Skip to content

Releases: veertuinc/anklet

0.12.1

08 Jan 14:52
Compare
Choose a tag to compare
  • [metrics]: Added healthcheck endpoint to aggregator

0.12.0

08 Jan 00:29
Compare
Choose a tag to compare
  • [github | handler]: Added skip_pull for users who don't have a registry.
  • [metrics]: Added last_update timestamp showing when the last time the DB keys were updated with new metrics. This can be used to see if there is a problem with Anklet services pushing metrics to the DB.
  • [metrics]: Added back JSON logs for Aggregator
  • [metrics]: On Anklet stop, the metrics are removed from the database so they don't pile up in the DB.

0.11.3

06 Jan 19:41
Compare
Choose a tag to compare
  • [github | handler] registration_timeout_seconds is no longer required and will default to 60s if not set in your config.

0.11.2

17 Dec 23:37
Compare
Choose a tag to compare
  • [github] Fixed minor bug where a failing install script would shut down all of anklet.

0.11.1

16 Dec 20:12
Compare
Choose a tag to compare
  • [metrics/github] Quick fix for canceled metrics not showing the proper timestamp.

0.11.0

14 Dec 01:49
Compare
Choose a tag to compare
  • #41 Handler metrics are now stored in the database which allows the aggregator to obtain them without knowing an IP/URL.
  • Various metrics improvements.
  • Removed json metrics. We can always add them back upon request.

0.10.0

13 Dec 16:48
Compare
Choose a tag to compare
  • [github] Added metrics for Cancellations
  • [github] Added timeout for registration failures

    At the moment of writing this, Github has a bug that assigns the same internal ID to runners that are registered around the same time. See actions/runner#3621 for more info. This means we need to count the time between when we register and start the runner in the VM and when the webhook/job goes into in_progress status, indicating the job in Github is now running inside of the VM properly. IF it doesn't go into in_progress within 1 minute, we consider it the github bug and will retry it on a brand new VM.

  • [github] Ability to copy out the _diag folder from the VM each run when LOG_LEVEL either DEBUG or DEV
  • log::split_by_plugin added so you can output each plugin to a separate file
     log:
       file_dir: /Users/nathanpierce/Library/Logs/
       split_by_plugin: true
    

0.9.4

02 Dec 23:57
Compare
Choose a tag to compare
  • Fixed aggregator

0.9.3

25 Nov 20:11
Compare
Choose a tag to compare
  • Better logging

0.9.2

21 Nov 23:28
Compare
Choose a tag to compare
  • [Github Receiver]: Optimized the redelivery logic to make as few api calls as possible.