…ment (ray-project#2995)

* simplify vec batch requirements

* Update rllib-training.rst

* Update rllib-training.rst

* Update rllib-training.rst

* Update rllib-training.rst

* Update rllib-training.rst

* Update rllib-models.rst

…ors (ray-project#2967)

* update

* link it

* warn about truncation

* fix

* Update rllib-training.rst

* deprecate tests failing

… of PPO (ray-project#2974)

* fix

* fix

* fix it

* propagate conf to action dist

* move carla example too

* rr

* Update policies.py

* wip

* lint

…ava worker. (ray-project#3002)

This fixes a bug in which Java actor methods inherit the resource requirements of the actor creation task.

* remove legacy

* remove reshaper

This tests the case in which a worker is blocked in a call to ray.get or ray.wait, and then the worker dies. Then later, the object that the worker was waiting for becomes available. We need to make sure not to try to send a message to the dead worker and then die. Related to ray-project#2790.

before fix,RAY_FUN_CACHE use only get method ,can only get null
  fix : put after create

…in. (ray-project#2862)

…t#3018)

…ay-project#3003)

Move function/actor exporting & loading code to function_manager.py to prepare the code change for function descriptor for python.

This commit fix some small defects. 
1. Remove a comment that should have been removed in ray-project#3003
2. Remove `redis_protected_mode` that is never used in `ray.init()`
3. Fix `object_id_seed` that is forgotten to be passed into `ray._init()`
4. Remove several redundant brackets.

…-project#2935)

…ect#2837)

* Introduce concept of resources required for placement.
* Add placement resources to task spec
* Update java worker
* Update taskinfo.java

…ject#3029)

Improve logging message when plasma store is started.

* Update rsync command

* Escape rsync locations

* Fix the accidental variable move

* Update rsync to use -s flag

## What do these changes do?
1. Add a configuration item `driver.resource-path`.
2. Load driver resources from the local path which is specified in the `ray.conf`.

Before this change, we should add all driver resources(like user's jar package, dependencies package and config files) into `classpath`.

After this change, we should add the driver resources into the mount path which we can configure it in `ray.conf`, and we shouldn't configure `classpath` for driver resources any more.

## Related issue number
N/A

* bugfix: env exists check error

* support to avoid re-build pyarrow in project

* bugfix: adapt gtest for centos lib64

* bugfix: check gtest lib exists in the directory

* bugfix: find gtest with checking all libs exists

* prefix RAY_ to thirdparty env variables to avoid conflicts with other module

* arrow use glog from ray

* change the glog and gtest install dir

This PR improves some java codes,  and removes some duplicated code.

## What do these changes do?
Fix the issue how we load driver resources by a specified path.
Also this addressed the comments from the related PR [3044](ray-project#3044).

## Related PRs:
 [ray-project#3044](ray-project#3044) and [ray-project#3001](ray-project#3001).

…l plasma java lib (ray-project#3047)

…y states (ray-project#3032)

* fix er

* update

…roject#2766)

…ight (ray-project#3061)

…oject#3068)

## What do these changes do?

Fix the misleading comments in code for:
 - `EPISODES_THIS_ITER`
 - `EPISODES_TOTAL`

Had noted it before and planned to fix it along with some other changes but seemed very relevant to stay next to ray-project#3058 so sending this now.

* Implement Redis authentication

* Throw exception for legacy Ray

* Add test

* Formatting

* Fix bugs in CLI

* Fix bugs in Raylet

* Move default password to constants.h

* Use pytest.fixture

* Fix bug

* Authenticate using formatted strings

* Add missing passwords

* Add test

* Improve authentication of async contexts

* Disable Redis authentication for credis

* Update test for credis

* Fix rebase artifacts

* Fix formatting

* Add workaround for issue ray-project#3045

* Increase timeout for test

* Improve C++ readability

* Fixes for CLI

* Add security docs

* Address comments

* Address comments

* Adress comments

* Use ray.get

* Fix lint

…mizer (ray-project#2993)

…ct#3074)

Follow-up to ray-project#2925 and ray-project#2952. Removes the Redis protected mode implementation from Ray which was replaced by Redis port authentication.

* Support setting logging level from env variable

* Remove Env Variable related code

* lint

* Adding Python3.7 wheels support

* Adding Mac wheels update

* fix

* numpy version

* choose different numpy versions depending on python version

* fix

…onously (ray-project#3023)

* copy over ref code

* wip async writes

* compiles

* fix error handling

* add test

* amend

* fix test

* clang fmgt

* clang format

* wip

* yapf

* rename format script

* test error

* clangfmt

* add test to list

* warn

* ref test

* fix test

* comment

* add capture

* Update client_connection.cc

* wip

* fix compile

…ay-project#3070)

## What do these changes do?
Before this PR, if we want to specify some resources, we must do as following codes:
```java
@rayremote(Resources={ResourceItem("CPU", 10)})
public static void f1() {
// do sth
}

@rayremote(Resources={ResourceItem("CPU", 10)})
class Demo {
// sth
}
```
Unfortunately, it's no way for us to create another actor or task with different resources required.

After this PR, the thing will be:
```java
ActorCreationOptions option = new ActorCreationOptions(); 
option.resources.put("CPU", 4.0);
RayActor<Echo> echo1 = Ray.createActor(Echo::new, option);
option.resources.put("Res-A", 4.0);
RayActor<Echo> echo2 = Ray.createActor(Echo::new, option);


//if we don't specify resource,  the resources will be `{"cpu":0.0}` by default.
Ray.call(Echo::echo, echo2, 100);
```


## Related issue number
N/A

<!--
Thank you for your contribution!

