Simon Mo
b125cd95b4
[Serve] Cleanup ThreadProxiedRouter and strictly scope endpoint to router ( #14911 )
2021-04-07 16:59:24 -07:00
Simon Mo
f0ba15c2d3
[Serve] Check for endpoint in the controller ( #15087 )
2021-04-07 14:16:19 -07:00
Edward Oakes
259813d8bd
[serve] Add list_deployments API ( #15152 )
2021-04-07 15:36:08 -05:00
Dmitri Gekhtman
158c3aa26f
[client] support ray.get_gpu_ids() ( #15173 )
...
since this is running on the driver, it will just return an empty list `[]`
2021-04-07 23:02:07 +03:00
Edward Oakes
59cdc30615
[packaging] Remove stub_file deprecation path ( #15171 )
2021-04-07 14:04:25 -05:00
Edward Oakes
eed34092f2
[serve] Defer building starlette request to backend replica ( #15169 )
2021-04-07 13:28:53 -05:00
Edward Oakes
195f818224
[serve] Move route to deployment options ( #15124 )
2021-04-07 10:50:11 -05:00
Edward Oakes
70195e2114
[serve] Make serve.start() idempotent ( #15148 )
2021-04-07 08:51:37 -05:00
Edward Oakes
65f0bae47f
[serve] Default deployment name to callable name ( #15121 )
2021-04-07 08:48:27 -05:00
SangBin Cho
1df60a61a2
Skip serve test that is not working in the master ( #15163 )
2021-04-06 23:17:22 -07:00
Dmitri Gekhtman
d0c83d5ecd
[kubernetes][autoscaler] Underscore for rsync file ( #15155 )
2021-04-06 23:04:31 -07:00
SangBin Cho
e0872083b8
[Pubsub] Generalize pubsub impl part 1 ( #15116 )
...
* Finished the implementation. Cpp tests are left.
* Fix cpp tests.
* Addressed code review.
* Addressed code review.
* Change the destruction order.
* Addressed code review part 2.
2021-04-06 20:59:32 -07:00
SangBin Cho
e624476b78
[Test] skip failing tests in test_actor_resources. ( #15151 )
2021-04-06 20:57:03 -07:00
Edward Oakes
2274a94453
[serve] Remove deprecated system-level batching ( #15065 )
2021-04-06 17:58:42 -05:00
Yi Cheng
35b45a03cf
[runtime_env] Add working_directory into sys.path ( #15145 )
2021-04-06 17:49:06 -05:00
Alex Wu
10fdb9e9ac
[metrics] Scheduler metrics ( #14716 )
2021-04-06 11:27:54 -07:00
Stefan Schneider
49ba51979e
Functions for restoring from last or best checkpoint ( #14735 )
...
Adds a helper function to retrieve the latest checkpoint after selecting the best trial according to a metric.
2021-04-06 12:19:09 +02:00
Edward Oakes
43932f5330
[serve] Health checking in the controller instead of using max_restarts ( #15047 )
2021-04-05 21:22:40 -05:00
Clark Zinzow
f17e59093d
[dask-on-ray] Fix Dask-on-Ray scheduler break caused by changing internal Dask API. ( #15100 )
2021-04-05 17:58:44 -07:00
Yi Cheng
afc92130fa
Unpack runtime env to runtime_resource ( #15111 )
2021-04-05 17:35:31 -07:00
Dmitri Gekhtman
ee291ae762
[kubernetes][test][minor] Fix K8s test by an adding an underscore. ( #15122 )
2021-04-05 15:18:20 -07:00
Ian Rodney
29566c3527
[Code Cleanup] Rename to Allowlist + Denylist ( #15117 )
...
Co-authored-by: Simon Mo <simon.mo@hey.com>
2021-04-05 13:58:18 -07:00
Ian Rodney
c1bffc2784
[Core] Pick Up GCS Port From Env Variable ( #15097 )
2021-04-05 13:57:55 -07:00
Richard Liaw
6269140ef8
[minor] improve warning message for Ray ( #15005 )
2021-04-05 12:39:42 -07:00
Yi Cheng
5806e726f4
[core] Internal kv support in gcs ( #14656 )
...
* server side ready
* client size
* py
* fix
* up
* format
* add files
* add pyx
* up
* up
* up
* add keys
* format
* update
* format
* add unittests
* add files
* up
* up
* fix
* up
* fix thread issue
* format
* fix
* Fix
* format
* fix
* more
* fix conflict
* fix
* fix order
* format
* compiling fix
* lint
* fix
* fix some
* some fix
* fix comment
* fix name
* format
* fix compatible issue
* fix name
* fix lint
* disconnect safe
* up
* format
* fix
Co-authored-by: Yi Cheng <singye888@gmail.com>
2021-04-05 10:26:46 -07:00
Richard Liaw
e72f6b0377
Fix ray[full] -> ray[cluster] #15112
...
