big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
13,230 Commits 2 Branches 2 Tags 264 MiB
552237d4a3
Commit Graph

1808 Commits

Author SHA1 Message Date
Robert Kanter
954dd57043 YARN-4697. NM aggregation thread pool is not bound by limits (haibochen via rkanter) 2016-02-24 15:00:24 -08:00
Sangjin Lee
553b591ba0 YARN-4722. AsyncDispatcher logs redundant event queue sizes (Jason Lowe via sjlee) 2016-02-24 09:29:41 -08:00
Jason Lowe
d284e187b8 YARN-2046. Out of band heartbeats are sent only on container kill and possibly too early. Contributed by Ming Ma 2016-02-23 20:49:09 +00:00
Junping Du
9ed17f181d YARN-3223. Resource update during NM graceful decommission. Contributed by Brook Zhou. 2016-02-23 03:30:26 -08:00
Tsuyoshi Ozawa
0e12114c9c YARN-4648. Move preemption related tests from TestFairScheduler to TestFairSchedulerPreemption. Contributed by Kai Sasaki. 2016-02-23 19:50:08 +09:00
Varun Vasudev
140cb5d745 YARN-4709. NMWebServices produces incorrect JSON for containers. Contributed by Varun Saxena. 2016-02-23 12:29:25 +05:30
Junping Du
3fab88540f YARN-4386. refreshNodesGracefully() should send recommission event to active RMNodes only. Contributed by Kuhu Shukla. 2016-02-22 07:04:19 -08:00
Sangjin Lee
7de70680fe YARN-4690. Skip object allocation in FSAppAttempt#getResourceUsage when possible (Ming Ma via sjlee) 2016-02-17 20:55:21 -08:00
Karthik Kambatla
2ab4c476ed YARN-4689. FairScheduler: Cleanup preemptContainer to be more readable. (Kai Sasaki via kasha) 2016-02-17 18:16:15 -08:00
Arun Suresh
23f937e3b7 YARN-2575. Create separate ACLs for Reservation create/update/delete/list ops (Sean Po via asuresh) 2016-02-11 10:47:43 -08:00
Varun Vasudev
fa00d3e205 YARN-4655. Log uncaught exceptions/errors in various thread pools in YARN. Contributed by Sidharta Seethana. 2016-02-11 12:06:42 +05:30
Jian He
d16b17b4d2 YARN-4138. Roll back container resource allocation after resource increase token expires. Contributed by Meng Ding 2016-02-11 10:06:27 +08:00
=
b706cbc1bc YARN-4420. Add REST API for List Reservations (Sean Po via curino) 2016-02-10 10:19:26 -08:00
Arun Suresh
5cf5c41a89 YARN-4360. Improve GreedyReservationAgent to support "early" allocations, and performance improvements (curino via asuresh) 2016-02-10 09:11:15 -08:00
Devaraj K
565af873d5 YARN-4667. RM Admin CLI for refreshNodesResources throws NPE when nothing 
is configured. Contributed by Naganarasimha G R.
