Commit graph

2625 commits

Author SHA1 Message Date
Anil Choudhary
1dda659918
[tune]TrialRunner wait on global checkpoint syncdown (#8798) 2020-06-05 10:39:59 -07:00
Edward Oakes
4155d5830f
[serve] Replace actor error retry logic with max_task_retries (#8768) 2020-06-05 10:45:28 -05:00
Amog Kamsetty
9410e5884d
[Tune] Parametrize Cloud Syncing Frequency (#8771) 2020-06-04 18:55:50 -07:00
Ian Rodney
d452932740
[autoscaler] Improve Logtimer log messages (#8753) 2020-06-04 18:07:27 -07:00
Ameer Haj Ali
d966d98729
cleanup to support provider's custom ssh command runner (#8720)
* cleanup to support provider's custom ssh command runner

* clean up

* trailing white spaces fix

* k8s signature fix

Co-authored-by: Ameer Haj Ali <ameerhajali@Ameers-MacBook-Pro.local>
2020-06-04 13:27:17 -07:00
Ian Rodney
09f89ff49d
[autoscaler] Improve SSH Command Failure Logging (#8751) 2020-06-04 12:38:20 -07:00
SangBin Cho
e372c06257
Hotfix dashboard broken tests. (#8757) 2020-06-04 09:44:00 -07:00
fangfengbin
84a8f2ccb5
Support reloading storage data when gcs server restarts (#8650) 2020-06-04 14:53:20 +08:00
Eric Liang
a24d117c68
[autoscaler] Refactor code in preparation for multi instance type support (#8632)
* wip refactor

* add util

* wip

* fix

* fix

* remove

* remove extraneous string type for sg
2020-06-03 12:53:55 -07:00
Ian Rodney
474bbc28bf
Warn if Autoscaling-config flag not set. (#8677) 2020-06-03 12:21:07 -07:00
Ian Rodney
7a2c9524d1
[Core] Randomize and 'Reserve' Port Generated for Node Manager (#8628) 2020-06-03 12:19:03 -07:00
Siyuan (Ryans) Zhuang
7fa64f2b24
Clean up unused Python code (#8755) 2020-06-03 12:09:19 -07:00
Max Fitton
b9f0f7ae5b
Dashboard minor refactor and first unit tests (#8705) 2020-06-03 11:04:55 -05:00
krfricke
f4ee3e76d8
[tune] last-n-avg
Co-authored-by: Kai Fricke <kai@anyscale.com>
2020-06-02 20:06:04 -07:00
SangBin Cho
7c43991100
[GCS] Monitor.py bug fix (#8725)
* comment.

* Fix bugs.

* Used pubsub message instead.

* Added a ray.actors test
2020-06-02 16:06:36 -07:00
Edward Oakes
0306e4d589
[serve] Refer to serve "instances," not "clusters" (#8746) 2020-06-02 15:16:29 -07:00
Edward Oakes
2e82e05e4b
[serve] Add list_backends and list_endpoints (#8737) 2020-06-02 15:14:10 -07:00
Simon Mo
c5544eb070
[Async] Remove Monitor + Cleanup Code (#8691) 2020-06-02 14:11:16 -07:00
Edward Oakes
e91f095d98
[Serve] Remove ray_init_kwargs in serve.init (#8747) 2020-06-02 14:05:35 -07:00
krfricke
4d0e9f3c71
[Dashboard] Sort IDLE workers to bottom in dashboard (#8708)
* Sort IDLE workers to bottom in dashboard

* Fixed linting error

Co-authored-by: Kai Fricke <kai@anyscale.com>
2020-06-02 14:00:59 -07:00
Stephanie Wang
aa06c3b15a
Eager eviction even when object pinning is disabled (#8561)
* Eager eviction even when object pinning is disabled, add regression test

