[jira] [Commented] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metri
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16950008#comment-16950008 ] jiulongzhu commented on YARN-9838: -- Thanks [~Tao Yang] for suggestions and updated a new patch file: YARN-9838.0002.patch Changes done: 1.check-style warnings were fixed and run "test-patch" 2.[LeafQueue.java] fixed the indentation and restored the deletion of a blank line 3.[CapacityScheduler.java] delete the useless annotation "sync ResourceUsageByLabel ResourceUsageByUser and numContainer" 4.[TestCapacityScheduler.java] deleted "before fixed" code block ,fixed the indentation and line length errors ,added whitespace after ',' Please help to add any points if I missed something ,thank you! > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Bug > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch, YARN-9838.0002.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Commented] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metri
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16933137#comment-16933137 ] jiulongzhu commented on YARN-9838: -- Test case failure is unrelated with this patch. > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Affects Version/s: (was: 3.2.0) 2.7.3 > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Fix Version/s: (was: 3.2.0) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 3.2.0 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Fix Version/s: 3.2.0 > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3, 3.2.0 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3, 3.2.0 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Affects Version/s: 3.2.0 > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3, 3.2.0 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Affects Version/s: (was: 2.7.3) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 3.2.0 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3, 3.2.0 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: YARN-9838.0001.patch > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: (was: YARN-9838.0001.patch) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: YARN-9838.0001.patch > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: (was: YARN-9838.0001.patch) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Commented] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metri
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16931239#comment-16931239 ] jiulongzhu commented on YARN-9838: -- [~sunilg] Sorry for that, it wouldn't happen again. # I checked Hadoop-3.2.0 and YARN-5932, Neither solves the problem # Patch file name has been renamed > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > YARN-9838.0001.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: (was: bug_fix_capacityScheduler_moveApplication.patch) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Reopened] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metric
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu reopened YARN-9838: -- > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > bug_fix_capacityScheduler_moveApplication.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Commented] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metri
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16931184#comment-16931184 ] jiulongzhu commented on YARN-9838: -- I need a review.Thanks! [~genericqa] [~lavkesh] [~aaron.hadoop] [~sunilg] [~hadoopqa] > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > bug_fix_capacityScheduler_moveApplication.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Resolved] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metric
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu resolved YARN-9838. -- Resolution: Resolved > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > bug_fix_capacityScheduler_moveApplication.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Reopened] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metric
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu reopened YARN-9838: -- > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > bug_fix_capacityScheduler_moveApplication.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Attachment: (was: bug_fix_capacityScheduler_moveApplication.patch) > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, > bug_fix_capacityScheduler_moveApplication.patch > > > In some clusters of ours, we are seeing "Used Resource","Used > Capacity","Absolute Used Capacity" and "Num Container" is positive or > negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In > extreme cases, apps couldn't be submitted to the queue that is actually idle > but the "Used Resource" is far more than zero, just like "Container Leak". > Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used > Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and > "Num Container" use the "numContainer" value kept by LeafQueue.And > AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will > change the state value of "numContainer" and "Used". Secondly, by comparing > the values numContainer and ResourceUsageByLabel and QueueMetrics > changed(#allocateContainer and #releaseContainer) logic of applications with > and without "movetoqueue",i found that moving the reservedContainers didn't > modify the "numContainer" value in AbstractCSQueue and "used" value in > ResourceUsage when the application was moved from a queue to another queue. > The metric values changed logic of reservedContainers are allocated, > and moved from $FROM queue to $TO queue, and released.The degree of increase > and decrease is not conservative, the Resource allocated from $FROM queue and > release to $TO queue. > ||move reversedContainer||allocate||movetoqueue||release|| > |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the > same,$TO queue stay the same{color}|decrease in $TO queue| > |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM > queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | > |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in > $TO queue|decrease in $TO queue| > The metric values changed logic of allocatedContainer(allocated, > acquired, running) are allocated, and movetoqueue, and released are > absolutely conservative. > -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
[jira] [Updated] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
