673 Commits

Author SHA1 Message Date
subru
fec639cda1 YARN-1712. Plan follower that synchronizes the current state of reservation subsystem with the scheduler. Contributed by Subru Krishnan and Carlo Curino.
(cherry picked from commit 169085319b8b76641f8b9f6840a3fef06d221e2b)
(cherry picked from commit 3418c56bcf4bbddaf483bdaa1a15a8bbc4039bfe)
2014-10-06 10:21:06 -07:00
carlo curino
6bfdaf06c4 YARN-1711. Policy to enforce instantaneous and over-time quotas on user reservation. Contributed by Carlo Curino and Subru Krishnan.
(cherry picked from commit c4918cb4cb5a267a8cfd6eace28fcfe7ad6174e8)
(cherry picked from commit b6df0dddcdafd7ec67c76ea92aea3ff3e94db247)
2014-10-06 10:21:05 -07:00
carlo curino
6a3c167175 YARN-1710. Logic to find allocations within a Plan that satisfy user ReservationRequest(s). Contributed by Carlo Curino and Subru Krishnan.
(cherry picked from commit aef7928899b37262773f3dc117157bb746bf8918)
(cherry picked from commit f66ffcf832235e0da0bb050fff08e248b547c360)
2014-10-06 10:21:05 -07:00
subru
63250ef9d6 YARN-1709. In-memory data structures used to track resources over time to enable reservations.
(cherry picked from commit 0d8b2cd88b958b1e602fd4ea4078ef8d4742a7c3)
(cherry picked from commit cf4b34282aafee9f6b09d3433c4de1ae4b359168)
2014-10-06 10:21:05 -07:00
carlo curino
b81f571e60 YARN-2475. Logic for responding to capacity drops for the ReservationSystem. Contributed by Carlo Curino and Subru Krishnan.
(cherry picked from commit f83a07f266f2c5e6eead554d8a331ed7e75e10d5)
(cherry picked from commit 1c6950354f3c35a7824770dc251d5aec3be4876a)
2014-10-06 10:21:05 -07:00
carlo curino
cf5ef00b96 YARN-1707. Introduce APIs to add/remove/resize queues in the CapacityScheduler. Contributed by Carlo Curino and Subru Krishnan
(cherry picked from commit aac47fda7fecda9fc18ade34d633eca895865a70)

Conflicts:
	hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/scheduler/capacity/CapacityScheduler.java
(cherry picked from commit eb3e40b833b9d82c1556843f960194dc42e482f3)
2014-10-06 10:21:04 -07:00
Karthik Kambatla
b6ce0a1f69 YARN-2635. TestRM, TestRMRestart, TestClientToAMTokens should run with both CS and FS. (Wei Yan and kasha via kasha)
(cherry picked from commit 80d11eb68e60f88e16d7d41edecbddfc935a6b10)
2014-10-03 11:52:05 -07:00
Jian He
6f4c77409b YARN-2628. Capacity scheduler with DominantResourceCalculator carries out reservation even though slots are free. Contributed by Varun Vasudev
(cherry picked from commit 054f28552687e9b9859c0126e16a2066e20ead3f)
2014-10-02 15:16:07 -07:00
Karthik Kambatla
b6f4d1ef42 YARN-2254. TestRMWebServicesAppsModification should run against both CS and FS. (Zhihai Xu via kasha)
(cherry picked from commit 5e0b49da9caa53814581508e589f3704592cf335)
2014-10-02 10:15:41 -07:00
Zhijie Shen
4b50e23271 YARN-2630. Prevented previous AM container status from being acquired by the current restarted AM. Contributed by Jian He.
(cherry picked from commit 52bbe0f11bc8e97df78a1ab9b63f4eff65fd7a76)
2014-10-01 15:39:36 -07:00
Jian He
823f02725f YARN-2602. Fixed possible NPE in ApplicationHistoryManagerOnTimelineStore. Contributed by Zhijie Shen
(cherry picked from commit bbff96be48119774688981d04baf444639135977)
2014-09-30 16:44:44 -07:00
Karthik Kambatla
5a43b795b7 YARN-2594. Potential deadlock in RM when querying ApplicationResourceUsageReport. (Wangda Tan via kasha)
(cherry picked from commit 14d60dadc25b044a2887bf912ba5872367f2dffb)
2014-09-30 16:17:38 -07:00
Zhijie Shen
d977205b9b YARN-2627. Added the info logs of attemptFailuresValidityInterval and number of previous failed attempts. Contributed by Xuan Gong.
(cherry picked from commit 9582a50176800433ad3fa8829a50c28b859812a3)
2014-09-30 12:59:34 -07:00
Jason Lowe
4f6a367166 YARN-1769. CapacityScheduler: Improve reservations. Contributed by Thomas Graves
(cherry picked from commit 9c22065109a77681bc2534063eabe8692fbcb3cd)
2014-09-29 14:13:21 +00:00
Jian He
cb08ed1484 YARN-668. Changed NMTokenIdentifier/AMRMTokenIdentifier/ContainerTokenIdentifier to use protobuf object as the payload. Contributed by Junping Du.
(cherry picked from commit 5391919b09ce9549d13c897aa89bb0a0536760fe)
2014-09-26 17:53:35 -07:00
Karthik Kambatla
b923c291b4 YARN-2608. FairScheduler: Potential deadlocks in loading alloc files and clock access. (Wei Yan via kasha)
(cherry picked from commit c9811af09a3d3f9f2f1b86fc9d6f2763d3225e44)
2014-09-25 17:43:08 -07:00
Jason Lowe
e4d46e5ff6 YARN-2523. ResourceManager UI showing negative value for "Decommissioned Nodes" field. Contributed by Rohith
(cherry picked from commit 8269bfa613999f71767de3c0369817b58cfe1416)
2014-09-25 22:39:57 +00:00
Zhijie Shen
83926b3c57 YARN-2546. Made REST API for application creation/submission use numeric and boolean types instead of the string of them. Contributed by Varun Vasudev.
(cherry picked from commit 72b0881ca641fa830c907823f674a5c5e39aa15a)
2014-09-24 17:58:48 -07:00
Zhijie Shen
3a2e400377 YARN-2581. Passed LogAggregationContext to NM via ContainerTokenIdentifier. Contributed by Xuan Gong.
(cherry picked from commit c86674a3a4d99aa56bb8ed3f6df51e3fef215eba)
2014-09-24 17:51:54 -07:00
Karthik Kambatla
30165d4cdd YARN-2596. TestWorkPreservingRMRestart fails with FairScheduler. (kasha)
(cherry picked from commit 39c87344e16a08ab69e25345b3bce92aec92db47)
2014-09-24 15:24:17 -07:00
Karthik Kambatla
024bcfdc0d YARN-2252. Intermittent failure of TestFairScheduler.testContinuousScheduling. (Ratandeep Ratti and kasha via kasha)
(cherry picked from commit f5578207d2d20aa71adf12aa9f6c37b00ee9b8dc)
2014-09-23 00:03:47 -07:00
Karthik Kambatla
8ba8521de5 YARN-1959. Fix headroom calculation in FairScheduler. (Anubhav Dhoot via kasha)
(cherry picked from commit 568d3dc2bbe43b7d2833d5da2b0e6d75eb86e5dd)
2014-09-22 23:51:45 -07:00
Karthik Kambatla
3984435661 YARN-2539. FairScheduler: Set the default value for maxAMShare to 0.5. (Wei Yan via kasha)
(cherry picked from commit 43efdd30b52f4ee8967a03dac4d548095d6cafdb)
2014-09-22 16:10:20 -07:00
Karthik Kambatla
24a814771e YARN-2540. FairScheduler: Queue filters not working on scheduler page in RM UI. (Ashwin Shankar via kasha)
(cherry picked from commit 23e17ce0ffba11a08c033b753573a756532cd1c2)
2014-09-22 15:42:36 -07:00
Jian He
3ce97a9efd YARN-1372. Ensure all completed containers are reported to the AMs across RM restart. Contributed by Anubhav Dhoot
