You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "Priyo Mustafi (JIRA)" <ji...@apache.org> on 2011/01/11 19:06:47 UTC

[jira] Created: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
-----------------------------------------------------------------------------------------------------------------------------------

                 Key: MAPREDUCE-2256
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
             Project: Hadoop Map/Reduce
          Issue Type: Bug
          Components: contrib/fair-share
    Affects Versions: 0.21.1, 0.22.0
            Reporter: Priyo Mustafi
             Fix For: 0.22.0


Scenarios:
You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
1)  Pool1 schedules 300 map tasks first
2)  Pool2 then schedules another 300 map tasks
3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  

The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.

The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Matei Zaharia (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980541#action_12980541 ] 

Matei Zaharia commented on MAPREDUCE-2256:
------------------------------------------

+1 looks good to me. Let's wait for Hudson to run its tests, and I'll commit it to trunk, 0.22 and 0.21 unless anyone brings up concerns.

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Todd Lipcon (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980327#action_12980327 ] 

Todd Lipcon commented on MAPREDUCE-2256:
----------------------------------------

Looks good to me, good catch, Priyo. +1 pending test results (and should probably give Matei a day or two to take a look as well)

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Assigned: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Todd Lipcon (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Todd Lipcon reassigned MAPREDUCE-2256:
--------------------------------------

    Assignee: Priyo Mustafi

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Matei Zaharia (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981566#action_12981566 ] 

Matei Zaharia commented on MAPREDUCE-2256:
------------------------------------------

Agreed on putting it in 0.22.

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12993912#comment-12993912 ] 

Hudson commented on MAPREDUCE-2256:
-----------------------------------

Integrated in Hadoop-Mapreduce-22-branch #33 (See [https://hudson.apache.org/hudson/job/Hadoop-Mapreduce-22-branch/33/])
    

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12987968#action_12987968 ] 

Hadoop QA commented on MAPREDUCE-2256:
--------------------------------------

-1 overall.  Here are the results of testing the latest attachment 
  http://issues.apache.org/jira/secure/attachment/12468031/mapreduce-2256_0_22.txt
  against trunk revision 1064287.

    +1 @author.  The patch does not contain any @author tags.

    +1 tests included.  The patch appears to include 3 new or modified tests.

    +1 javadoc.  The javadoc tool did not generate any warning messages.

    +1 javac.  The applied patch does not increase the total number of javac compiler warnings.

    +1 findbugs.  The patch does not introduce any new Findbugs (version 1.3.9) warnings.

    +1 release audit.  The applied patch does not increase the total number of release audit warnings.

    +1 core tests.  The patch passed core unit tests.

    -1 contrib tests.  The patch failed contrib unit tests.

    +1 system test framework.  The patch passed system test framework compile.

Test results: https://hudson.apache.org/hudson/job/PreCommit-MAPREDUCE-Build/19//testReport/
Findbugs warnings: https://hudson.apache.org/hudson/job/PreCommit-MAPREDUCE-Build/19//artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Console output: https://hudson.apache.org/hudson/job/PreCommit-MAPREDUCE-Build/19//console

This message is automatically generated.

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Priyo Mustafi (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Priyo Mustafi updated MAPREDUCE-2256:
-------------------------------------

    Status: Patch Available  (was: Open)

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>             Fix For: 0.22.0
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Commented: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12989444#comment-12989444 ] 

Hudson commented on MAPREDUCE-2256:
-----------------------------------

Integrated in Hadoop-Mapreduce-trunk-Commit #595 (See [https://hudson.apache.org/hudson/job/Hadoop-Mapreduce-trunk-Commit/595/])
    

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Updated: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Priyo Mustafi (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Priyo Mustafi updated MAPREDUCE-2256:
-------------------------------------

    Attachment: mapreduce-2256_0_22.txt

Patch

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] Updated: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Konstantin Shvachko (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Konstantin Shvachko updated MAPREDUCE-2256:
-------------------------------------------

      Resolution: Fixed
    Hadoop Flags: [Reviewed]
          Status: Resolved  (was: Patch Available)

I just committed this. Thank you Priyo.

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

[jira] Updated: (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Priyo Mustafi (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Priyo Mustafi updated MAPREDUCE-2256:
-------------------------------------

    Fix Version/s: 0.22.0

This is an important bug in the FairScheduler.  I think this needs to be fixed in 0.22

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


[jira] [Commented] (MAPREDUCE-2256) FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/MAPREDUCE-2256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13016953#comment-13016953 ] 

Hudson commented on MAPREDUCE-2256:
-----------------------------------

Integrated in Hadoop-Mapreduce-trunk #643 (See [https://hudson.apache.org/hudson/job/Hadoop-Mapreduce-trunk/643/])
    

> FairScheduler fairshare preemption from multiple pools may preempt all tasks from one pool causing that pool to go below fairshare.
> -----------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-2256
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2256
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/fair-share
>    Affects Versions: 0.21.1, 0.22.0
>            Reporter: Priyo Mustafi
>            Assignee: Priyo Mustafi
>             Fix For: 0.22.0
>
>         Attachments: mapreduce-2256_0_22.txt
>
>
> Scenarios:
> You have a cluster with 600 map slots and 3 pools.  Fairshare for each pool is 200 to start with.  Fairsharepreemption timeout is 5 mins.
> 1)  Pool1 schedules 300 map tasks first
> 2)  Pool2 then schedules another 300 map tasks
> 3)  Pool3 demands 300 map tasks but doesn't get any slot as all slots are taken.
> 4)  After 5 mins pool3 should preempt 200 map-slots.  Instead of peempting 100 slots each from pool1 and pool2, the bug would cause it to preempt all 200 slots from pool2 (last started) causing it to go below fairshare.  This is happening because the preemptTask method is not reducing the tasks left from a pool while preempting the tasks.  
> The above scenario could be an extreme case but some amount of excess preemption would happen because of this bug.
> The patch I created was for 0.22.0 but the code fix should work on 0.21  as well as looks like it has the same bug.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira