[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-17 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.010.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch, YARN-5077.008.patch, 
> YARN-5077.009.patch, YARN-5077.010.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.009.patch

Upload patch 009 to fix the patch cannot apply issue.

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch, YARN-5077.008.patch, 
> YARN-5077.009.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.008.patch

[~kasha], thanks you very much for your comments. I uploaded patch 008 for all 
comments. 

The new patch consolidates YARN-4866, add a test case with tiny weight besides 
the zero weight, and remove useless functions. 

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch, YARN-5077.008.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: (was: YARN-5077.008.patch)

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.008.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch, YARN-5077.008.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: (was: YARN-5077.008.patch)

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-16 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.008.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch, YARN-5077.008.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-06-02 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.007.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch, YARN-5077.007.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-27 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.006.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch, 
> YARN-5077.006.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-25 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.005.patch

[~kasha], thanks for the detailed review. I uploaded patch 005 with the 
modification you suggested. 

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Karthik Kambatla
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch, YARN-5077.005.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-24 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.004.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Karthik Kambatla
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch, YARN-5077.004.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-24 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: (was: YARN-5077.004.patch)

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Karthik Kambatla
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-24 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.004.patch

[~kasha], thanks a lot for the reviewing. I agree with you completely and 
uploaded a new patch for it.

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Karthik Kambatla
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-18 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.003.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch, 
> YARN-5077.003.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-18 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: (was: YARN-5077.003.patch)

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-18 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.003.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-17 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.002.patch

The patch 002 solves the style issues. All failed tests are unrelated. 

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch, YARN-5077.002.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-17 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Attachment: YARN-5077.001.patch

> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
> Attachments: YARN-5077.001.patch
>
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-12 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Description: 
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
- root.dev 0.0 
- root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.

  was:
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
# root.dev 0.0 
# root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.


> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> - root.dev 0.0 
> - root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-12 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Description: 
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
# root.dev 0.0 
# root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.

  was:
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
# root.dev 0.0 
root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.


> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> # root.dev 0.0 
> # root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-12 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Description: 
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
# root.dev 0.0 
root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.

  was:
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
root.dev 0.0 
root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.


> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> # root.dev 0.0 
> root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

[jira] [Updated] (YARN-5077) Fix FSLeafQueue#getFairShare() for queues with weight 0.0

2016-05-12 Thread Yufei Gu (JIRA) 

 [ 
https://issues.apache.org/jira/browse/YARN-5077?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Yufei Gu updated YARN-5077:
---
Description: 
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.

For example, we have a pool with the following weights: 
root.dev 0.0 
root.product 1.0

The root.dev is a best effort pool and should only get resources if 
root.product is not running. In our tests, with no jobs running under 
root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
never start.

  was:
1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
 
2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 

In case 1), that means no container ever gets allocated for an AM because from 
the viewpoint of the RM, there is never any headroom to allocate a container on 
that queue.


> Fix FSLeafQueue#getFairShare() for queues with weight 0.0
> -
>
> Key: YARN-5077
> URL: https://issues.apache.org/jira/browse/YARN-5077
> Project: Hadoop YARN
>  Issue Type: Bug
>Reporter: Yufei Gu
>Assignee: Yufei Gu
>
> 1) When a queue's weight is set to 0.0, FSLeafQueue#getFairShare() returns 
>  
> 2) When a queue's weight is nonzero, FSLeafQueue#getFairShare() returns 
> 
> In case 1), that means no container ever gets allocated for an AM because 
> from the viewpoint of the RM, there is never any headroom to allocate a 
> container on that queue.
> For example, we have a pool with the following weights: 
> root.dev 0.0 
> root.product 1.0
> The root.dev is a best effort pool and should only get resources if 
> root.product is not running. In our tests, with no jobs running under 
> root.product, jobs started in root.dev queue stay stuck in ACCEPT phase and 
> never start.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

-
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org