 2016-02-08 15:01:54 +05:30
Varun Vasudev
22a2b2231d YARN-4669. Fix logging statements in resource manager's Application class. Contributed by Sidharta Seethana. 2016-02-04 13:51:25 +05:30
Varun Vasudev
308d63f382 YARN-4307. Display blacklisted nodes for AM container in the RM web UI. Contributed by Naganarasimha G R. 2016-02-04 13:32:54 +05:30
Jason Lowe
fa328e2d39 YARN-4594. container-executor fails to remove directory tree when chmod required. Contributed by Colin Patrick McCabe 2016-02-03 17:21:12 +00:00
Varun Vasudev
1adb64e09b YARN-4625. Make ApplicationSubmissionContext and ApplicationSubmissionContextInfo more consistent. Contributed by Xuan Gong. 2016-02-03 16:26:28 +05:30
Varun Vasudev
1cd55e0c17 YARN-4649. Add additional logging to some NM state store operations. Contributed by Sidharta Seethana 2016-02-02 09:25:23 +05:30
Wangda Tan
9875325d5c YARN-4340. Add list API to reservation system. (Sean Po via wangda) 2016-02-02 10:17:33 +08:00
Jason Lowe
ed55950164 YARN-3102. Decommisioned Nodes not listed in Web UI. Contributed by Kuhu Shukla 2016-02-01 23:15:26 +00:00
Rohith Sharma K S
2673cbaf55 YARN-4615. Fix random test failure in TestAbstractYarnScheduler#testResourceRequestRecoveryToTheRightAppAttempt. (Sunil G via rohithsharmaks) 2016-02-01 10:43:56 +05:30
Jason Lowe
772ea7b41b YARN-4428. Redirect RM page to AHS page when AHS turned on and RM page is not available. Contributed by Chang Li 2016-01-29 21:48:54 +00:00
Jian He
f4a57d4a53 YARN-4617. LeafQueue#pendingOrderingPolicy should always use fixed ordering policy instead of using same as active applications ordering policy. Contributed by Rohith Sharma K S 2016-01-29 12:22:23 -08:00
Karthik Kambatla
c9a09d6926 YARN-4647. Make RegisterNodeManagerRequestPBImpl thread-safe. (kasha) 2016-01-29 08:12:54 -08:00
Devaraj K
a277bdc9ed YARN-4411. RMAppAttemptImpl#createApplicationAttemptReport throws 
IllegalArgumentException. Contributed by Bibin A Chundatt and yarntime.
 2016-01-29 13:51:37 +05:30
Rohith Sharma K S
ac68666803 YARN-4543. Fix random test failure in TestNodeStatusUpdater.testStopReentrant. (Akihiro Suda via rohithsharmaks) 2016-01-29 12:29:54 +05:30
Rohith Sharma K S
09d831c95b YARN-4219. addendum patch to fix javadoc errors 2016-01-29 11:51:47 +05:30
Jian He
7f46636495 YARN-4519. Potential deadlock of CapacityScheduler between decrease container and assign containers. Contributed by Meng Ding 2016-01-28 14:51:00 -08:00
Xuan
9fab22b366 YARN-4219. New levelDB cache storage for timeline v1.5. Contributed by 
Li Lu
 2016-01-28 14:24:22 -08:00
Jason Lowe
61382ff8fa YARN-4643. Container recovery is broken with delegating container runtime. Contributed by Sidharta Seethana 2016-01-28 18:59:35 +00:00
Rohith Sharma K S
ef343be82b YARN-4633. Fix random test failure in TestRMRestart#testRMRestartAfterPreemption. (Bibin A Chundatt via rohithsharmaks) 2016-01-28 21:53:45 +05:30
Karthik Kambatla
fb238d7e5d YARN-4462. FairScheduler: Disallow preemption from a queue. (Tao Jie via kasha) 2016-01-27 12:29:06 -08:00
Rohith Sharma K S
c01bee0108 YARN-4573. Fix test failure in TestRMAppTransitions#testAppRunningKill and testAppKilledKilled. (Takashi Ohnishi via rohithsharmaks) 2016-01-27 08:23:02 +05:30
Vinod Kumar Vavilapalli (I am also known as @tshooter.)
2085e60a96 YARN-3542. Refactored existing CPU cgroups support to use the newer and integrated ResourceHandler mechanism, and also deprecated the old LCEResourceHandler inteface hierarchy. Contributed by Varun Vasudev. 2016-01-25 16:19:36 -08:00
Jian He
992dd2f783 YARN-4520. Finished app info is unnecessarily persisted in NM state-store if container is acquired but not lunched on this node. Contributed by sandflee 2016-01-25 15:36:14 -08:00
rohithsharmaks
10dc2c0493 YARN-4613. Fix test failure in TestClientRMService#testGetClusterNodes. (Takashi Ohnishi via rohithsharmaks) 2016-01-24 23:36:15 +05:30
rohithsharmaks
99829eb221 YARN-4614. Fix random failure in TestApplicationPriority#testApplicationPriorityAllocationWithChangeInPriority. (Sunil G via rohithsharmaks) 2016-01-23 07:56:57 +05:30
Xuan
618bfd6ac2 YARN-4496. Improve HA ResourceManager Failover detection on the client. 
