[jira] [Created] (YARN-501) Application Master getting killed randomly reporting excess usage of memory

Krishna Kishore Bonagiri created YARN-501: - Summary: Application Master getting killed randomly reporting excess usage of memory Key: YARN-501 URL: https://issues.apache.org/jira/browse/YARN-501

[jira] [Commented] (YARN-490) TestDistributedShell fails on Windows

[ https://issues.apache.org/jira/browse/YARN-490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610130#comment-13610130 ] Hudson commented on YARN-490: - Integrated in Hadoop-Yarn-trunk #163 (See

[jira] [Commented] (YARN-488) TestContainerManagerSecurity fails on Windows

[ https://issues.apache.org/jira/browse/YARN-488?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610133#comment-13610133 ] Hudson commented on YARN-488: - Integrated in Hadoop-Yarn-trunk #163 (See

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610141#comment-13610141 ] Hudson commented on YARN-417: - Integrated in Hadoop-Yarn-trunk #163 (See

[jira] [Commented] (YARN-490) TestDistributedShell fails on Windows

[ https://issues.apache.org/jira/browse/YARN-490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610227#comment-13610227 ] Hudson commented on YARN-490: - Integrated in Hadoop-Hdfs-trunk #1352 (See

[jira] [Commented] (YARN-491) TestContainerLogsPage fails on Windows

[ https://issues.apache.org/jira/browse/YARN-491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610228#comment-13610228 ] Hudson commented on YARN-491: - Integrated in Hadoop-Hdfs-trunk #1352 (See

[jira] [Commented] (YARN-488) TestContainerManagerSecurity fails on Windows

[ https://issues.apache.org/jira/browse/YARN-488?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610230#comment-13610230 ] Hudson commented on YARN-488: - Integrated in Hadoop-Hdfs-trunk #1352 (See

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610238#comment-13610238 ] Hudson commented on YARN-417: - Integrated in Hadoop-Hdfs-trunk #1352 (See

[jira] [Commented] (YARN-490) TestDistributedShell fails on Windows

[ https://issues.apache.org/jira/browse/YARN-490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610288#comment-13610288 ] Hudson commented on YARN-490: - Integrated in Hadoop-Mapreduce-trunk #1380 (See

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610299#comment-13610299 ] Hudson commented on YARN-417: - Integrated in Hadoop-Mapreduce-trunk #1380 (See

[jira] [Commented] (YARN-497) yarn unmanaged-am launcher jar does not define a main class in its manifest

[ https://issues.apache.org/jira/browse/YARN-497?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610349#comment-13610349 ] Hitesh Shah commented on YARN-497: -- No tests as just a pom file change.

[jira] [Updated] (YARN-493) NodeManager job control logic flaws on Windows

[ https://issues.apache.org/jira/browse/YARN-493?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Chris Nauroth updated YARN-493: --- Attachment: YARN-493.1.patch This patch addresses the bugs that I found. I've verified that the tests

[jira] [Updated] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Hitesh Shah updated YARN-498: - Summary: Unmanaged AM launcher does not set various constants in env for an AM, also does not handle

[jira] [Resolved] (YARN-501) Application Master getting killed randomly reporting excess usage of memory

[ https://issues.apache.org/jira/browse/YARN-501?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ravi Prakash resolved YARN-501. --- Resolution: Not A Problem Hi Krishna, Please set yarn.app.mapreduce.am.command-opts to include a more

[jira] [Commented] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610410#comment-13610410 ] Hadoop QA commented on YARN-498: {color:red}-1 overall{color}. Here are the results of

[jira] [Updated] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Hitesh Shah updated YARN-498: - Attachment: YARN-498.2.patch Unmanaged AM launcher does not set various constants in env for an AM,

[jira] [Commented] (YARN-493) NodeManager job control logic flaws on Windows

[ https://issues.apache.org/jira/browse/YARN-493?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610564#comment-13610564 ] Hadoop QA commented on YARN-493: {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610645#comment-13610645 ] Hadoop QA commented on YARN-498: {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (YARN-493) NodeManager job control logic flaws on Windows

[ https://issues.apache.org/jira/browse/YARN-493?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610927#comment-13610927 ] Chris Nauroth commented on YARN-493: Jenkins gave -1 for failure in mvn eclipse:eclipse.