[ https://issues.apache.org/jira/browse/YARN-9838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] jiulongzhu updated YARN-9838: - Description: In some clusters of ours, we are seeing "Used Resource","Used Capacity","Absolute Used Capacity" and "Num Container" is positive or negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In extreme cases, apps couldn't be submitted to the queue that is actually idle but the "Used Resource" is far more than zero, just like "Container Leak". Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and "Num Container" use the "numContainer" value kept by LeafQueue.And AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will change the state value of "numContainer" and "Used". Secondly, by comparing the values numContainer and ResourceUsageByLabel and QueueMetrics changed(#allocateContainer and #releaseContainer) logic of applications with and without "movetoqueue",i found that moving the reservedContainers didn't modify the "numContainer" value in AbstractCSQueue and "used" value in ResourceUsage when the application was moved from a queue to another queue. The metric values changed logic of reservedContainers are allocated, and moved from $FROM queue to $TO queue, and released.The degree of increase and decrease is not conservative, the Resource allocated from $FROM queue and release to $TO queue. ||move reversedContainer||allocate||movetoqueue||release|| |numContainer|increase in $FROM queue|{color:#FF}$FROM queue stay the same,$TO queue stay the same{color}|decrease in $TO queue| |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#FF}$FROM queue stay the same,$TO queue stay the same{color}|decrease in $TO queue | |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in $TO queue|decrease in $TO queue| The metric values changed logic of allocatedContainer(allocated, acquired, running) are allocated, and movetoqueue, and released are absolutely conservative. was: In some clusters of ours, we are seeing "Used Resource","Used Capacity","Absolute Used Capacity" and "Num Container" is positive or negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In extreme cases, apps couldn't be submitted to the queue that is actually idle but the "Used Resource" is far more than zero, just like "Container Leak". Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and "Num Container" use the "numContainer" value kept by LeafQueue.And AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will change the state value of "numContainer" and "Used". Secondly, by comparing the values numContainer and ResourceUsageByLabel and QueueMetrics changed(#allocateContainer and #releaseContainer) logic of applications with and without "movetoqueue",i found that moving the reservedContainers didn't modify the "numContainer" value in AbstractCSQueue and "used" value in ResourceUsage when the application was moved from a queue to another queue. The metric values changed logic of reservedContainers are allocated, and moved from $FROM queue to $TO queue, and released.The degree of increase and decrease is not conservative, the Resource allocated from $FROM queue and release to $TO queue. ||move reversedContainer||allocate||movetoqueue||release|| |numContainer|increase in $FROM queue|{color:#66}{color:#66}$FROM queue stay the same,$TO queue stay the same{color}{color}|decrease in $TO queue| |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#66}$FROM queue stay the same,$TO queue stay the same{color}|decrease in $TO queue| |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in $TO queue|decrease in $TO queue| The metric values changed logic of allocatedContainer(allocated, acquired, running) are allocated, and movetoqueue, and released are absolutely conservative. > Using the CapacityScheduler,Apply "movetoqueue" on the application which CS > reserved containers for,will cause "Num Container" and "Used Resource" in > ResourceUsage metrics error > -- > > Key: YARN-9838 > URL: https://issues.apache.org/jira/browse/YARN-9838 > Project: Hadoop YARN > Issue Type: Improvement > Components: capacity scheduler >Affects Versions: 2.7.3 >Reporter: jiulongzhu >Priority: Critical > Labels: patch > Fix For: 2.7.3 > > Attachments:
[jira] [Created] (YARN-9838) Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics
jiulongzhu created YARN-9838: Summary: Using the CapacityScheduler,Apply "movetoqueue" on the application which CS reserved containers for,will cause "Num Container" and "Used Resource" in ResourceUsage metrics error Key: YARN-9838 URL: https://issues.apache.org/jira/browse/YARN-9838 Project: Hadoop YARN Issue Type: Improvement Components: capacity scheduler Affects Versions: 2.7.3 Reporter: jiulongzhu Fix For: 2.7.3 Attachments: RM_UI_metric_negative.png, RM_UI_metric_positive.png, bug_fix_capacityScheduler_moveApplication.patch In some clusters of ours, we are seeing "Used Resource","Used Capacity","Absolute Used Capacity" and "Num Container" is positive or negative when the queue is absolutely idle(no RUNNING, no NEW apps...).In extreme cases, apps couldn't be submitted to the queue that is actually idle but the "Used Resource" is far more than zero, just like "Container Leak". Firstly,I found that "Used Resource","Used Capacity" and "Absolute Used Capacity" use the "Used" value of ResourceUsage kept by AbstractCSQueue, and "Num Container" use the "numContainer" value kept by LeafQueue.And AbstractCSQueue#allocateResource and AbstractCSQueue#releaseResource will change the state value of "numContainer" and "Used". Secondly, by comparing the values numContainer and ResourceUsageByLabel and QueueMetrics changed(#allocateContainer and #releaseContainer) logic of applications with and without "movetoqueue",i found that moving the reservedContainers didn't modify the "numContainer" value in AbstractCSQueue and "used" value in ResourceUsage when the application was moved from a queue to another queue. The metric values changed logic of reservedContainers are allocated, and moved from $FROM queue to $TO queue, and released.The degree of increase and decrease is not conservative, the Resource allocated from $FROM queue and release to $TO queue. ||move reversedContainer||allocate||movetoqueue||release|| |numContainer|increase in $FROM queue|{color:#66}{color:#66}$FROM queue stay the same,$TO queue stay the same{color}{color}|decrease in $TO queue| |ResourceUsageByLabel(USED)|increase in $FROM queue|{color:#66}$FROM queue stay the same,$TO queue stay the same{color}|decrease in $TO queue| |QueueMetrics|increase in $FROM queue|decrease in $FROM queue, increase in $TO queue|decrease in $TO queue| The metric values changed logic of allocatedContainer(allocated, acquired, running) are allocated, and movetoqueue, and released are absolutely conservative. -- This message was sent by Atlassian Jira (v8.3.2#803003) - To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org