Contributed by Jian He
 2016-01-22 18:20:38 -08:00
Jason Lowe
46e5ea81e0 YARN-4598. Invalid event: RESOURCE_FAILED at CONTAINER_CLEANEDUP_AFTER_KILL. Contributed by tangshangwen 2016-01-22 21:55:01 +00:00
rohithsharmaks
d6258b33a7 YARN-4497. RM might fail to restart when recovering apps whose attempts are missing. (Jun Gong via rohithsharmaks) 2016-01-22 20:27:38 +05:30
Akira Ajisaka
8f58f742ae YARN-4605. Spelling mistake in the help message of "yarn applicationattempt" command. Contributed by Weiwei Yang. 2016-01-22 19:43:06 +09:00
Varun Vasudev
b41a7e89d1 YARN-4578. Directories that are mounted in docker containers need to be more restrictive/container-specific. Contributed by Sidharta Seethana. 2016-01-22 14:43:14 +05:30
Rohith Sharma K S
e30668106d YARN-4584. RM startup failure when AM attempts greater than max-attempts. (Bibin A Chundatt via rohithsharmaks) 2016-01-22 10:14:46 +05:30
Jason Lowe
468a53b22f YARN-4610. Reservations continue looking for one app causes other apps to starve. Contributed by Jason Lowe 2016-01-21 18:31:29 +00:00
Karthik Kambatla
4992398aee YARN-4603. FairScheduler should mention user requested queuename in error message when failed in queue ACL check. (Tao Jie via kasha) 2016-01-21 17:40:59 +01:00
Wangda Tan
5ff5f67332 YARN-4557. Fix improper Queues sorting in PartitionedQueueComparator when accessible-node-labels=*. (Naganarasimha G R via wangda) 2016-01-21 11:21:06 +08:00
Xuan
890a2ebd1a YARN-4559. Make leader elector and zk store share the same curator 
client. Contributed by Jian He
 2016-01-20 14:48:10 -08:00
Wangda Tan
89d1fd5dac HADOOP-12356. Fix computing CPU usage statistics on Windows. (Inigo Goiri via wangda) 2016-01-19 21:27:38 +08:00
Jian He
edc43a9097 YARN-4565. Fix a bug that leads to AM resource limit not hornored when sizeBasedWeight enabled for FairOrderingPolicy. Contributed by Wangda Tan 2016-01-18 21:04:36 -08:00
Wangda Tan
a44ce3f14f YARN-4502. Fix two AM containers get allocated when AM restart. (Vinod Kumar Vavilapalli via wangda) 2016-01-19 09:30:04 +08:00
Wangda Tan
150f5ae034 Revert "YARN-4502. Fix two AM containers get allocated when AM restart. (Vinod Kumar Vavilapalli via wangda)" 
This reverts commit 3fe5728563.

Conflicts:
	hadoop-yarn-project/CHANGES.txt
 2016-01-19 09:27:36 +08:00
Jian He
f385851141 YARN-4596. SystemMetricPublisher should not swallow error messages from TimelineClient#putEntities. Contributed by Li Lu 2016-01-18 16:58:39 -08:00
Karthik Kambatla
d40859fab1 YARN-4526. Make SystemClock singleton so AppSchedulingInfo could use it. (kasha) 2016-01-18 10:58:14 +01:00
Wangda Tan
3fe5728563 YARN-4502. Fix two AM containers get allocated when AM restart. (Vinod Kumar Vavilapalli via wangda) 
(cherry picked from commit 805a9ed85e)
 2016-01-18 17:06:05 +08:00
Wangda Tan
adf260a728 Revert "YARN-4502. Fix two AM containers get allocated when AM restart. (Vinod Kumar Vavilapalli via wangda)" 
This reverts commit 805a9ed85e.
