc/hosts likely caused runtime stuck at java.net.Inet4AddressImpl.getLocalHostName(Native Method)
Date Fri, 01 Aug 2014 16:39:38 GMT

    [ https://issues.apache.org/jira/browse/INFRA-8097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14082456#comment-14082456
] 

Yongjun Zhang commented on INFRA-8097:
--------------------------------------

Thanks a lot for resolving the issue [~gkesavan]!


> Extra newline in /etc/hosts likely caused runtime stuck at java.net.Inet4AddressImpl.getLocalHostName(Native
Method)
> --------------------------------------------------------------------------------------------------------------------
>
>                 Key: INFRA-8097
>                 URL: https://issues.apache.org/jira/browse/INFRA-8097
>             Project: Infrastructure
>          Issue Type: Bug
>          Components: Jenkins
>            Reporter: Yongjun Zhang
>            Assignee: Giridharan Kesavan
>            Priority: Critical
>
> Per jenkins test failure investigation, it's identified that /etc/hosts files in some
of the jenkins test slaves have an extra newline, which likely caused runtime stuck at 
> {code}
> java.net.Inet4AddressImpl.getLocalHostName(Native Method)
> {code}
> for some of the testcases (e.g. org.apache.hadoop.ipc.TestIPC.testRetryProxy failure
reported in HADOOP-10888, and org.apache.hadoop.hdfs.server.balancer.TestBalancer.testBalancerWithNonZeroThreadsForMove
failure seen in HDFS-6588).
> Specifically, the /etc/hosts of https://builds.apache.org/computer/H0/ contains the following,
> {code}
> 127.0.0.1	localhost
>    asf900.ygridcore.net       asf900
> {code}
> which should be on the same line
> {code}
> 127.0.0.1	localhost asf900.ygridcore.net  asf900
> {code}
> Other hosts https://builds.apache.org/computer/H1/, https://builds.apache.org/computer/H2/
etc likely have same problem. Would anyone who has administrator access please help to take
a look? thanks a lot.
> Many thanks to [~arpitagarwal] for the help in HADOOP-10888 investigation and the findings
reported here. Thanks [~ted_yu] for suggesting filing this infra jira.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message