[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15406477#comment-15406477 ] Hudson commented on YARN-4280: -- SUCCESS: Integrated in Hadoop-trunk-Commit #10205 (See

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15406467#comment-15406467 ] Kuhu Shukla commented on YARN-4280: --- Thank you all for the detailed reviews. > CapacityScheduler

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15406411#comment-15406411 ] Jason Lowe commented on YARN-4280: -- +1 for the branch-2.8 patch as well. Committing this. >

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15403112#comment-15403112 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15402490#comment-15402490 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15399958#comment-15399958 ] Jason Lowe commented on YARN-4280: -- +1 for the latest patch. I'll commit this sometime next week unless

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15398671#comment-15398671 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15398328#comment-15398328 ] Jason Lowe commented on YARN-4280: -- Thanks for updating the patch! The parent queue code is now

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15397714#comment-15397714 ] Kuhu Shukla commented on YARN-4280: --- [~jlowe], Request for comments on latest patch. Thanks a lot! >

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15397678#comment-15397678 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15397582#comment-15397582 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15397148#comment-15397148 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15396743#comment-15396743 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15396200#comment-15396200 ] Jason Lowe commented on YARN-4280: -- Thanks for updating the patch, Kuhu! The copy constructor for

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15393794#comment-15393794 ] Kuhu Shukla commented on YARN-4280: --- Test failures are known and unrelated. Checkstyle warning is about

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15393173#comment-15393173 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392975#comment-15392975 ] Kuhu Shukla commented on YARN-4280: --- Test failures are known and unrelated. Checkstyle warning is about

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392970#comment-15392970 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392553#comment-15392553 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392545#comment-15392545 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392378#comment-15392378 ] Kuhu Shukla commented on YARN-4280: --- Thanks a lot [~leftnoteasy]. The branch 2.8 patch is attached as :

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392371#comment-15392371 ] Wangda Tan commented on YARN-4280: -- Thanks [~kshukla], patch LGTM. But it seems the patch doesn't apply to

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392300#comment-15392300 ] Kuhu Shukla commented on YARN-4280: --- [~leftnoteasy], [~jlowe], Precommit for trunk looks fine now.

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392266#comment-15392266 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15392120#comment-15392120 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15390429#comment-15390429 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15390409#comment-15390409 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15390406#comment-15390406 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15390363#comment-15390363 ] Wangda Tan commented on YARN-4280: -- Thanks [~kshukla], rekicked both. > CapacityScheduler reservations

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15390195#comment-15390195 ] Kuhu Shukla commented on YARN-4280: --- Test failures seem unrelated for the 2.8 patch precommit.

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15388544#comment-15388544 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15386574#comment-15386574 ] Wangda Tan commented on YARN-4280: -- Looks good to me, thanks [~kshukla]. The patch needs a minor rebase

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15364647#comment-15364647 ] Kuhu Shukla commented on YARN-4280: --- {{TestRMRestart.testQueueMetricsOnRMRestart}} is tracked through

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15364583#comment-15364583 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15351537#comment-15351537 ] Wangda Tan commented on YARN-4280: -- If I understand correctly, current logic is: For Application, if it

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15348222#comment-15348222 ] Kuhu Shukla commented on YARN-4280: --- Requesting [~jlowe], [~leftnoteasy], [~sunilg] for more

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15347653#comment-15347653 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15347320#comment-15347320 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15347296#comment-15347296 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15345623#comment-15345623 ] Sunil G commented on YARN-4280: --- bq.So yes, we need the other queues to stop allocating until the

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15345266#comment-15345266 ] Jason Lowe commented on YARN-4280: -- bq. IIRC the headroom is a combination of the user limits and the

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15345203#comment-15345203 ] Jason Lowe commented on YARN-4280: -- Thanks for updating the patch, Kuhu! I'm confused by the

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15344888#comment-15344888 ] Sunil G commented on YARN-4280: --- Thanks [~kshukla] for the patch and its an interesting problem. I have one

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15344626#comment-15344626 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15343702#comment-15343702 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15343248#comment-15343248 ] Kuhu Shukla commented on YARN-4280: --- Thank you so much [~leftnoteasy], [~jlowe]. I am updating the patch

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15342666#comment-15342666 ] Jason Lowe commented on YARN-4280: -- I haven't had a chance to look at the patch yet, but I'm not thrilled

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15337163#comment-15337163 ] Wangda Tan commented on YARN-4280: -- Thanks [~kshukla] for working on this patch. My understanding of how

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15336745#comment-15336745 ] Kuhu Shukla commented on YARN-4280: --- TestAMRestart passes locally with the patch. The checkstyle issue is

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15336467#comment-15336467 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15334467#comment-15334467 ] Hadoop QA commented on YARN-4280: - | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem ||

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15274661#comment-15274661 ] Wangda Tan commented on YARN-4280: -- [~jlowe], thanks for sharing your thoughts, agree with your plan about

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15268908#comment-15268908 ] Jason Lowe commented on YARN-4280: -- The proposed algorithm does not change how reserved containers work --

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15267623#comment-15267623 ] Wangda Tan commented on YARN-4280: -- Thanks [~jlowe], your comments are all make sense to me. [~kshukla],

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15267527#comment-15267527 ] Kuhu Shukla commented on YARN-4280: --- I will try to code up a prototype for this asap. Thank you all. >

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15266711#comment-15266711 ] Jason Lowe commented on YARN-4280: -- Clarification: I meant to say "absolute max capacity" above whenever I

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15266673#comment-15266673 ] Jason Lowe commented on YARN-4280: -- bq. The problem of allowing one container reserved exceed queue's max

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15265860#comment-15265860 ] Wangda Tan commented on YARN-4280: -- [~jlowe], Agree, and actually even after enable preemption, sometimes

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15264079#comment-15264079 ] Jason Lowe commented on YARN-4280: -- I'm not thrilled with the idea of preemption to solve this issue.

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14971154#comment-14971154 ] Jason Lowe commented on YARN-4280: -- It's a sticky problem. The problem with doing the resource check is

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14972110#comment-14972110 ] Wangda Tan commented on YARN-4280: -- [~jlowe], agree, this is a hard problem to solve, allowing reservation

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14968607#comment-14968607 ] Wangda Tan commented on YARN-4280: -- Thanks for sharing thoughts, [~jlowe]. Agree, and this requires to do

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14967740#comment-14967740 ] Wangda Tan commented on YARN-4280: -- Thanks [~kshukla] reporting this issue. I think this problem could be

[jira] [Commented] (YARN-4280) CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster

[ https://issues.apache.org/jira/browse/YARN-4280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14967904#comment-14967904 ] Jason Lowe commented on YARN-4280: -- I also thought originally that relaxing the queue max cap check could