You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Fabian Deutsch (JIRA)" <ji...@apache.org> on 2008/08/04 13:30:45 UTC
[jira] Created: (HADOOP-3895) namenode fails to run on ppc
namenode fails to run on ppc
----------------------------
Key: HADOOP-3895
URL: https://issues.apache.org/jira/browse/HADOOP-3895
Project: Hadoop Core
Issue Type: Bug
Affects Versions: 0.17.1
Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
Reporter: Fabian Deutsch
Priority: Minor
Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: hadoop-site.xml
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Raghu Angadi (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619918#action_12619918 ]
Raghu Angadi commented on HADOOP-3895:
--------------------------------------
This mostly not because of mismatch since this is running inside the NameNode it wants to contact. Output of '-QUIT' seems to show that RPC server in NameNode is stuck in side Seletctor, which is unusual. We could take thread stacks multiple times with 'jstack'. Do unit tests under core/io and core/net pass?
{noformat}
"IPC Server listener on 9000" daemon prio=10 tid=0x1006d570 nid=0x8d5 runnable [0x8e3c2000..0x8e3c2ae0]
java.lang.Thread.State: RUNNABLE
at java.util.HashMap.keySet(HashMap.java:886)
at java.util.HashSet.iterator(HashSet.java:170)
at sun.nio.ch.SelectorImpl.processDeregisterQueue(SelectorImpl.java:145)
- locked <0x56d679d8> (a java.util.HashSet)
at sun.nio.ch.EPollSelectorImpl.doSelect(EPollSelectorImpl.java:87)
at sun.nio.ch.SelectorImpl.lockAndDoSelect(SelectorImpl.java:87)
- locked <0x56d61c20> (a sun.nio.ch.Util$1)
- locked <0x56d61c30> (a java.util.Collections$UnmodifiableSet)
- locked <0x56d61be0> (a sun.nio.ch.EPollSelectorImpl)
at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:98)
at sun.nio.ch.SelectorImpl.select(SelectorImpl.java:102)
at org.apache.hadoop.ipc.Server$Listener.run(Server.java:253)
{noformat}
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619599#action_12619599 ]
Fabian Deutsch commented on HADOOP-3895:
----------------------------------------
just replaced hadoop-env.sh with correct one. i wasn't using openjdk but gcj.
when using openjdk hadoop dies after a couple of seconds, quietly.
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: build.log
# ant -l build.log -Dtest.output=yes test
Killed after a couple of minutes.
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: build.log, hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: jdb-namenode-QUIT.log
jdb session of a running namenode, which receives a kill -QUIT
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: (was: hadoop-env.sh)
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Steve Loughran (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619506#action_12619506 ]
Steve Loughran commented on HADOOP-3895:
----------------------------------------
what stack trace does a kill -QUIT on the namenode show up?
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Robert Chansler (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Chansler updated HADOOP-3895:
------------------------------------
Component/s: dfs
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Components: dfs
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: build.log, hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log, netstat.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Steve Loughran (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619824#action_12619824 ]
Steve Loughran commented on HADOOP-3895:
----------------------------------------
The last stack trace shows that main() is waiting for a response from the server about filesystem versions; if it quits after this then it may be that the versions of the different processes are inconsistent. Nothing else looks busy
at java.lang.Object.wait(Native Method)
- waiting on <0x56d716d8> (a org.apache.hadoop.ipc.Client$Call)
at org.apache.hadoop.ipc.Client.call(Client.java:465)
- locked <0x56d716d8> (a org.apache.hadoop.ipc.Client$Call)
at org.apache.hadoop.ipc.RPC$Invoker.invoke(RPC.java:163)
at org.apache.hadoop.dfs.$Proxy0.getProtocolVersion(Unknown Source)
at org.apache.hadoop.ipc.RPC.getProxy(RPC.java:247)
at org.apache.hadoop.dfs.DFSClient.<init>(DFSClient.java:105)
at org.apache.hadoop.dfs.DistributedFileSystem$RawDistributedFileSystem.initialize(DistributedFileSystem.java:67)
at org.apache.hadoop.fs.FilterFileSystem.initialize(FilterFileSystem.java:57)
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:160)
- locked <0x88f65668> (a java.lang.Class for org.apache.hadoop.fs.FileSystem)
at org.apache.hadoop.fs.FileSystem.getNamed(FileSystem.java:119)
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:91)
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:144)
- locked <0x88f65668> (a java.lang.Class for org.apache.hadoop.fs.FileSystem)
at org.apache.hadoop.fs.Path.getFileSystem(Path.java:159)
at org.apache.hadoop.fs.Trash.<init>(Trash.java:58)
at org.apache.hadoop.dfs.NameNode.init(NameNode.java:184)
at org.apache.hadoop.dfs.NameNode.<init>(NameNode.java:195)
at org.apache.hadoop.dfs.NameNode.main(NameNode.java:728)
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12620257#action_12620257 ]
Fabian Deutsch commented on HADOOP-3895:
----------------------------------------
During the following lines, java is hogging on the CPU:
{noformat}
[junit] 2008-08-06 15:12:41,355 INFO ipc.Client (Client.java:setupIOstreams(190)) - Retrying connect to server: power.local/127.0.0.1:44005. Already tried 1 time(s).
