Skip to content

v2.2.0

Choose a tag to compare

View all tags
@luocfprime luocfprime released this 09 Sep 15:09
· 18 commits to main since this release
v2.2.0
ae930c9

Important

Upgrade the Server first to enable the new observation and query features, then upgrade Clients and Workers. Existing v2 Clients remain compatible; older Workers do not report observations. New Workers can still execute Tasks against older v2 Servers.

The Server automatically migrates the database. Back it up before upgrading if rollback is needed: older Servers cannot open the migrated database, and no downgrade command is provided.

Changes

  • Add Worker visibility: list and count idle/busy Workers through HTTP, Python, and the CLI, without changing Task scheduling or ownership.
  • Add Queue inspection tools: grouped Task counts by status and routes, plus case-insensitive fuzzy Task name search.
  • Change Worker failure protection: stop after five consecutive execution failures by default, configurable with max_consecutive_failures. Successful completion resets the count; transport retries do not count as workload failures.
  • Fix Server reliability: database locking on macOS, local daemon recovery, atomic schema upgrades, and consistent Task reads and pagination.
  • Fix Worker recovery: cancellation and result-reporting races, process cleanup, and local log failures that could interrupt execution or reporting.

Full Changelog: v2.1.0...v2.2.0

Assets 2
Loading