You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Prabhu Joseph (JIRA)" <ji...@apache.org> on 2019/06/11 08:50:00 UTC
[jira] [Comment Edited] (YARN-9609) Nodemanager Web Service should
return logAggregationType for each file
[ https://issues.apache.org/jira/browse/YARN-9609?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16860683#comment-16860683 ]
Prabhu Joseph edited comment on YARN-9609 at 6/11/19 8:49 AM:
--------------------------------------------------------------
[~yeshavora] {{NMWebService}} groups the LOCAL files in one {{containerLogsInfo}} and groups the AGGREGATED files in another {{containerLogsInfo}}. Below is the output when both LOCAL and AGGREGATED files are present for a container. I think the below structure looks fine, showing logAggregationType for each log file will be verbose and redundant.
{code:xml}
<containerLogsInfoes>
<containerLogsInfo>
<containerLogInfo>
<fileName>prelaunch.out</fileName>
<fileSize>100</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<containerLogInfo>
<fileName>launch_container.sh</fileName>
<fileSize>5403</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<logAggregationType>LOCAL</logAggregationType>
<containerId>container_e01_1559302665385_0004_01_000002</containerId>
<nodeId>yarn-ats-1:45454</nodeId>
</containerLogsInfo>
<containerLogsInfo>
<containerLogInfo>
<fileName>auditlog</fileName>
<fileSize>123</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<containerLogInfo>
<fileName>auditlog</fileName>
<fileSize>123</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<logAggregationType>AGGREGATED</logAggregationType>
<containerId>container_e01_1559302665385_0004_01_000002</containerId>
<nodeId>yarn-ats-1:45454</nodeId>
</containerLogsInfo>
</containerLogsInfoes>
{code}
was (Author: prabhu joseph):
[~yvora] {{NMWebService}} groups the LOCAL files in one {{containerLogsInfo}} and groups the AGGREGATED files in another {{containerLogsInfo}}. Below is the output when both LOCAL and AGGREGATED files are present for a container. I think the below structure looks fine, showing logAggregationType for each log file will be verbose and redundant.
{code:xml}
<containerLogsInfoes>
<containerLogsInfo>
<containerLogInfo>
<fileName>prelaunch.out</fileName>
<fileSize>100</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<containerLogInfo>
<fileName>launch_container.sh</fileName>
<fileSize>5403</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<logAggregationType>LOCAL</logAggregationType>
<containerId>container_e01_1559302665385_0004_01_000002</containerId>
<nodeId>yarn-ats-1:45454</nodeId>
</containerLogsInfo>
<containerLogsInfo>
<containerLogInfo>
<fileName>auditlog</fileName>
<fileSize>123</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<containerLogInfo>
<fileName>auditlog</fileName>
<fileSize>123</fileSize>
<lastModifiedTime>Tue Jun 11 08:20:10 +0000 2019</lastModifiedTime>
</containerLogInfo>
<logAggregationType>AGGREGATED</logAggregationType>
<containerId>container_e01_1559302665385_0004_01_000002</containerId>
<nodeId>yarn-ats-1:45454</nodeId>
</containerLogsInfo>
</containerLogsInfoes>
{code}
> Nodemanager Web Service should return logAggregationType for each file
> ----------------------------------------------------------------------
>
> Key: YARN-9609
> URL: https://issues.apache.org/jira/browse/YARN-9609
> Project: Hadoop YARN
> Issue Type: Bug
> Components: nodemanager
> Affects Versions: 3.1.1
> Reporter: Yesha Vora
> Assignee: Prabhu Joseph
> Priority: Critical
>
> Steps:
> 1) Launch sleeper yarn service
> 2) When sleeper component is in READY state, call NM web service to list the container files and its log aggregation status.
> http://NMHost:NMPort/ws/v1/node/containers/CONTAINERID/logs
> NM web service response shows a common log aggregation type response for all files.
> Instead, NM web service should return a log aggregation type for each file.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org