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610944#comment-13610944 ] Chris Nauroth commented on YARN-417: We would also need to add a timeout to

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13610954#comment-13610954 ] Sandy Ryza commented on YARN-417: - Agreed about fixing the Jenkins script. In general, I

[jira] [Commented] (YARN-389) Infinitely assigning containers when the required resource exceeds the cluster's absolute capacity

[ https://issues.apache.org/jira/browse/YARN-389?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611037#comment-13611037 ] Bikas Saha commented on YARN-389: - bq. I've checked reason why the requested AM size is

[jira] [Commented] (YARN-378) ApplicationMaster retry times should be set by Client

[ https://issues.apache.org/jira/browse/YARN-378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611042#comment-13611042 ] Bikas Saha commented on YARN-378: - I have small request. If the application being submitted

[jira] [Commented] (YARN-450) Define value for * in the scheduling protocol

[ https://issues.apache.org/jira/browse/YARN-450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611054#comment-13611054 ] Bikas Saha commented on YARN-450: - Zhijie, you will need to make sure the string comparisons

[jira] [Updated] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Hitesh Shah updated YARN-498: - Attachment: YARN-498.3.patch Addressed Bikas's comments. Unmanaged AM launcher does not

[jira] [Commented] (YARN-498) Unmanaged AM launcher does not set various constants in env for an AM, also does not handle failed AMs properly

[ https://issues.apache.org/jira/browse/YARN-498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611071#comment-13611071 ] Hadoop QA commented on YARN-498: {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (YARN-494) RM should be able to hard stop a lingering app on a NM

[ https://issues.apache.org/jira/browse/YARN-494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611077#comment-13611077 ] Daryn Sharp commented on YARN-494: -- Yes, log aggregation is a yarn service provided by the

[jira] [Created] (YARN-502) RM crash with NPE on NODE_REMOVED event

Lohit Vijayarenu created YARN-502: - Summary: RM crash with NPE on NODE_REMOVED event Key: YARN-502 URL: https://issues.apache.org/jira/browse/YARN-502 Project: Hadoop YARN Issue Type: Bug

[jira] [Assigned] (YARN-193) Scheduler.normalizeRequest does not account for allocation requests that exceed maximumAllocation limits

[ https://issues.apache.org/jira/browse/YARN-193?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zhijie Shen reassigned YARN-193: Assignee: Zhijie Shen (was: Hitesh Shah) Scheduler.normalizeRequest does not account for

[jira] [Updated] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

[ https://issues.apache.org/jira/browse/YARN-467?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] omkar vinit joshi updated YARN-467: --- Attachment: yarn-467-20130322.patch Fixing directory limit problem for public cache. Creating

[jira] [Commented] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12575078/yarn-467-20130322.patch against trunk revision . {color:green}+1 @author{color}. The patch does not contain any @author tags. {color:green}+1 tests included{color}. The patch appears to include 2

[jira] [Updated] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory limit reaches before reaching the cache size and fails to create any directories in file cache (PUBLIC). The jobs start failing with the below exception

[jira] [Commented] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

Reporter: omkar vinit joshi Assignee: omkar vinit joshi Attachments: yarn-467-20130322.1.patch, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory limit reaches before reaching the cache size and fails

[jira] [Updated] (YARN-450) Define value for * in the scheduling protocol

[ https://issues.apache.org/jira/browse/YARN-450?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zhijie Shen updated YARN-450: - Attachment: YARN-450_5.patch Addressed Bikas' comments. The bug of ==/!= was fixed. In the future,

[jira] [Updated] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

-467-20130322.2.patch, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory limit reaches before reaching the cache size and fails to create any directories in file cache (PUBLIC). The jobs start failing with the below

[jira] [Commented] (YARN-450) Define value for * in the scheduling protocol

[ https://issues.apache.org/jira/browse/YARN-450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611250#comment-13611250 ] Hadoop QA commented on YARN-450: {color:green}+1 overall{color}. Here are the results of

[jira] [Updated] (YARN-71) Ensure/confirm that the NodeManager cleans up local-dirs on restart

[ https://issues.apache.org/jira/browse/YARN-71?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Xuan Gong updated YARN-71: -- Attachment: YARN-71.13.patch 1. remove the extra system.out and printStack 2. get containerState from

