We have been using the github deployment extension to manage our Auth0 tenants for a year or two now. Sometimes deployments fail for various reasons and we were hoping to create some alerts for this occurrence downstream in datadog. We have our Auth0 logs streaming to datadog successfully but are not able to determine when a deployment fails from them. It seems that events are logged for ‘successful API operations’ with the log event sapi, but failed API operations are not logged. We tested this by merging/deploying malformed JSON files, creating grants that reference non-existence clients, etc.
Does anyone have recommendations for determining when a deployment fails from Auth0 logs?
Are there any plans to add log events for failed API operations?
When it comes to your second point I believe it can be already there. Pretty much what you can accomplish via the dashboard can also be accomplished via the Management API and dashboard events are in the logs so I presume API ones should also be there.
@karen1 The ‘Real-time Webtask Logs’ are certainly useful once I know I have a problem but I am looking to automate something that detects when problems happen in the first place.
@konrad.sopala I believe you are correct that anything I can do with the dashboard I can automate, but my issue is that not all API errors are logged as events, so I can’t figure this out from the dashboard either. The only place I can see that a deployment failed is in the github-deployment extension.
A simple example case is deploying a ‘client’ with malformed JSON. The management API will return an HTTP 400 (or another error status) to the caller but doesn’t seem to log an error event. This is the type of failure I am looking to catch automatically.