(cherry picked from commit 0a641496c706fc175e7bf66d69ebf71c7d078e84)
2014-09-22 10:32:44 -07:00
Karthik Kambatla
c1a3819a4d YARN-2453. TestProportionalCapacityPreemptionPolicy fails with FairScheduler. (Zhihai Xu via kasha)
(cherry picked from commit 9721e2c1feb5aecea3a6dab5bda96af1cd0f8de3)
2014-09-21 23:14:27 -07:00
Karthik Kambatla
1714af0618 YARN-2452. TestRMApplicationHistoryWriter fails with FairScheduler. (Zhihai Xu via kasha)
(cherry picked from commit c50fc92502934aa2a8f84ea2466d4da1e3eace9d)
2014-09-21 13:15:45 -07:00
Jian He
07981c8f60 YARN-2565. Fixed RM to not use FileSystemApplicationHistoryStore unless explicitly set. Contributed by Zhijie Shen
(cherry picked from commit 444acf8ea795e4bc782f1ce3b5ef7a1a47d1d27d)
2014-09-19 11:28:22 -07:00
Zhijie Shen
9333ee3068 YARN-2568. Fixed the potential test failures due to race conditions when RM work-preserving recovery is enabled. Contributed by Jian He.
(cherry picked from commit 6fe5c6b746a40019b9a43676c33efec0f971c4b9)
2014-09-18 22:02:53 -07:00
Jason Lowe
d9273a9547 YARN-2561. MR job client cannot reconnect to AM after NM restart. Contributed by Junping Du
(cherry picked from commit a337f0e3549351344bce70cb23ddc0a256c894b0)
2014-09-18 21:36:20 +00:00
Jason Lowe
44c22c3d8d YARN-2363. Submitted applications occasionally lack a tracking URL. Contributed by Jason Lowe
(cherry picked from commit 9ea7b6c063c0bdd4551962e21d0173f671e9df03)
2014-09-18 20:15:46 +00:00
Vinod Kumar Vavilapalli
d24ef142b7 YARN-2001. Added a time threshold for RM to wait before starting container allocations after restart/failover. Contributed by Jian He.
(cherry picked from commit 485c96e3cb9b0b05d6e490b4773506da83ebc61d)
2014-09-18 11:04:27 -07:00
Jian He
d61cdd66db YARN-2559. Fixed NPE in SystemMetricsPublisher when retrieving FinalApplicationStatus. Contributed by Zhijie Shen
(cherry picked from commit ee21b13cbd4654d7181306404174329f12193613)
2014-09-17 21:44:49 -07:00
junping_du
1e1f4c2fdf YARN-1250. Generic history service should support application-acls. (Contributed by Zhijie Shen)
(cherry picked from commit 90a0c03f0a696d32e871a5da4560828edea8cfa9)
2014-09-16 18:30:06 -07:00
Vinod Kumar Vavilapalli
5cdb24d4b1 YARN-611. Added an API to let apps specify an interval beyond which AM failures should be ignored towards counting max-attempts. Contributed by Xuan Gong.
(cherry picked from commit 14e2639fd0d53f7e0b58f2f4744af44983d4e867)
2014-09-13 18:05:02 -07:00
XuanGong
add5ac6a73 YARN-2456. Possible livelock in CapacityScheduler when RM is recovering
apps. Contributed by Jian He
2014-09-12 15:24:02 -07:00
Jian He
f21a5236e5 YARN-2229. Changed the integer field of ContainerId to be long type. Contributed by Tsuyoshi OZAWA
(cherry picked from commit 3122daa80261b466e309e88d88d1e2c030525e3f)

Conflicts:
	hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-app/src/main/java/org/apache/hadoop/mapreduce/v2/app/rm/preemption/CheckpointAMPreemptionPolicy.java
2014-09-12 10:45:42 -07:00
Jason Lowe
7e69dc8771 YARN-2484. FileSystemRMStateStore#readFile/writeFile should close FSData(In|Out)putStream in final block. Contributed by Tsuyoshi OZAWA
(cherry picked from commit 78b048393a80a9bd1399d08525590bb211a32d8c)
2014-09-12 14:10:51 +00:00
Zhijie Shen
466bdf79c5 YARN-2538. Added logs when RM sends roll-overed AMRMToken to AM. Contributed by Xuan Gong.