Signed-off-by: Richard Liaw <rliaw@berkeley.edu>
2021-04-05 09:55:00 -07:00
valiantljk-tt
9b85b0b82b
[runtime_env] Continue hashing if data is none ( #14906 )
2021-04-05 11:24:09 -05:00
architkulkarni
0d1b30620b
[Serve] Un-revert #15000 Set controller and HTTP proxy num_cpus=0 by default ( #15092 )
2021-04-05 09:58:55 -05:00
Dmitri Gekhtman
47a3bd7b67
[kubernetes][autoscaler] Rename autoscaler K8s module ( #15101 )
2021-04-04 15:30:00 -07:00
Dmitri Gekhtman
e0ed549006
return False, don't raise ( #15107 )
2021-04-03 14:04:50 -07:00
Yi Cheng
1920db0a64
Deflak dask related task on win32 ( #15102 )
...
* disable dask test on win32
* disable dask test on win32
* disable dask test on win32
* disable dask test on win32
2021-04-03 12:26:29 -07:00
Richard Liaw
e9f365ad2e
[tune] support skopt/bayesopt search restore ( #15075 )
2021-04-03 12:24:30 -07:00
Alex Wu
aea28c53ce
. ( #15093 )
...
Co-authored-by: Alex <alex@anyscale.com>
2021-04-02 16:39:38 -07:00
Yi Cheng
4caf7a511d
Deflaky test failure in win32 ( #15090 )
...
* deflake win32 failure
* skip dask on w32
2021-04-02 14:56:24 -07:00
architkulkarni
d8f8583e80
Revert "[Serve] Set controller and HTTP proxy num_cpus=0 by default ( #15000 )" ( #15091 )
...
This reverts commit 39aa01fc2c
2021-04-02 13:01:57 -07:00
Yi Cheng
ecb94b3fe9
Add test case to check job conf compatible issue ( #15082 )
2021-04-02 12:03:21 -07:00
Dmitri Gekhtman
42565d5bbe
[autoscaler] Fix update/terminate race condition ( #15019 )
...
Co-authored-by: AmeerHajAli <ameerh@berkeley.edu>
2021-04-02 11:57:02 -07:00
Edward Oakes
96cc7897f7
[serve] Use longest prefix matching for path routing ( #15041 )
2021-04-02 12:01:47 -05:00
architkulkarni
39aa01fc2c
[Serve] Set controller and HTTP proxy num_cpus=0 by default ( #15000 )
2021-04-02 12:01:22 -05:00
Dmitri Gekhtman
6f81ec1998
[kubernetes][test] Operator test tweaks. ( #15074 )
2021-04-02 09:20:52 -07:00
Kai Fricke
8de66fce3d
[tune] Improve BOHB/ConfigSpace dependency check ( #15064 )
2021-04-02 10:19:49 +02:00
Alex Wu
f52c855704
[core] Fix placement group GPU assignment bug ( #15049 )
2021-04-01 17:46:09 -07:00
Yi Cheng
d4c20c970b
[core] Fix UTIL worker issue ( #14925 )
...
* Fix
* format
* more
* format
* fix
* fix
* fix comment
* fix test failure
2021-04-01 17:36:45 -07:00
Simon Mo
c9dac9328e
[Serve] Fix serializing nested fields in Pydantic ( #15069 )
2021-04-01 17:20:34 -07:00
Alex Wu
4fba05ae4d
[core] Hybrid scheduling policy. ( #14790 )
2021-04-01 16:59:59 -07:00
Edward Oakes
346994745a
[serve] Get handle in starlette endpoint constructor instead of lazily ( #15066 )
2021-04-01 16:07:28 -05:00
Ian Rodney
22c1aeb240
[Tests] Skip autoscaler tests on Windows ( #15033 )
2021-04-01 10:16:42 -07:00
SangBin Cho
005cff0092
Revert "Revert "[Core] Implement long polling-based pubsub to reduce … ( #14909 )
2021-04-01 09:03:15 -07:00
Kai Fricke
d33b0e4bc3
[tune] Reconcile placement groups every N seconds to avoid bottlenecks when running many short trials ( #15011 )
...
Closes a release blocking issue
2021-04-01 17:04:44 +02:00
Hao Chen
3e1a0439b7
Fix concurrent actor starting too many threads. ( #14927 )
2021-04-01 19:58:18 +08:00