[junit] 2008-08-06 15:13:44,390 INFO ipc.Client (Client.java:setupIOstreams(190)) - Retrying connect to server: power.local/127.0.0.1:44005. Already tried 2 time(s).
[junit] 2008-08-06 15:14:47,439 INFO ipc.Client (Client.java:setupIOstreams(190)) - Retrying connect to server: power.local/127.0.0.1:44005. Already tried 3 time(s).
[junit] 2008-08-06 15:15:50,225 INFO ipc.Client (Client.java:setupIOstreams(190)) - Retrying connect to server: power.local/127.0.0.1:44005. Already tried 4 time(s).
[junit] 2008-08-06 15:16:53,196 INFO ipc.Client (Client.java:setupIOstreams(190)) - Retrying connect to server: power.local/127.0.0.1:44005. Already tried 5 time(s).
{noformat}
I also attached a log of netstat. Interesting might be, that there is only a process bound to port 44005 on an ipv6 address.
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: build.log, hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: hadoop-env.sh
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: java.hprof.txt
HPROF data.
hadoop dies automagically after a while.
It seems as if the problem is related to some socket/ipc/rpc things.
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Description:
Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
was:
Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0 (was: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6)
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml, jdb-namenode-QUIT.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: netstat.log
netstat during test
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and gcj-1.5.0.0
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: build.log, hadoop-env.sh, hadoop-site.xml, java.hprof.txt, jdb-namenode-QUIT.log, netstat.log
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and gcj-1.5.0.0 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619612#action_12619612 ]
Fabian Deutsch commented on HADOOP-3895:
----------------------------------------
I don't know about a different way on debuging this, so I tried strace.
It hangs at "getdents64(3, ..." all the 180seconds.
{noformat}
[fabiand@power search]$ rm logs/* ; strace bin/hadoop namenode & sleep 180 ; kill -QUIT $!
[1] 1907
...
open("/home/fabiand/work/nutch/search/bin/../", O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_DIRECTORY|0x80000) = 3
fstat64(3, {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 3
fcntl64(3, F_GETFD) = 0x3 (flags FD_CLOEXEC|0x2)
getdents64(3,
[fabiand@power search]$
{noformat}
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12619603#action_12619603 ]
Fabian Deutsch commented on HADOOP-3895:
----------------------------------------
I get nothing.
This is what I did:
{noformat}
[fabiand@power search]$ rm logs/*
[fabiand@power search]$ bin/hadoop namenode & sleep 180 ; kill -QUIT $!
[1] 1861
[fabiand@power search]$ cat logs/*
2008-08-04 19:28:52,319 INFO util.Credential - Checking Resource aliases
2008-08-04 19:28:52,499 INFO http.HttpServer - Version Jetty/5.1.4
2008-08-04 19:28:52,509 INFO util.Container - Started HttpContext[/logs,/logs]
2008-08-04 19:28:55,903 INFO util.Container - Started org.mortbay.jetty.servlet.WebApplicationHandler@491f4f80
2008-08-04 19:28:56,222 INFO util.Container - Started WebApplicationContext[/,/]
2008-08-04 19:28:56,223 INFO util.Container - Started HttpContext[/static,/static]
2008-08-04 19:28:56,226 INFO http.SocketListener - Started SocketListener on 0.0.0.0:50070
2008-08-04 19:28:56,226 INFO util.Container - Started org.mortbay.jetty.Server@49623b00
[1]+ Verlassen bin/hadoop namenode
[fabiand@power search]$
{noformat}
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HADOOP-3895) namenode fails to run on ppc
Posted by "Fabian Deutsch (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3895?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Fabian Deutsch updated HADOOP-3895:
-----------------------------------
Attachment: hadoop-env.sh
> namenode fails to run on ppc
> ----------------------------
>
> Key: HADOOP-3895
> URL: https://issues.apache.org/jira/browse/HADOOP-3895
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.17.1
> Environment: PowerPC using Fedora 9 (all updates) and OpenJDK-1.6
> Reporter: Fabian Deutsch
> Priority: Minor
> Attachments: hadoop-env.sh, hadoop-env.sh, hadoop-site.xml
>
>
> Hadoop starts, but eats 100% CPU. Data- and Secondarynamenodes can not connect. No jobs were run, just trying to start the daemon. using bin/start-dfs.sh.
> Using the same simple configuration on an x86-arch - also using Fedora 9 and OpenJDK-1.6 - works perfectly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.