You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on May 3, 2022. It is now read-only.
To be able to effective logging, rate limiting, circuit breaking and alerting it is good to differentiate between
A hyperbahn worker generates a Busy frame because its busy
A service Foo generates a Busy frame and we are forwarding an error frame.
If we forward an error frame then we do not need to logger.warn().
If we are generating Busy frames then we should actually logger.warn() and alert above a threshold.
This would allows us to improve the smarts of our RelayRequest.logError() method to give us the right level of insight in terms of info/warn/error and as a consequence give us the right level of insight in terms of should we alert on the cluster being unhealthy.
The text was updated successfully, but these errors were encountered:
To be able to effective logging, rate limiting, circuit breaking and alerting it is good to differentiate between
If we forward an error frame then we do not need to
logger.warn()
.If we are generating Busy frames then we should actually
logger.warn()
and alert above a threshold.This would allows us to improve the smarts of our RelayRequest.logError() method to give us the right level of insight in terms of info/warn/error and as a consequence give us the right level of insight in terms of should we alert on the cluster being unhealthy.
The text was updated successfully, but these errors were encountered: