Flakey failure of the cluster script in tests
shachlanAmazon opened this issue · 1 comments
shachlanAmazon commented
https://github.com/aws/glide-for-redis/actions/runs/7961591871/job/21733197017?pr=971
All the node cluster tests fail, seemingly due to the cluster manager failure.
console.log
python3 ../utils/cluster_manager.py start --cluster-mode -r 0 -n 3
at tests/TestUtilities.ts:197:21
console.error
INFO:root:## Executing cluster_manager.py with the following args:
Namespace(host='127.0.0.1', tls=False, auth=None, log='info', logfile=None, action='start', cluster_mode=True, folder_path='/home/runner/work/glide-for-redis/glide-for-redis/utils/clusters', ports=None, shard_count=3, replica_count=0, prefix='redis-cluster', load_module=None)
INFO:root:2024-02-19 15:09:46.923248+00:00 Starting script for cluster /home/runner/work/glide-for-redis/glide-for-redis/utils/clusters/redis-cluster-2024-02-19T15-09-46Z-rcl2Ni
Traceback (most recent call last):
File "/home/runner/work/glide-for-redis/glide-for-redis/node/../utils/cluster_manager.py", line 1036, in <module>
main()
File "/home/runner/work/glide-for-redis/glide-for-redis/node/../utils/cluster_manager.py", line 973, in main
servers = create_servers(
File "/home/runner/work/glide-for-redis/glide-for-redis/node/../utils/cluster_manager.py", line 372, in create_servers
if is_address_already_in_use(server, f"{node_folder}/redis.log"):
File "/home/runner/work/glide-for-redis/glide-for-redis/node/../utils/cluster_manager.py", line 627, in is_address_already_in_use
with open(log_file, "r") as f:
FileNotFoundError: [Errno 2] No such file or directory: '/home/runner/work/glide-for-redis/glide-for-redis/utils/clusters/redis-cluster-2024-02-19T15-09-46Z-rcl2Ni/25032/redis.log'
198 | exec(command, (error, stdout, stderr) => {
199 | if (error) {
> 200 | console.error(stderr);
| ^
201 | reject(error);
202 | } else {
203 | const { clusterFolder, ports } = this.parseOutput(stdout);
asafpamzn commented
Closing - waiting for repro