Commit graph

5598 commits

Author SHA1 Message Date
Ian Rodney
d76d4822bd
[Docker] Fix building images (#10393) 2020-08-28 10:41:02 -07:00
Lixin Wei
eb66db3199
[Build] bug fixed for logging (#10364) 2020-08-28 09:17:08 -07:00
SangBin Cho
68c2dcd12b
Fix. (#10390) 2020-08-28 08:22:23 -07:00
Edward Oakes
c3ed403def
fix typo (#10382) 2020-08-28 09:57:04 -05:00
SangBin Cho
d206fbbc99
[Placement group] Scheduler map refactoring part 1. (#10381)
* In Progress

* done.

* Address code review.
2020-08-28 00:57:09 -07:00
SangBin Cho
7b29eb7949
[Build] Try parallel Python builds. (#10291)
* Trial 1.

* Parallelize even more.
2020-08-28 00:06:52 -07:00
SongGuyang
cb70864c04
[cpp worker] support cluster mode and object Put/Get works (#9682) 2020-08-28 13:53:36 +08:00
Richard Liaw
0d22c0b653
[tune] Avoid recreating actor multiple times (#10374) 2020-08-27 18:02:26 -07:00
Richard Liaw
922bf9f45a
[cli] improve error handling, don't swallow errors (#10370) 2020-08-27 17:59:44 -07:00
Richard Liaw
ed5de89470
FIX: Lint (#10384) 2020-08-27 17:56:39 -07:00
Ian Rodney
465d4c50b6
[docker push fix] (#10375) 2020-08-27 17:07:48 -07:00
SangBin Cho
f35339b5ff
[Dashboard] Change default ip address for the dashboard to ipv4 (#10287)
* Done.

* Add todo.

* Addressed code review.

* Fix issue.

* Fix test failure.

* Fix a test.
2020-08-27 14:43:10 -07:00
Alex Wu
7dbc1f439c
[hotfix] Autoscaler monitor fix unit tests 2020-08-27 14:26:41 -07:00
Alex Wu
76898d4ebc
[Autoscaler][hotfix] Remove additionalProperties from available_node_types schema (#10366) 2020-08-27 13:56:44 -07:00
Eric Liang
bd245a1c18
[api] Clean up and document Actor name / lifetime API (#10332) 2020-08-27 13:38:39 -07:00
SangBin Cho
17f465d5c1
[Core] Improve raylet failure error msg (#10345)
* Improve error message.

* Lint.

* Addressed code review.
2020-08-27 12:53:18 -07:00
Eric Liang
583ad38f8f
remove shellcheck bazel (#10369) 2020-08-27 12:36:57 -07:00
Clark Zinzow
0178d6318e
[Core] Expand job ID to 4 bytes by removing object flag bytes. (#10187) 2020-08-27 14:08:17 -05:00
SangBin Cho
f846b26165
[Doc] Add ports configuration & Move dashboard to the original place (#10281)
* Done.

* Address code review.

* Addressed code review.
2020-08-27 12:00:16 -07:00
Philipp Moritz
b8673e5697
[autoscaler] Make KeyName optional in AWS autoscaler (#10336) 2020-08-27 11:08:44 -07:00
architkulkarni
eea7a86163
[Serve] add type hints for controller and backend_worker (#10288) 2020-08-27 10:20:36 -07:00
Stephanie Wang
f75dfd60a3
[api] API deprecations and cleanups for 1.0 (internal_config and Checkpointable actor) (#10333)
* remove

* internal config updates, remove Checkpointable

* Lower object timeout default

* remove json

* Fix flaky test

* Fix unit test
2020-08-27 10:19:53 -07:00
Amog Kamsetty
0aec4cbccb
[Tune] Update PBT Transformers Example (#10289)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
Co-authored-by: krfricke <krfricke@users.noreply.github.com>
2020-08-27 08:25:05 -07:00
krfricke
53ab228b75
[tune] Fix log to file on actor reuse (#10363) 2020-08-27 08:22:19 -07:00
Alex Wu
6d2af33a01
[Autoscaler] Proper resource demand plumbing (#10329) 2020-08-26 23:36:01 -07:00
Ian Rodney
9056854c06
drop keep alive (#10347) 2020-08-26 21:15:48 -07:00
Edward Oakes
60665fc936
Clean up task dependency and scheduler metrics (#10340) 2020-08-26 22:56:03 -05:00
Lixin Wei
fe6daef85e
[Core]Add runtime context for python worker (#10309)
* add runtime context for python

* fixed

* code fixed

* test added

* lint

* lint
2020-08-26 20:11:42 -07:00
Ian Rodney
2526c06b5e
[WIP] [docker] Cleanup Docker Base-Deps (#9988)
* cleanup-base deps

* only build base-deps a bit

* remove parens

* formatting

* add ray-deps

* gpu enabled

* always include wheel

* fix script

* log new variables

* run tests for docker

* try to include env variables

* source files

* remove bash when sourcing

* add new lines

* use wget

* dual build autoscaler

* switch to gnupg

* add gcc cmake

* remove blist

* clarify build-docker-images
2020-08-26 19:36:11 -07:00
Ian Rodney
e2eef6469b
Deprecate Jenkins (#10314) 2020-08-26 15:43:27 -07:00
Ameer Haj Ali
17c8c63e7e
Metadata schema (#10328)
* metadata

* Eric

Co-authored-by: Ameer Haj Ali <ameerhajali@Ameers-MacBook-Pro.local>
2020-08-26 15:43:03 -07:00
Richard Liaw
29e8a664c4
[cli] make sure old-style works (#10344) 2020-08-26 15:26:24 -07:00
Lixin Wei
4b856fa416
[Core]Async updating issue fixed for actor's num_restart (#10176)
* bug fixed for num_restart updating

* add log

* log updated

* lint

* fixed

* Update src/ray/gcs/gcs_server/gcs_actor_manager.cc

Co-authored-by: Stephanie Wang <swang@cs.berkeley.edu>

* bug fixed

* bug fixed

* test passed

Co-authored-by: Stephanie Wang <swang@cs.berkeley.edu>
2020-08-26 11:49:26 -07:00
Edward Oakes
c35ad8237d
[metrics] Clean up object manager stats (#10316) 2020-08-26 13:43:06 -05:00
Ian Rodney
dc378a80b7
[autoscaler/docker] Docker Inititialization Revamp (#9515)
* Basic idea

* Small fixes

* dockerize start commands in Command Runner

* Remove run_init from CommandRunnerInterface

* Add Parens

Co-authored-by: Simon Mo <simon.mo@hey.com>

* Cleaning up

* Response to richards comments

* Further small fixes

* Fix Json

* schema format fix

* cleanup

* run more often

* fix indent

* Fix richards responses

* fix ups

* remove docker_commands from schema

* default to list

* fix docker cmd runner test

* lint fix

Co-authored-by: Simon Mo <simon.mo@hey.com>
2020-08-26 10:29:06 -07:00
Edward Oakes
916a19363f
Clean up actor metrics (#10317) 2020-08-26 10:21:15 -05:00
Sven Mika
93120e0347
Unity3D API Fixes (recent changes in Unity's MLAgents API caused errors on RLlib side). (#10285) 2020-08-26 14:16:08 +02:00
Michael Luo
4e9888ce2f
[RLlib] Dreamer (#10172) 2020-08-26 13:24:05 +02:00
Alex Wu
9ca159aa0b
[Autoscaler] Multi node commands (#10236) 2020-08-25 23:35:38 -07:00
Olli Huotari
0dae50b5eb
Fixed num_atoms>1 in pytorch (#10330) 2020-08-25 23:10:20 -07:00
Amog Kamsetty
8c0503ddd3
[Tune] Convert PBT DCGAN Example to Function API (#10246)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-08-25 22:34:19 -07:00
Antoni Baum
87ed20738e
[tune] Add on_pause, on_unpause to ConcurrencyLimiter (#10320)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-08-25 22:33:17 -07:00
Simon Mo
ed3fdd2c0b
[Serve] Remove register_custom_serializer (#10331) 2020-08-25 21:20:43 -07:00
Edward Oakes
cbd9632f3a
Fix wait timeout logic (#10199) 2020-08-25 22:41:39 -05:00
fyrestone
08adbb371f
Cross language exception (#10023) 2020-08-26 10:46:05 +08:00
Edward Oakes
1e99b814f0
Remove unused scheduler states (#10318)
* remove unused state

* remove unused states
2020-08-25 18:56:21 -07:00
Eric Liang
deea1861ab
[rllib] Try fixing torch GPU and masking errors (#10168) 2020-08-25 18:34:19 -07:00
Eric Liang
6fcb816fdd
Ray operator deprecation message (#10334) 2020-08-25 18:26:02 -07:00
Robert Nishihara
79eefbf357
Better checking that ray.init() has been called. (#10261) 2020-08-25 17:13:11 -07:00
Stephanie Wang
d4537ac1ce
[core] Try to schedule tasks locally before spilling over to remote nodes (#10302)
* Regression test

* Spillback

* Remove check for actor tasks
2020-08-25 15:01:59 -07:00