PrefectHQ/prefect - CodSpeed

prefect

Blog Docs Changelog

Performance History

Latest Results

fix: Missing task runs in flow runs. (closes #21057)

trinathone:fix/issue-21057-missing-task-runs-in-flow-runs-

12 hours ago

Add tests for untested functions in utilities.math Covers exponential_cdf, lower_clamp_multiple (including the large-k short-circuit to 0), bounded_poisson_interval (bounds and average), and the non-positive clamping_factor validation in clamped_poisson_interval.

eeshsaxena:test-math-utilities

1 day ago

docs: document relative-path invariant for UI version redirects in server AGENTS.md (#22408) Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

main

1 day ago

docs: Add release notes for 3.7.7 (#22434) Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Alex Streed <alex.s@prefect.io> Co-authored-by: alex.s <ajstreed1@gmail.com> Co-authored-by: Alex Streed <desertaxle@users.noreply.github.com>

main

1 day ago

Fix double backticks in docstring Co-authored-by: Alexander Streed <alex.s@prefect.io>

devin/1783082794-supervise-docket-worker

2 days ago

Supervise Docket worker to restart after Redis failover Docket's Worker._run() reconnect loop only catches redis.exceptions.ConnectionError. During Redis failovers, redis.exceptions.TimeoutError (not a ConnectionError subclass) can propagate out of run_forever(), silently killing the worker loop while the independent heartbeat task keeps the pod healthy. Wrap worker.run_forever() in _supervised_worker() that catches any unexpected exception and restarts after the configured reconnection_delay, so background scheduling resumes once Redis connectivity recovers. closes #22436 Co-authored-by: Alexander Streed <alex.s@prefect.io>

devin/1783082794-supervise-docket-worker

2 days ago

Fix parameter name mismatches in docstrings - server/models/work_queues.py: read_work_queue_by_name docstring said "by id" with a `work_queue_id` arg; it takes `name` - server/models/workers.py: `worker` -> `work_pool` in update_work_pool - server/schemas/core.py: `late_runs` -> `late_runs_count` in evaluate_health_status - server/models/artifacts.py: remove stale `work_pool_filter` entry from read_artifacts and read_latest_artifacts (not a parameter)

eeshsaxena:fix-docstring-param-names

2 days ago

Fix review findings in Redis Sentinel support - Tolerate pre-connection_url pickles in RedisLockManager.__setstate__ - Honor tls_insecure/tls_ca_file on single-node messaging URLs and retain them verbatim in RedisDatabase.from_connection_string - Detect Sentinel schemes without urlsplit (IPv6 member lists) and case-insensitively; same for cluster URLs - Forward caller socket timeouts to the Sentinel daemon connections - Lift the 0-15 database index cap (only negative indexes are invalid) - Close Sentinel daemon clients deterministically in read_path/write_path via new close_redis_client/aclose_redis_client helpers Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

fatih-acar:feat/redis-sentinel-on-22211

3 days ago

Latest Branches

0%

fix: Missing task runs in flow runs.#22440

12 hours ago

4c3f0de

trinathone:fix/issue-21057-missing-task-runs-in-flow-runs-

0%

Add tests for untested functions in utilities.math#22438

1 day ago

fef5c8c

eeshsaxena:test-math-utilities

0%

Supervise Docket worker to restart after Redis failover#22437

2 days ago

4c5efe0

devin/1783082794-supervise-docket-worker

© 2026 CodSpeed Technology

Home Terms Privacy Docs