 2016-01-18 16:50:45 +08:00
Wangda Tan
b08ecf5c75 YARN-4304. AM max resource configuration per partition to be displayed/updated correctly in UI and in various partition related metrics. (Sunil G via wangda) 2016-01-18 11:11:32 +08:00
Wangda Tan
805a9ed85e YARN-4502. Fix two AM containers get allocated when AM restart. (Vinod Kumar Vavilapalli via wangda) 2016-01-18 11:04:25 +08:00
Wangda Tan
9523648d57 YARN-4538. QueueMetrics pending cores and memory metrics wrong. (Bibin A Chundatt via wangda) 2016-01-18 10:57:14 +08:00
Junping Du
02f597c5db YARN-4265. Provide new timeline plugin storage to support fine-grained entity caching. Contributed by Li Lu and Jason Lowe 2016-01-17 17:37:40 -08:00
Colin Patrick Mccabe
b2c155f810 HADOOP-12712. Fix some cmake plugin and native build warnings (cmccabe) 2016-01-15 12:29:18 -08:00
Junping Du
fc6d3a3b23 YARN-4581. AHS writer thread leak makes RM crash while RM is recovering. Contributed by sandflee. 2016-01-15 08:40:56 -08:00
rohithsharmaks
f7736f464f YARN-4389. Allow application to enable or disable am blacklisting. (Sunil G via rohithsharmaks) 2016-01-15 21:38:26 +05:30
Colin Patrick Mccabe
b1ed28fa77 HADOOP-8887. Use a Maven plugin to build the native code using CMake (cmccabe) 2016-01-14 11:02:34 -08:00
Karthik Kambatla
9d04f26d4c YARN-3446. FairScheduler headroom calculation should exclude nodes in the blacklist. (Zhihai Xu via kasha) 2016-01-14 08:33:23 -08:00
Varun Vasudev
3ddb92bd30 YARN-4553. Add cgroups support for docker containers. Contributed by Sidharta Seethana. 2016-01-14 14:29:29 +05:30
Karthik Kambatla
321072ba81 YARN-4551. Address the duplication between StatusUpdateWhenHealthy and StatusUpdateWhenUnhealthy transitions. (Sunil G via kasha) 2016-01-13 12:09:34 -08:00
Wangda Tan
c0537bcd2c YARN-4571. Make app id/name available to the yarn authorizer provider for better auditing. (Jian He via wangda) 2016-01-13 13:18:31 +08:00
Jason Lowe
13de8359a1 YARN-4414. Nodemanager connection errors are retried at multiple levels. Contributed by Chang Li 2016-01-12 15:56:15 +00:00
Akira Ajisaka
da1e3e3c57 YARN-4567. javadoc failing on java 8. Contributed by Steve Loughran. This closes #67. 2016-01-12 15:12:17 +09:00
Wangda Tan
9e792da014 YARN-4582. Label-related invalid resource request exception should be able to properly handled by application. (Bibin A Chundatt via wangda) 2016-01-12 12:53:31 +08:00
Jian He
5fab4ec31c Missing file for YARN-4580 2016-01-11 17:00:44 -08:00
Jian He
b8942be888 YARN-4537. Pull out priority comparison from fifocomparator and use compound comparator for FifoOrdering policy. Contributed by Rohith Sharma K S 2016-01-11 16:44:28 -08:00
Jian He
109e528ef5 YARN-4479. Change CS LeafQueue pendingOrderingPolicy to hornor recovered apps. Contributed by Rohith Sharma K S 2016-01-08 15:51:10 -08:00
Xuan
89022f8d4b YARN-4438. Implement RM leader election with curator. Contributed by Jian He 2016-01-07 14:33:06 -08:00
Akira Ajisaka
52b77577c4 YARN-4544. All the log messages about rolling monitoring interval are shown with WARN level. Contributed by Takashi Ohnishi. 2016-01-08 01:12:46 +09:00
Steve Loughran
07d1cb612c YARN-4550. Some tests in TestContainerLanch fails on non-english locale environment. (Takashi Ohnishi via stevel) 2016-01-07 14:30:20 +00:00
rohithsharmaks
791c1639ae YARN-4393. Fix intermittent test failure for TestResourceLocalizationService#testFailedDirsResourceRelease (Varun Saxana via rohithsharmaks) 2016-01-07 09:38:47 +05:30
Junping Du
c1462a67ff YARN-4546. ResourceManager crash due to scheduling opportunity overflow. Contributed by Jason Lowe. 2016-01-06 05:49:24 -08:00
rohithsharmaks
6da6d87872 YARN-4535. Fix checkstyle error in CapacityScheduler.java (Naganarasimha G R via rohithsharmaks) 2016-01-05 12:09:57 +05:30
Rohith Sharma K S
f9e36dea96 YARN-4530. LocalizedResource trigger a NPE Cause the NodeManager exit. (tangshangwen via rohithsharmaks) 2016-01-04 09:58:13 +05:30
Wangda Tan
4e4b3a8465 YARN-4524. Cleanup AppSchedulingInfo. (Karthik Kambatla via wangda) 
(cherry picked from commit 05fa852d7567b7590d6b53bbf925f8f424736514)
 2015-12-30 15:39:34 -08:00
Wangda Tan
8310b2e9ff YARN-4522. Queue acl can be checked at app submission. (Jian He via wangda) 2015-12-30 15:30:12 -08:00
Junping Du
223ce323bb YARN-1382. Remove unusableRMNodesConcurrentSet (never used) in NodeListManager to get rid of memory leak. Contributed by Rohith Sharma K S. 2015-12-30 07:52:07 -08:00
Jian He
5273413411 YARN-3480. Remove attempts that are beyond max-attempt limit from state store. Contributed by Jun Gong 2015-12-29 15:58:39 -08:00
Wangda Tan
561abb9fee YARN-4315. NaN in Queue percentage for cluster apps page. (Bibin A Chundatt via wangda) 2015-12-29 13:28:00 -08:00
Jian He
d0a22bae9b YARN-4417. Make RM and Timeline-server REST APIs more consistent. Contributed by Wangda Tan 2015-12-28 15:52:45 -08:00
Gera Shegalov
2c17b81569 YARN-2934. Improve handling of container's stderr. (Naganarasimha G R via gera) 2015-12-24 23:48:05 -08:00
Karthik Kambatla
0af492b4bd YARN-4156. TestAMRestart#testAMBlacklistPreventsRestartOnSameNode assumes CapacityScheduler. (Anubhav Dhoot via kasha) 2015-12-23 17:52:36 -08:00
Junping Du
882f2f0464 YARN-4234. New put APIs in TimelineClient for ats v1.5. Contributed by Xuan Gong. 2015-12-23 05:26:51 -08:00
rohithsharmaks
8c180a13c8 YARN-4109. Exception on RM scheduler page loading with labels. (Mohammad Shahid Khan via rohithsharmaks) 2015-12-23 09:12:32 +05:30
Arun Suresh
e88422df45 YARN-4477. FairScheduler: Handle condition which can result in an infinite loop in attemptScheduling. (Tao Jie via asuresh) 2015-12-21 22:41:09 -08:00
Wangda Tan
bc038b382c YARN-4454. NM to nodelabel mapping going wrong after RM restart. (Bibin A Chundatt via wangda) 2015-12-21 11:30:13 -08:00
Uma Mahesh
0f82b5d878 YARN-4480. Clean up some inappropriate imports. (Kai Zheng via umamahesh) 2015-12-19 23:10:13 -08:00
Jian He
85c2466048 YARN-4164. Changed updateApplicationPriority API to return the updated application priority. Contributed by Rohith Sharma K S 2015-12-18 14:13:48 -08:00
Junping Du
1de56b0448 YARN-3226. UI changes for decommissioning node. Contributed by Sunil G. 2015-12-17 15:20:17 -08:00
Vinod Kumar Vavilapalli (I am also known as @tshooter.)
4e7d32c0db YARN-1856. Added cgroups based memory monitoring for containers as another alternative to custom memory-monitoring. Contributed by Varun Vasudev. 2015-12-17 12:13:03 -08:00
Jason Lowe
91828fef6b YARN-4461. Redundant nodeLocalityDelay log in LeafQueue. Contributed by Eric Payne 2015-12-16 23:22:31 +00:00
Wangda Tan
9b856d9787 YARN-4416. Deadlock due to synchronised get Methods in AbstractCSQueue. (Naganarasimha G R via wangda) 2015-12-16 13:22:37 -08:00