search

fluxcd / flux

Successor: https://github.com/fluxcd/flux2
https://fluxcd.io
Apache License 2.0
6.9k stars 1.08k forks source link

fix(repocachemanager): rate limit condition #3542

Closed fdpeiter closed 3 years ago

fdpeiter commented 3 years ago

Recently we've started seeing several messages related to the Dockerhub pull limit in our clusters:

{"auth":"{map[]}","caller":"repocachemanager.go:226","canonical_name":"index.docker.io/connecteverything/nats-operator","component":"warmer","err":{"errors":[{"code":"TOOMANYREQUESTS","message":"You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit"}]},"ref":"docker.io/connecteverything/nats-operator:0.8.0","ts":"2021-08-17T15:50:10.734988918Z"}

This is unexpected, as there is a specific switch case for handling rate-limiting here. I checked the content of err.Error() and got this: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit.

I've added this error message as a new condition for the switch case. I've tested it already and got the expected behaviour:

{"auth":"{map[]}","caller":"repocachemanager.go:216","canonical_name":"index.docker.io/tfgco/will-iam","component":"warmer","ts":"2021-08-20T17:20:46.220746763Z","warn":"aborting image tag fetching due to rate limiting, will try again later"}