feat(replay): Remove replay_relay_snuba_publish_disabled_sample_rate#5629
Conversation
| start_time: safe_timestamp(received_at), | ||
| payload, | ||
| }; | ||
| self.produce(KafkaTopic::ReplayEvents, KafkaMessage::ReplayEvent(message))?; |
There was a problem hiding this comment.
I think we can then also remove the ReplayEvents topic (and follow that up with an infra change to unconfigure the topic)
There was a problem hiding this comment.
I think for the time that we remove it here but not yet in the infra this will error here:
Line 50 in 88085ca
There was a problem hiding this comment.
Yeah that should happen and it's fine, this is the safest way to do it. Remove from code before removing from infra (e.g. to allow for rollbacks). In this case we could do something else considering the topic hasn't been used in a long time, but better to do it the 'right ' way.
We could lower that message from error to warn though. It's error because we didn't used to pipe warnings into Sentry.
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.
| start_time: safe_timestamp(received_at), | ||
| payload, | ||
| }; | ||
| self.produce(KafkaTopic::ReplayEvents, KafkaMessage::ReplayEvent(message))?; |
There was a problem hiding this comment.
Yeah that should happen and it's fine, this is the safest way to do it. Remove from code before removing from infra (e.g. to allow for rollbacks). In this case we could do something else considering the topic hasn't been used in a long time, but better to do it the 'right ' way.
We could lower that message from error to warn though. It's error because we didn't used to pipe warnings into Sentry.
jjbayer
left a comment
jjbayer
left a comment
There was a problem hiding this comment.
Might benefit from an "internal" changelog because it will change behavior in self-hosted.
tobias-wilfert
deleted the
tobias-wilfert/feat/hard-code-relay_snuba_publish_disabled
branch
…te` (#108110) As of this PR (getsentry/relay#5629) the option is no longer used. This PR unsets it in the option automator (getsentry/sentry-options-automator#6463)
…te` (#108110) As of this PR (getsentry/relay#5629) the option is no longer used. This PR unsets it in the option automator (getsentry/sentry-options-automator#6463)
…te` (#108110) As of this PR (getsentry/relay#5629) the option is no longer used. This PR unsets it in the option automator (getsentry/sentry-options-automator#6463)
3 participants
While working on the refactor of the
process_replaylogic we noticed thatrelay-snuba-publishing-disabledis perpetually set to 100%.This PR replaces the option with a hard coded true value and removes all the code that is dead due to this change. Specifically
produce_replay_eventwhich currently just early returns andReplayEventKafkaMessagewhich is only constructed in this function. Furthermore the tests that rely on the option are also modified or removed (if they now test behaviour that is no longer possible).Follow up:
replay_eventsin the processor since they are dropped silently in the store anyways.