(cherry picked from commit 469ea3dcef6e427d02fd08b859b2789cc25189f9)
2014-09-11 22:21:59 -07:00
junping_du
efa9ab3638 YARN-2033. Merging generic-history into the Timeline Store (Contributed by Zhijie Shen)
(cherry picked from commit 6b8b1608e64e300e4e1d23c60476febaca29ca38)
2014-09-12 10:08:06 +08:00
Karthik Kambatla
6f8b2a8e2e YARN-2534. FairScheduler: Potential integer overflow calculating totalMaxShare. (Zhihai Xu via kasha)
(cherry picked from commit c11ada5ea6d17321626e5a9a4152ff857d03aee2)
2014-09-11 12:06:28 -07:00
Jian He
ceae7be6b2 YARN-415. Capture aggregate memory allocation at the app-level for chargeback. Contributed by Eric Payne & Andrey Klochkov
(cherry picked from commit 83be3ad44484bf8a24cb90de4b9c26ab59d226a8)
2014-09-10 18:22:07 -07:00
Jian He
f58a076dbc YARN-2158. Fixed TestRMWebServicesAppsModification#testSingleAppKill test failure. Contributed by Varun Vasudev
(cherry picked from commit cbfe26370b85161c79fdd48bf69c95d5725d8f6a)
2014-09-10 12:48:40 -07:00
XUAN
b91ef0c502 YARN-2459. RM crashes if App gets rejected for any reason and HA is enabled. Contributed by Jian He 2014-09-10 11:51:57 -07:00
Vinod Kumar Vavilapalli
b4b59ef749 YARN-2448. Changed ApplicationMasterProtocol to expose RM-recognized resource types to the AMs. Contributed by Varun Vasudev.
(cherry picked from commit b67d5ba7842cc10695d987f217027848a5a8c3d8)
2014-09-10 10:17:30 -07:00
Karthik Kambatla
a19694f19d YARN-1458. FairScheduler: Zero weight can lead to livelock. (Zhihai Xu via kasha)
(cherry picked from commit 3072c83b38fd87318d502a7d1bc518963b5ccdf7)
2014-09-10 08:33:31 -07:00
Karthik Kambatla
96a13c6d0c YARN-2394. FairScheduler: Configure fairSharePreemptionThreshold per queue. (Wei Yan via kasha)
(cherry picked from commit 1dcaba9a7aa27f7ca4ba693e3abb56ab3c59c8a7)
2014-09-03 10:27:47 -07:00
Karthik Kambatla
7b9e763138 YARN-2395. FairScheduler: Preemption timeout should be configurable per queue. (Wei Yan via kasha)
(cherry picked from commit 0f34e6f3873aee0f4932740ca790c6dd2a13b5d9)
2014-08-30 01:22:44 -07:00
Jian He
38853c97c2 YARN-1506. Changed RMNode/SchedulerNode to update resource with event notification. Contributed by Junping Du
(cherry picked from commit 5c14bc426b4be381383018ebc2236be83eef15cd)

Conflicts:
	hadoop-yarn-project/CHANGES.txt
2014-08-29 23:39:58 -07:00
Jian He
73a0e46657 YARN-2447. RM web service app submission doesn't pass secrets correctly. Contributed by Varun Vasudev
(cherry picked from commit c686aa3533b42e1baf62a78bc1bfb0ac05be53bb)
2014-08-29 11:44:27 -07:00