OSS release tests currently run with hardcoded Python 3.7 base. In the future we will want to run tests on different python versions.
This PR adds support for a new `python` field in the test configuration. The python field will determine both the base image used in the Buildkite runner docker container (for Ray client compatibility) and the base image for the Anyscale cluster environments.
Note that in Buildkite, we will still only wait for the python 3.7 base image before kicking off tests. That is acceptable, as we can assume that most wheels finish in a similar time, so even if we wait for the 3.7 image and kick off a 3.8 test, that runner will wait maybe for 5-10 more minutes.
* Revert "Revert "[tune] Also interrupt training when SIGUSR1 received" (#24085)"
This reverts commit 00595653ed.
Failure in windows has been addressed by conditionally registering the signal handler if available.
Ray Tune currently gracefully stops training on SIGINT. However, the Ray core worker prevents SIGINT (and SIGTERM) to be processed by child tasks, which means that Ray Tune runs that are started in remote tasks (e.g. via Ray client) cannot be gracefully interrupted.
In k8s-based cloud tests that used the Ray client to kick off a Ray Tune run, this lead to test flakiness, as final experiment state could not be gracefully persisted to cloud storage.
This PR adds support for SIGUSR1 in addition to SIGINT to interrupt training gracefully.
We use tarfile to pack/unpack directories in several locations. Instead of using temporary files, we can just use io.BytesIO to avoid unnecessary disk writes.
Note that this functionality is present in 3 different modules - in Ray (AIR), in the release test package, and in a specific release test. The implementations should live in the three modules independently, so we don't add a common utility for this (e.g. the ray_release package should be independent of the Ray package).
This PR addresses recent failures in the tune cloud tests.
In particular, this PR changes the following:
The trial runner will now wait for potential previous syncs to finish before syncing once more if force=True is supplied. This is to make sure that the final experiment checkpoints exist in the most recent version on remote storage. This likely fixes some flakiness in the tests.
We switched to new cloud buckets that don't interfere with other tests (and are less likely to be garbage collected)
We're now using dated subdirectories in the cloud buckets so that we don't interfere if two tests are run in parallel. Objects are cleaned up afterwards. The buckets are configured to remove objects after 30 days.
Lastly, we fix an issue in the cloud tests where the RELEASE_TEST_OUTPUT file was unavailable when run in Ray client mode (as e.g. in kubernetes).
Local release test runs succeeded.
https://buildkite.com/ray-project/release-tests-branch/builds/189https://buildkite.com/ray-project/release-tests-branch/builds/191
This fixes the previous problems from team column revert.
This has 2 additional changes;
alert handler receives the team argument, which was the root cause of breakage; https://github.com/ray-project/ray/pull/21289
Previously, tests without a team column were raising an exception, but I made the condition weaker (warning logs). I will eventually change it to raise an exception, but for smoother transition, we will log warning instead for a short time
Please review **e2e.py and test_suite belonging to your team**!
This is the first part of https://docs.google.com/document/d/16IrwerYi2oJugnRf5hvzukgpJ6FAVEpB6stH_CiNMjY/edit#
This PR adds a team name to each test suite.
If the name is not specified, it will be reported as unspecified.
If you are running a local test, and if the new test suite doesn't have a team name specified, it will raise an exception (in this way, we can avoid missing team names in the future).
Note that we will aggregate all of test config into a single file, nightly_test.yaml.
## Why are these changes needed?
The .boto files are already added to the base image and ACL'ed to root, adding them again during app config build causes permission issues.
## Related issue number
* use nightly
* switch ml cpu to ray cpu
* fix
* add pytest
* add more pytest
* add constraint
* add tensorflow
* fix merge conflict
* add tblib
* fix
* add back uninstall
* prepare for head node
* move command runner interface outside _private
* remove space
* Eric
* flake
* min_workers in multi node type
* fixing edge cases
* eric not idle
* fix target_workers to consider min_workers of node types
* idle timeout
* minor
* minor fix
* test
* lint
* eric v2
* eric 3
* min_workers constraint before bin packing
* Update resource_demand_scheduler.py
* Revert "Update resource_demand_scheduler.py"
This reverts commit 818a63a2c86d8437b3ef21c5035d701c1d1127b5.
* reducing diff
* make get_nodes_to_launch return a dict
* merge
* weird merge fix
* auto fill instance types for AWS
* Alex/Eric
* Update doc/source/cluster/autoscaling.rst
* merge autofill and input from user
* logger.exception
* make the yaml use the default autofill
* docs Eric
* remove test_autoscaler_yaml from windows tests
* lets try changing the test a bit
* return test
* lets see
* edward
* Limit max launch concurrency
* commenting frac TODO
* move to resource demand scheduler
* use STATUS UP TO DATE
* Eric
* make logger of gc freed refs debug instead of info
* add cluster name to docker mount prefix directory
* grrR
* fix tests
* moving docker directory to sdk
* move the import to prevent circular dependency
* smallf fix
* ian
* fix max launch concurrency bug to assume failing nodes as pending and consider only load_metric's connected nodes as running
* small fix
* deflake test_joblib
* lint
* placement groups bypass
* remove space
* Eric
* first ocmmit
* lint
* exmaple
* documentation
* hmm
* file path fix
* fix test
* some format issue in docs
* modified docs
* joblib strikes again on windows
* add ability to not start autoscaler/monitor
* a
* remove worker_default
* Remove default pod type from operator
* Remove worker_default_node_type from rewrite_legacy_yaml_to_availble_node_types
* deprecate useless fields
Co-authored-by: Ameer Haj Ali <ameerhajali@ameers-mbp.lan>
Co-authored-by: Alex Wu <alex@anyscale.io>
Co-authored-by: Alex Wu <itswu.alex@gmail.com>
Co-authored-by: Eric Liang <ekhliang@gmail.com>
Co-authored-by: Ameer Haj Ali <ameerhajali@Ameers-MacBook-Pro.local>
Co-authored-by: root <root@ip-172-31-56-188.us-west-2.compute.internal>
Co-authored-by: Dmitri Gekhtman <dmitri.m.gekhtman@gmail.com>
* Working prototype
* Pass buffer length, fix tests
* Don't buffer per default
* Dispatch and process save in one go, added tests
* Fix tests
* Pass adaptive seconds to train_buffered, stop result processing after STOP decision
* Fix tests, add release test
* Update tests
* Added detailed logs for slow operations
* Update python/ray/tune/trial_runner.py
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
* Apply suggestions from code review
* Revert tests and go back to old tuning loop
* nit
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
