[jira] [Commented] (YARN-3364) Clarify Naming of yarn.client.nodemanager-connect.max-wait-ms and yarn.resourcemanager.connect.max-wait.ms
[ https://issues.apache.org/jira/browse/YARN-3364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14367167#comment-14367167 ] Andrew Johnson commented on YARN-3364: -- No, I did not have YARN-3238 applied. Thanks for that! Given that and HADOOP-11398 I think this can can be closed. Clarify Naming of yarn.client.nodemanager-connect.max-wait-ms and yarn.resourcemanager.connect.max-wait.ms --- Key: YARN-3364 URL: https://issues.apache.org/jira/browse/YARN-3364 Project: Hadoop YARN Issue Type: Improvement Components: yarn Reporter: Andrew Johnson I encountered an issue recently where the ApplicationMaster for MapReduce jobs would spend hours attempting to connect to a node in my cluster that had died due to a hardware fault. After debugging this, I found that the yarn.client.nodemanager-connect.max-wait-ms property did not behave as I had expected. Based on the name I had thought this would set a maximum time limit for attempting to connect to a NodeManager. The code in org.apache.hadoop.yarn.client.NMProxy corroborated this thought - it used a RetryUpToMaximumTimeWithFixedSleep policy when a ConnectTimeoutException was thrown, as it was in my case with a dead node. However, the RetryUpToMaximumTimeWithFixedSleep policy doesn't actually set a time limit, but instead divides the maximum time by the sleep period to set a total number of retries, regardless of how long those retries take. As such I was seeing the ApplicationMaster spend much longer attempting to make a connection than I had anticipated. The yarn.resourcemanager.connect.max-wait.ms would have the same behavior. These properties would be better named like yarn.client.nodemanager-connect.max.retries and yarn.resourcemanager.connect.max.retries to better align with the actual behavior. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (YARN-3364) Clarify Naming of yarn.client.nodemanager-connect.max-wait-ms and yarn.resourcemanager.connect.max-wait.ms
Andrew Johnson created YARN-3364: Summary: Clarify Naming of yarn.client.nodemanager-connect.max-wait-ms and yarn.resourcemanager.connect.max-wait.ms Key: YARN-3364 URL: https://issues.apache.org/jira/browse/YARN-3364 Project: Hadoop YARN Issue Type: Improvement Components: yarn Reporter: Andrew Johnson I encountered an issue recently where the ApplicationMaster for MapReduce jobs would spend hours attempting to connect to a node in my cluster that had died due to a hardware fault. After debugging this, I found that the yarn.client.nodemanager-connect.max-wait-ms property did not behave as I had expected. Based on the name I had thought this would set a maximum time limit for attempting to connect to a NodeManager. The code in org.apache.hadoop.yarn.client.NMProxy corroborated this thought - it used a RetryUpToMaximumTimeWithFixedSleep policy when a ConnectTimeoutException was thrown, as it was in my case with a dead node. However, the RetryUpToMaximumTimeWithFixedSleep policy doesn't actually set a time limit, but instead divides the maximum time by the sleep period to set a total number of retries, regardless of how long those retries take. As such I was seeing the ApplicationMaster spend much longer attempting to make a connection than I had anticipated. The yarn.resourcemanager.connect.max-wait.ms would have the same behavior. These properties would be better named like yarn.client.nodemanager-connect.max.retries and yarn.resourcemanager.connect.max.retries to better align with the actual behavior. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2419) RM applications page doesn't sort application id properly
[ https://issues.apache.org/jira/browse/YARN-2419?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14278696#comment-14278696 ] Andrew Johnson commented on YARN-2419: -- I am encountering this same problem. Is there a fix in the works? RM applications page doesn't sort application id properly - Key: YARN-2419 URL: https://issues.apache.org/jira/browse/YARN-2419 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Thomas Graves The ResourceManager apps page doesn't sort the application ids properly when the app id rolls over from to 1. When it rolls over the 1+ application ids end up being many pages down by the 0XXX numbers. I assume we just sort alphabetically so we would need a special sorter that knows about application ids. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14269373#comment-14269373 ] Andrew Johnson commented on YARN-2893: -- Yeah, that definitely seems like its worth a look. Is there anything specific I should look out for? AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14268449#comment-14268449 ] Andrew Johnson commented on YARN-2893: -- I'm seeing this error on a non-secure cluster. AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14268193#comment-14268193 ] Andrew Johnson commented on YARN-2893: -- I've also noticed that if multiple jobs are submitted at the same time and this error occurs, all the jobs will fail. AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14268183#comment-14268183 ] Andrew Johnson commented on YARN-2893: -- I am also encountering this same error. The failures are pretty sporadic and I've never been able to reproduce it. Resubmitting the failed job always works, however. AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14268359#comment-14268359 ] Andrew Johnson commented on YARN-2893: -- No, it's at least 95% Scalding jobs. AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Commented] (YARN-2893) AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream
[ https://issues.apache.org/jira/browse/YARN-2893?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14268340#comment-14268340 ] Andrew Johnson commented on YARN-2893: -- [~jira.shegalov] This is always with Scalding jobs. AMLaucher: sporadic job failures due to EOFException in readTokenStorageStream -- Key: YARN-2893 URL: https://issues.apache.org/jira/browse/YARN-2893 Project: Hadoop YARN Issue Type: Bug Components: resourcemanager Affects Versions: 2.4.0 Reporter: Gera Shegalov MapReduce jobs on our clusters experience sporadic failures due to corrupt tokens in the AM launch context. -- This message was sent by Atlassian JIRA (v6.3.4#6332)