You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-user@hadoop.apache.org by Boyu Zhang <bo...@gmail.com> on 2015/10/01 16:03:48 UTC
Finished Jobs web UI
Hello users,
I have the job history server daemon running, and disabled the history
cleaner.
I enabled the yarn log aggregation, with the following setting:
<property>
<name>yarn.log.server.url</name>
<value>http://xxxx:19888/jobhistory/logs/</value>
</property>
<property>
<name>yarn.log-aggregation-enable</name>
<value>true</value>
</property>
The settings in mapred-site.xml regarding job history server is default.
Now that I see the finished job logs on hdfs:///tmp/logs/user/appID, but is
there a way to see that through the web UI? Accessing this:
http://xxxx:19888/jobhistory/logs/ from safari give me the following error:
Cannot get container logs without a ContainerId
Cannot get container logs without a NodeId
Cannot get container logs without an app owner
Any suggestions/pointers would be greatly appreciated, thank you!!
Boyu
Re: Finished Jobs web UI
Posted by Varun Saxena <vs...@gmail.com>.
This configuration is used for redirection by NM Web UI.
NM will append container id, node id and app owner depending on the
container whose logs you want to see.
Regards,
Varun Saxena.
On Thu, Oct 1, 2015 at 7:33 PM, Boyu Zhang <bo...@gmail.com> wrote:
> Hello users,
>
> I have the job history server daemon running, and disabled the history
> cleaner.
>
> I enabled the yarn log aggregation, with the following setting:
>
> <property>
> <name>yarn.log.server.url</name>
> <value>http://xxxx:19888/jobhistory/logs/</value>
> </property>
> <property>
> <name>yarn.log-aggregation-enable</name>
> <value>true</value>
> </property>
>
> The settings in mapred-site.xml regarding job history server is default.
>
> Now that I see the finished job logs on hdfs:///tmp/logs/user/appID, but
> is there a way to see that through the web UI? Accessing this:
> http://xxxx:19888/jobhistory/logs/ from safari give me the following
> error:
>
> Cannot get container logs without a ContainerId
> Cannot get container logs without a NodeId
> Cannot get container logs without an app owner
>
> Any suggestions/pointers would be greatly appreciated, thank you!!
>
> Boyu
>
>
>
>
>
Re: Finished Jobs web UI
Posted by Varun Saxena <vs...@gmail.com>.
This configuration is used for redirection by NM Web UI.
NM will append container id, node id and app owner depending on the
container whose logs you want to see.
Regards,
Varun Saxena.
On Thu, Oct 1, 2015 at 7:33 PM, Boyu Zhang <bo...@gmail.com> wrote:
> Hello users,
>
> I have the job history server daemon running, and disabled the history
> cleaner.
>
> I enabled the yarn log aggregation, with the following setting:
>
> <property>
> <name>yarn.log.server.url</name>
> <value>http://xxxx:19888/jobhistory/logs/</value>
> </property>
> <property>
> <name>yarn.log-aggregation-enable</name>
> <value>true</value>
> </property>
>
> The settings in mapred-site.xml regarding job history server is default.
>
> Now that I see the finished job logs on hdfs:///tmp/logs/user/appID, but
> is there a way to see that through the web UI? Accessing this:
> http://xxxx:19888/jobhistory/logs/ from safari give me the following
> error:
>
> Cannot get container logs without a ContainerId
> Cannot get container logs without a NodeId
> Cannot get container logs without an app owner
>
> Any suggestions/pointers would be greatly appreciated, thank you!!
>
> Boyu
>
>
>
>
>
Re: Finished Jobs web UI
Posted by Varun Saxena <vs...@gmail.com>.
This configuration is used for redirection by NM Web UI.
NM will append container id, node id and app owner depending on the
container whose logs you want to see.
Regards,
Varun Saxena.
On Thu, Oct 1, 2015 at 7:33 PM, Boyu Zhang <bo...@gmail.com> wrote:
> Hello users,
>
> I have the job history server daemon running, and disabled the history
> cleaner.
>
> I enabled the yarn log aggregation, with the following setting:
>
> <property>
> <name>yarn.log.server.url</name>
> <value>http://xxxx:19888/jobhistory/logs/</value>
> </property>
> <property>
> <name>yarn.log-aggregation-enable</name>
> <value>true</value>
> </property>
>
> The settings in mapred-site.xml regarding job history server is default.
>
> Now that I see the finished job logs on hdfs:///tmp/logs/user/appID, but
> is there a way to see that through the web UI? Accessing this:
> http://xxxx:19888/jobhistory/logs/ from safari give me the following
> error:
>
> Cannot get container logs without a ContainerId
> Cannot get container logs without a NodeId
> Cannot get container logs without an app owner
>
> Any suggestions/pointers would be greatly appreciated, thank you!!
>
> Boyu
>
>
>
>
>
Re: Finished Jobs web UI
Posted by Varun Saxena <vs...@gmail.com>.
This configuration is used for redirection by NM Web UI.
NM will append container id, node id and app owner depending on the
container whose logs you want to see.
Regards,
Varun Saxena.
On Thu, Oct 1, 2015 at 7:33 PM, Boyu Zhang <bo...@gmail.com> wrote:
> Hello users,
>
> I have the job history server daemon running, and disabled the history
> cleaner.
>
> I enabled the yarn log aggregation, with the following setting:
>
> <property>
> <name>yarn.log.server.url</name>
> <value>http://xxxx:19888/jobhistory/logs/</value>
> </property>
> <property>
> <name>yarn.log-aggregation-enable</name>
> <value>true</value>
> </property>
>
> The settings in mapred-site.xml regarding job history server is default.
>
> Now that I see the finished job logs on hdfs:///tmp/logs/user/appID, but
> is there a way to see that through the web UI? Accessing this:
> http://xxxx:19888/jobhistory/logs/ from safari give me the following
> error:
>
> Cannot get container logs without a ContainerId
> Cannot get container logs without a NodeId
> Cannot get container logs without an app owner
>
> Any suggestions/pointers would be greatly appreciated, thank you!!
>
> Boyu
>
>
>
>
>