GuidesAPIChangelogSupportStatusCodecov HomeTry Codecov for Free
Back to All

Release Notes for Codecov v4.5.10

about 3 years ago by Eli Hooten

🚧

This update runs database migrations

This update runs a database migration to add a new column to the owners table. While no issues are expected to occur with this migration, we recommend ensuring you have recent database backups in place before upgrading to v4.5.10.

v4.5.10

New

  • Better support for flag management has been added to the yaml. Users can now setup flag_management rules that allow for the automatic configuration of new flags. This is useful for those repositories that have many flags, or add new flags often and do not wish to configure each one. More about this feature can be seen in our Flags Documentation.
  • (On-premises) Added the ability to increase queue timeout intervals for Notify tasks. This is useful in the event that Notify tasks are encountering timeout issues. See more in the Notify Timeout Limit Changes section below.

Fixes

  • Fixed a bug that sometimes caused GitHub app integration reinstalls to not properly update the database.
  • Fixed issues related to improper pagination of GitLab endpoints when using the GitLab API.

Minor

  • Adds internal support for better associating GitLab subgroups with their root owning organization, even in the case of deeply nested subgroups. This functionality will be leveraged for future user-facing product improvements, but is now only used internally by the application.
  • Removed all support for the v3 version of the GitLab API.
  • (Cloud only) Improved wording of in-application alerts that better describe auto-activation and when it occurs.

Notify Timeout Limit Changes

📘

On-Premises Only

This timeout is only configurable from the codecov install yaml file included with on-premises installations.

In some cases, particularly for those repositories that use a large number of flags or have many (i.e., <50) uploads per commit, notifications and statuses can timeout. Prior to v4.5.10, a 45 second hard timeout existed on status update tasks. v4.5.10 now makes this timeout configurable.

In v4.5.10, you will receive an error in your worker log output if you hit this timeout issue. The log line will look as follows:

{"message": "Soft time limit (45s) exceeded for app.tasks.notify.Notify ...}

If you encounter this error after upgrading to v4.5.10, you can increase the timeout by updating your codecov.yml as follows:

setup:
  tasks:
    notify:
      timeout: 45 #default in seconds

and supplying a higher value for timeout.

It is recommended to incrementally increase this value to prevent notify tasks from consuming an inordinate amount of time in the queue. For example, if the 45 second default is not sufficient, increasing to a 60s timeout may be a good first step.

Docker Image Information

The following Docker digests comprise the v4.5.10 release:

  • codecov/enterprise-web:v4.5.10 - sha256-11cbac6acc7e8caa1add89c73d434cd6a2eedf754e4af303ae3cbfb815e0433c
  • codecov/enterprise-worker:v4.5.10 -sha256: 08a8ba4a8a245fdd48070f50b0447f1dcbe0cddf1b60d73253853ef0a1cad145

Note: due to a user impacting bug in codecov/enterprise-web:v4.5.10, a point release was issued pushed over the existing v4.5.10 tag. The new digest is indicated above, the previous digests are as follows:

  • codecov/enterprise-web:v4.5.10 - sha256:8812f70abf6e5bcf84d637eadf1608b04780199895891913d28dd2563cf88a4c