Please review https://github.com/ray-project/ray/blob/master/CONTRIBUTING.rst before opening a pull request.
-->

## What do these changes do?
remove TaskExecutionException, use RayException instead
<!-- Please give a short brief about these changes. -->

## Related issue number

<!-- Are there any issues opened that will be resolved by merging this change? -->

…#3091)

…aces; add prev a, r to LSTM (ray-project#3051)

…ay-project#3080)

## What do these changes do?
Fill driver id into actor notification when finishing assigned task.
Also it improves codes.

* logg

* set rllib logger

* comment

* info

* rlib

* comment

* add format

* fix lint

* add file info

* update

* add ts

* lint

* better docs

* fix value error

* soft log level

* Fix trial search alg finishing early

* Fix lint

* fix lint

* nit fix

…ect#3105)

This is fixing a problem that @devin-petersohn observed on the windows subsystem for linux.

In theory, redis should be up once the async connect is happening and there should be no retries needed for the async connect. However on the windows subsystem for linux, the async connect was failing even though the synchronous one was working. Maybe windows has a different semantics here than linux.

* Use XRay backend by default.

* Remove irrelevant valgrind tests.

* Fix

* Move tests around.

* Fix

* Fix test

* Fix test.

* String/unicode fix.

* Fix test

* Fix unicode issue.

* Minor changes

* Fix bug in test_global_state.py.

* Fix test.

* Linting

* Try arrow change and other object manager changes.

* Use newer plasma client API

* Small updates

* Revert plasma client api change.

* Update

* Update arrow and allow SendObjectHeaders to fail.

* Update arrow

* Update python/ray/experimental/state.py

Co-Authored-By: robertnishihara <[email protected]>

* Address comments.

<!--
Thank you for your contribution!

Please review https://github.com/ray-project/ray/blob/master/CONTRIBUTING.rst before opening a pull request.
-->

## What do these changes do?

support raylet, which is started by java runManager, to start python default_worker.py . 

So when doing local test of java call python task, it helps auto start python worker.

## Related issue number

<!-- Are there any issues opened that will be resolved by merging this change? -->

…bers of variables (ray-project#3139)

users can add custom path of ray.config by using -Dray.config=/path/to/ray.conf

* Remove legacy Ray code.

* Fix cmake and simplify monitor.

* Fix linting

* Updates

* Fix

* Implement some methods.

* Remove more plasma manager references.

* Fix

* Linting

* Fix

* Fix

* Make sure class IDs are strings.

* Some path fixes

* Fix

* Path fixes and update arrow

* Fixes.

* linting

* Fixes

* Java fixes

* Some java fixes

* TaskLanguage -> Language

* Minor

* Fix python test and remove unused method signature.

* Fix java tests

* Fix jenkins tests

* Remove commented out code.

…ect#3119)

We found that there are large amount of pub-sub keys with no content in it (This case is worse when wait-id is used in the key name.).
This logic of deleting empty pub-sub keys from GCS was in legacy ray but not in raylet.

…t#3033)

This includes most of the TF code used for the OSDI experiment. Perf sanity check on p3.16xl instances: Overall scaling looks ok, with the multi-node results within 5% of OSDI final numbers. This seems reasonable given that hugepages are not enabled here, and the param server shards are placed randomly.

$ RAY_USE_XRAY=1 ./test_sgd.py --gpu --batch-size=64 --num-workers=N \
  --devices-per-worker=M --strategy=<simple|ps> \
  --warmup --object-store-memory=10000000000

Images per second total
gpus total              | simple | ps
========================================
1                       | 218
2 (1 worker)            | 388
4 (1 worker)            | 759
4 (2 workers)           | 176    | 623
8 (1 worker)            | 985
8 (2 workers)           | 349    | 1031
16 (2 nodes, 2 workers) | 600    | 1661
16 (2 nodes, 4 workers) | 468    | 1712   <--- OSDI perf was 1817

ray-project#3130)

* Allow the node manager port and object manager port to be set through ray start.

* Linting

* Fix Java test

* Address comments.

… too small. (ray-project#3149)

* Use /tmp instead of /dev/shm for object store on Linux if /dev/shm is too small.

* Add logging statement and address comments.

* Fix

…t#3114)

* Remove num_workers argument.

* Fix

* Fix

…ariable sharing (ray-project#3151)

)

* Policy that flushes the lineage stash immediately

* Fix bug where remote tasks in uncommitted lineage weren't getting subscribed to, add reg test

* test

* Fix bug where waiting task was getting subscribed

* Cleanup

* Update src/ray/raylet/lineage_cache.cc

Co-Authored-By: stephanie-wang <[email protected]>

* Update src/ray/raylet/lineage_cache.cc

Co-Authored-By: stephanie-wang <[email protected]>

* cleanup

* cleanup

* Add another test for task with many parents

* fix, unsubscribe to new waiting tasks

* Unsubscribe as soon as the commit notification is handled

Modify `training_iteraion` to `timesteps_total` because only `timesteps_total` is inside the reporter.

* Update task_table and object_table API.

* Fix

* Add example for fractional GPU support

* Update tune_mnist_keras.py

* Update doc/source/tune-usage.rst

…y-project#3147)

* Add use_raylet option for backwards compatibility.

* Update message.

)

This reverts commit 55d161b.

* Rename get_task -> worker_idle in timeline.

* Fix test.

* Fix docs.

* Update doc/source/tempfile.rst

Co-Authored-By: suquark <[email protected]>

* Remove doc for raylet socket.

)

## What do these changes do?

Clean up the checkpointing to handle the new checkpoint dirs. Add a test for rollout.py

## Related issue number

ray-project#3206
ray-project#3204

* caching resource requirements

* small fixes

* avoid copying the resource map