From e2f027dddcbbf7f339f518476e2d43c123b64abc Mon Sep 17 00:00:00 2001 From: Tsuyoshi Ozawa Date: Wed, 21 Oct 2015 23:15:10 +0900 Subject: [PATCH] Moved CHANGES.txt entries to 2.6.2 YARN-2019, YARN-4092, YARN-4101, YARN-4087, YARN-3727, YARN-4005, YARN-3780, YARN-3802, YARN-3896, YARN-3554, YARN-3194, YARN-3798 (cherry picked from commit 14f0e2787c7c590d8ec1fc52f437866f34b098bd) --- hadoop-yarn-project/CHANGES.txt | 54 +++++++++++++-------------------- 1 file changed, 21 insertions(+), 33 deletions(-) diff --git a/hadoop-yarn-project/CHANGES.txt b/hadoop-yarn-project/CHANGES.txt index 8afc2d2bfd6..5d4945a69e2 100644 --- a/hadoop-yarn-project/CHANGES.txt +++ b/hadoop-yarn-project/CHANGES.txt @@ -926,9 +926,6 @@ Release 2.7.2 - UNRELEASED NEW FEATURES - YARN-2019. Retrospect on decision of making RM crashed if any exception throw - in ZKRMStateStore. (Jian He via junping_du) - IMPROVEMENTS YARN-3170. YARN architecture document needs updating. (Brahma Reddy Battula @@ -937,9 +934,6 @@ Release 2.7.2 - UNRELEASED YARN-3967. Fetch the application report from the AHS if the RM does not know about it. (Mit Desai via xgong) - YARN-4092. Fixed UI redirection to print useful messages when both RMs are - in standby mode. (Xuan Gong via jianhe) - YARN-2801. Add documentation for node labels feature. (Wangda Tan and Naganarasimha G R via ozawa) @@ -976,18 +970,12 @@ Release 2.7.2 - UNRELEASED YARN-3893. Both RM in active state when Admin#transitionToActive failure from refeshAll() (Bibin A Chundatt via rohithsharmaks) - YARN-4101. RM should print alert messages if Zookeeper and Resourcemanager - gets connection issue. (Xuan Gong via jianhe) - YARN-4103. RM WebServices missing scheme for appattempts logLinks. (Jonathan Eagles via vvasudeb) YARN-4105. Capacity Scheduler headroom for DRF is wrong (Chang Li via jlowe) - YARN-4087. Followup fixes after YARN-2019 regarding RM behavior when - state-store error occurs. (Jian He via xgong) - YARN-4096. App local logs are leaked if log aggregation fails to initialize for the app. (Jason Lowe via zxu) @@ -1015,27 +1003,12 @@ Release 2.7.2 - UNRELEASED YARN-4180. AMLauncher does not retry on failures when talking to NM. (adhoot) - YARN-3727. For better error recovery, check if the directory exists before - using it for localization. (Zhihai Xu via jlowe) - YARN-3619. ContainerMetrics unregisters during getMetrics and leads to ConcurrentModificationException (Zhihai Xu via jlowe) YARN-4209. RMStateStore FENCED state doesn’t work due to updateFencedState called by stateMachine.doTransition. (Zhihai Xu via rohithsharmaks) - YARN-4005. Completed container whose app is finished is possibly not - removed from NMStateStore. (Jun Gong via jianhe) - - YARN-3780. Should use equals when compare Resource in RMNodeImpl#ReconnectNodeTransition. - (zhihai xu via devaraj) - - YARN-3802. Two RMNodes for the same NodeId are used in RM sometimes - after NM is reconnected. (zhihai xu via xgong) - - YARN-3896. RMNode transitioned from RUNNING to REBOOTED because its response id - has not been reset synchronously. (Jun Gong via rohithsharmaks) - Release 2.7.1 - 2015-07-06 INCOMPATIBLE CHANGES @@ -1101,9 +1074,6 @@ Release 2.7.1 - 2015-07-06 YARN-3358. Audit log not present while refreshing Service ACLs. (Varun Saxena via devaraj) - YARN-3554. Default value for maximum nodemanager connect wait time is too - high (Naganarasimha G R via jlowe) - YARN-3476. Nodemanager can fail to delete local logs if log aggregation fails (Rohith via jlowe) @@ -1709,9 +1679,6 @@ Release 2.7.0 - 2015-04-20 YARN-933. Fixed InvalidStateTransitonException at FINAL_SAVING state in RMApp. (Rohith Sharmaks via jianhe) - YARN-3194. RM should handle NMContainerStatuses sent by NM while - registering if NM is Reconnected node (Rohith via jlowe) - YARN-3247. TestQueueMappings should use CapacityScheduler explicitly. (Zhihai Xu via ozawa) @@ -1776,12 +1743,30 @@ Release 2.6.2 - UNRELEASED NEW FEATURES + YARN-2019. Retrospect on decision of making RM crashed if any exception throw + in ZKRMStateStore. (Jian He via junping_du) + IMPROVEMENTS + YARN-4092. Fixed UI redirection to print useful messages when both RMs are + in standby mode. (Xuan Gong via jianhe) + + YARN-4101. RM should print alert messages if Zookeeper and Resourcemanager + gets connection issue. (Xuan Gong via jianhe) + OPTIMIZATIONS BUG FIXES + YARN-4087. Followup fixes after YARN-2019 regarding RM behavior when + state-store error occurs. (Jian He via xgong) + + YARN-3554. Default value for maximum nodemanager connect wait time is too + high (Naganarasimha G R via jlowe) + + YARN-3727. For better error recovery, check if the directory exists before + using it for localization. (Zhihai Xu via jlowe) + YARN-4005. Completed container whose app is finished is possibly not removed from NMStateStore. (Jun Gong via jianhe) @@ -1797,6 +1782,9 @@ Release 2.6.2 - UNRELEASED YARN-3896. RMNode transitioned from RUNNING to REBOOTED because its response id has not been reset synchronously. (Jun Gong via rohithsharmaks) + YARN-3798. ZKRMStateStore shouldn't create new session without occurrance of + SESSIONEXPIED. (ozawa and Varun Saxena) + Release 2.6.1 - 2015-09-23 INCOMPATIBLE CHANGES