mirror of
https://github.com/vale981/ray
synced 2025-03-09 12:56:46 -04:00
658c14282c
* 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.
62 lines
1.8 KiB
Python
62 lines
1.8 KiB
Python
from __future__ import absolute_import
|
|
from __future__ import division
|
|
from __future__ import print_function
|
|
|
|
import logging
|
|
import pytest
|
|
|
|
import ray
|
|
import ray.services as services
|
|
from ray.test.cluster_utils import Cluster
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
@pytest.fixture
|
|
def start_connected_cluster():
|
|
# Start the Ray processes.
|
|
g = Cluster(initialize_head=True, connect=True)
|
|
yield g
|
|
# The code after the yield will run as teardown code.
|
|
ray.shutdown()
|
|
g.shutdown()
|
|
|
|
|
|
def test_cluster():
|
|
"""Basic test for adding and removing nodes in cluster."""
|
|
g = Cluster(initialize_head=False)
|
|
node = g.add_node()
|
|
node2 = g.add_node()
|
|
assert node.all_processes_alive()
|
|
assert node2.all_processes_alive()
|
|
g.remove_node(node2)
|
|
g.remove_node(node)
|
|
assert not any(node.any_processes_alive() for node in g.list_all_nodes())
|
|
|
|
|
|
def test_wait_for_nodes(start_connected_cluster):
|
|
"""Unit test for `Cluster.wait_for_nodes`.
|
|
|
|
Adds 4 workers, waits, then removes 4 workers, waits,
|
|
then adds 1 worker, waits, and removes 1 worker, waits.
|
|
"""
|
|
cluster = start_connected_cluster
|
|
workers = [cluster.add_node() for i in range(4)]
|
|
cluster.wait_for_nodes()
|
|
[cluster.remove_node(w) for w in workers]
|
|
cluster.wait_for_nodes()
|
|
worker2 = cluster.add_node()
|
|
cluster.wait_for_nodes()
|
|
cluster.remove_node(worker2)
|
|
cluster.wait_for_nodes()
|
|
|
|
|
|
def test_worker_plasma_store_failure(start_connected_cluster):
|
|
cluster = start_connected_cluster
|
|
worker = cluster.add_node()
|
|
cluster.wait_for_nodes()
|
|
# Log monitor doesn't die for some reason
|
|
worker.kill_log_monitor()
|
|
worker.kill_plasma_store()
|
|
worker.process_dict[services.PROCESS_TYPE_RAYLET][0].wait()
|
|
assert not worker.any_processes_alive(), worker.live_processes()
|