Andy Lee

[Jobs] Refactor: Extract task failure state update helper

1

Extract repeated task failure state update logic into `_update_failed_task_state()` helper to reduce code duplication. Good catch! Could submit this change to master first. _Originally posted by @cblmemo in https://github.com/skypilot-org/skypilot/pull/4128#discussion_r1817404695_ Tested...

Refactor `server.controller`: Replace Process Pool with Thread Pool for Better Performance

2

### Background Initially, a process pool was used in `server.controller` to handle replica launching and termination due to limitations with redirecting `sys.stdout` and `sys.stderr` in multi-threaded environments. The global nature...

Currently, replicas can get stuck during shutdown, requiring manual intervention into code logic. This is very unfriendly to users. We propose implement a timeout mechanism for replica termination. Replicas that...

Refactor optimizer: Ensure proper removal of sink/dummy nodes in DAG

3

Our optimizer currently removes sink and dummy nodes without checking if they're still depended on by other nodes. This violates DAG rules and forces us to use a workaround in...

Implement Automated Weekly Smoke Tests

2

# Implement Automated Weekly Smoke Tests ## Problem Currently, smoke tests for SkyPilot (implemented in `test_smoke.py`) are being run manually. This process could be improved by automating these tests on...

Implement automated weekly smoke tests with GitHub Actions

4

Fixes #4112 Implement an automated weekly smoke test run using GitHub Actions leveraging the existing `tests/test_smoke.py` script. * Add a new GitHub Actions workflow file `.github/workflows/weekly-smoke-tests.yml`. * Schedule the workflow...