From da42b01645d30f0a243731396bc2f392d344da48 Mon Sep 17 00:00:00 2001 From: Rohith Sharma K S Date: Mon, 11 Jan 2016 12:07:16 +0530 Subject: [PATCH] Add YARN-3849 to Release 2.6.4 entry in CHANGES.txt (cherry picked from commit 76e72708511100dbfaba910e120892b87b87edee) --- hadoop-yarn-project/CHANGES.txt | 37 ++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 17 deletions(-) diff --git a/hadoop-yarn-project/CHANGES.txt b/hadoop-yarn-project/CHANGES.txt index 19688665d50..e4edd60a8f7 100644 --- a/hadoop-yarn-project/CHANGES.txt +++ b/hadoop-yarn-project/CHANGES.txt @@ -2089,31 +2089,34 @@ Release 2.6.4 - UNRELEASED BUG FIXES - YARN-2975. FSLeafQueue app lists are accessed without required locks. (kasha) + YARN-2975. FSLeafQueue app lists are accessed without required locks. (kasha) - YARN-3857: Memory leak in ResourceManager with SIMPLE mode. - (mujunchao via zxu) + YARN-3857: Memory leak in ResourceManager with SIMPLE mode. + (mujunchao via zxu) - YARN-3535. Scheduler must re-request container resources when RMContainer transitions - from ALLOCATED to KILLED (rohithsharma and peng.zhang via asuresh) + YARN-3535. Scheduler must re-request container resources when RMContainer transitions + from ALLOCATED to KILLED (rohithsharma and peng.zhang via asuresh) - YARN-4452. NPE when submit Unmanaged application. (Naganarasimha G R - via junping_du) + YARN-4452. NPE when submit Unmanaged application. (Naganarasimha G R + via junping_du) - YARN-3893. Both RM in active state when Admin#transitionToActive failure - from refeshAll() (Bibin A Chundatt via rohithsharmaks) + YARN-3893. Both RM in active state when Admin#transitionToActive failure + from refeshAll() (Bibin A Chundatt via rohithsharmaks) - YARN-3697. FairScheduler: ContinuousSchedulingThread can fail to shutdown. - (Zhihai Xu via kasha) + YARN-3697. FairScheduler: ContinuousSchedulingThread can fail to shutdown. + (Zhihai Xu via kasha) - YARN-4546. ResourceManager crash due to scheduling opportunity overflow. - (Jason Lowe via junping_du) + YARN-4546. ResourceManager crash due to scheduling opportunity overflow. + (Jason Lowe via junping_du) - YARN-2902. Killing a container that is localizing can orphan resources in - the DOWNLOADING state (Varun Saxena via jlowe) + YARN-2902. Killing a container that is localizing can orphan resources in + the DOWNLOADING state (Varun Saxena via jlowe) - YARN-4354. Public resource localization fails with NPE. (Jason Lowe via - junping_du) + YARN-4354. Public resource localization fails with NPE. (Jason Lowe via + junping_du) + + YARN-3849. Too much of preemption activity causing continuos killing of + containers across queues. (Sunil G via wangda) Release 2.6.3 - 2015-12-17