[jira] [Commented] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

Attachments: yarn-467-20130322.1.patch, yarn-467-20130322.2.patch, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory limit reaches before reaching the cache size and fails to create any directories in file cache (PUBLIC). The jobs

[jira] [Commented] (YARN-389) Infinitely assigning containers when the required resource exceeds the cluster's absolute capacity

[ https://issues.apache.org/jira/browse/YARN-389?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611266#comment-13611266 ] Zhijie Shen commented on YARN-389: -- @Bikas, the problem was produced by two issues: 1. The

[jira] [Commented] (YARN-71) Ensure/confirm that the NodeManager cleans up local-dirs on restart

[ https://issues.apache.org/jira/browse/YARN-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611267#comment-13611267 ] Xuan Gong commented on YARN-71: --- when I do the test, I manually create some folders under

[jira] [Commented] (YARN-470) Support a way to disable resource monitoring on the NodeManager

[ https://issues.apache.org/jira/browse/YARN-470?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611286#comment-13611286 ] Hitesh Shah commented on YARN-470: -- +1. Committing shortly. Support a way

[jira] [Updated] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

-20130322.3.patch, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory limit reaches before reaching the cache size and fails to create any directories in file cache (PUBLIC). The jobs start failing with the below exception

[jira] [Commented] (YARN-470) Support a way to disable resource monitoring on the NodeManager

[ https://issues.apache.org/jira/browse/YARN-470?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611293#comment-13611293 ] Hudson commented on YARN-470: - Integrated in Hadoop-trunk-Commit #3514 (See

[jira] [Moved] (YARN-503) DelegationTokens will be renewed forever if multiple jobs share tokens and the first one sets JOB_CANCEL_DELEGATION_TOKEN to false

[ https://issues.apache.org/jira/browse/YARN-503?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Daryn Sharp moved MAPREDUCE-3979 to YARN-503: - Component/s: (was: resourcemanager)

[jira] [Commented] (YARN-494) RM should be able to hard stop a lingering app on a NM

[ https://issues.apache.org/jira/browse/YARN-494?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611298#comment-13611298 ] Bikas Saha commented on YARN-494: - Thanks! That makes more sense. RM

[jira] [Updated] (YARN-439) Flatten NodeHeartbeatResponse

[ https://issues.apache.org/jira/browse/YARN-439?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Xuan Gong updated YARN-439: --- Attachment: YARN-439.4.patch fix the javadoc warning Flatten NodeHeartbeatResponse

[jira] [Commented] (YARN-467) Jobs fail during resource localization when directories in file cache reaches to unix directory limit for public cache

Reporter: omkar vinit joshi Assignee: omkar vinit joshi Attachments: yarn-467-20130322.1.patch, yarn-467-20130322.2.patch, yarn-467-20130322.3.patch, yarn-467-20130322.patch If we have multiple jobs which uses distributed cache with small size of files, the directory

[jira] [Updated] (YARN-503) DelegationTokens will be renewed forever if multiple jobs share tokens and the first one sets JOB_CANCEL_DELEGATION_TOKEN to false

[ https://issues.apache.org/jira/browse/YARN-503?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Daryn Sharp updated YARN-503: - Attachment: YARN-503.patch Overhaul and simplify the RM's token renewer. Apps now track their tokens, and

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611332#comment-13611332 ] Chris Nauroth commented on YARN-417: I am +1 for the addendum patch, assuming there is

[jira] [Commented] (YARN-503) DelegationTokens will be renewed forever if multiple jobs share tokens and the first one sets JOB_CANCEL_DELEGATION_TOKEN to false

[ https://issues.apache.org/jira/browse/YARN-503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611344#comment-13611344 ] Hadoop QA commented on YARN-503: {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (YARN-439) Flatten NodeHeartbeatResponse

[ https://issues.apache.org/jira/browse/YARN-439?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611354#comment-13611354 ] Hadoop QA commented on YARN-439: {color:red}-1 overall{color}. Here are the results of

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611461#comment-13611461 ] Bikas Saha commented on YARN-417: - Thanks Sandy for the fix and Chris for testing.

[jira] [Commented] (YARN-417) Create AMRMClient wrapper that provides asynchronous callbacks

[ https://issues.apache.org/jira/browse/YARN-417?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13611465#comment-13611465 ] Hudson commented on YARN-417: - Integrated in Hadoop-trunk-Commit #3515 (See