* Make test more robust

* lint
2020-06-02 11:48:03 -07:00
Edward Oakes
ae312af435
Remove accidental passes in rllib, tune (#8742) 2020-06-02 12:29:17 -05:00
Edward Oakes
57bf0e43f0
fix docstring (#8736) 2020-06-02 08:55:20 -07:00
Lingxuan Zuo
4cbbc15ca7
[GCS] Global state accessor from node resource table (#8658) 2020-06-02 14:01:00 +08:00
Alec Brickner
207ab44129
Raise major version limit for msgpack (#8466) 2020-06-01 20:00:36 -07:00
Alex Wu
a2ec282033
[Doc] Dataset lint fix (#8719) 2020-06-01 19:43:06 -07:00
Simon Mo
4cef1ee591
[Serve] Cleanup Router Implementation (#8718) 2020-06-01 19:21:28 -07:00
Alex Wu
dcf58a43dc
[SGD] Dataset API (#7839) 2020-06-01 15:48:15 -07:00
SangBin Cho
cd5a207d69
[Dashboard] Frontend Lint Fix. (#8696) 2020-06-01 11:29:01 -07:00
fangfengbin
016337d4eb
Heartbeat table uses gcs pub-sub instead of redis accessor (#8655) 2020-05-30 23:17:25 +08:00
Siyuan (Ryans) Zhuang
ebea5c4111
Update cloudpickle to version 1.4.1 (#8577) 2020-05-29 17:55:48 -07:00
SangBin Cho
3ee3e64de0
[Dashboard] Ray memory frontend (#8563) 2020-05-29 19:02:09 -05:00
SangBin Cho
1115231e7c
[Test] Fix test_reconstruction OOM error (#8636) 2020-05-29 18:56:19 -05:00
Edward Oakes
5bec951ece
[docs] [serve] Deployment as a service on k8s docs (#8663) 2020-05-29 14:53:42 -07:00
krfricke
e5b6566d28
Remove blocking flag from serve.init() (#8654) 2020-05-29 13:25:35 -07:00
Thomas Desrosiers
457a66ae9c
Reverts setup.py changes from 76450c8d4 (#8670) 2020-05-29 13:24:32 -07:00
Edward Oakes
30ed20405a
[autoscaler] Support creating services in k8s backend (#8659) 2020-05-29 15:19:21 -05:00
Simon Mo
6b04664645
[Serve] Add Tutorial for Batch Inference (#8490) 2020-05-29 09:55:47 -07:00
fangfengbin
35eeec5647
Add C++ global state for actor table (#8501)
* add global state actors

* fix code style

* fix GcsActorManagerTest bug

* rebase master

* add jni code

* add get checkpoint id code

* add debug code

* add debug code

* change log level

* fix compile bug

* return null in jni

* fix crash bug

* change import seq

Co-authored-by: 灵洵 <fengbin.ffb@antfin.com>
Co-authored-by: Hao Chen <chenh1024@gmail.com>
2020-05-29 21:10:42 +08:00
Sven Mika
d483ed28ba
[RLlib] Fix broken tune tests in master due to framework=auto errors. (#8672) 2020-05-29 11:55:47 +02:00
Edward Oakes
c64b694560
Update RaySGD test to use ray.kill instead of __ray_kill__ (#8662) 2020-05-28 22:38:05 -05:00
Patrick Ames
76450c8d47
[autoscaler] Honor separate head and worker node subnet IDs (#8374) 2020-05-28 18:16:46 -07:00
Ian Rodney
99cc2e28b3
[Core] Fix test_cancel (#8644) 2020-05-28 09:40:37 -07:00
Hao Chen
08fee00bc8
Increase rayelt client connect timeout to fix test_debug_tools (#8605) 2020-05-28 20:57:30 +08:00
Lingxuan Zuo
e594524ed3
[GCS] global state query node info table from GCS. (#8498) 2020-05-28 16:39:13 +08:00
Ujval Misra
e958d261b6
Fix ray.available_resources bug (#8537) 2020-05-27 17:55:08 -07:00
Simon Mo
38399c9885
[Hotfix] [Serve] Disable Deployment Tutorial Test (#8641) 2020-05-27 10:40:40 -07:00
Bill Chambers
fadd47e44e
[docs] Ray Serve Documentation Overhaul (#8524) 2020-05-27 11:03:28 -05:00
fyrestone
b0bb0584fb
Fix fix_test_actor_method_metadata_cache (#8633) 2020-05-27 15:49:40 +08:00
mehrdadn
79a4eac48c
Make more tests run on Windows (#8553) 2020-05-26 18:43:34 -05:00