keptn
Dynatrace provider: improve troubleshooting capabilities
When working with the DQL provider, I noticed that it is quite hard to figure out why a query might be failing. This is because in case of an error when communicating with the Dynatrace API, only a static error message saying that there was an error is being returned and visible in the logs or the status of an Analysis - This is where the error is being returned: https://github.com/keptn/lifecycle-toolkit/blob/2cd4491fa49876534b0f5344c1e3dd4fcab7e540/metrics-operator/controllers/common/providers/dynatrace/client/client.go#L120C18-L120C18
To make debugging these kinds of issues easier, we should give more information (at least the HTTP status code that has been returned).
DoD
- [ ] Returned errors by the DQL handler include at least the HTTP status returned by the API + more information if possible
Acceptance Criteria
- [ ] Unit tests
- [ ] chainsaw test using mockserver
Additional Remarks
Implementation wise this will likely be very simple, so this might be a good first issue. Proper e2e testing will be more complex however, as this requires access to a DT tenant with. So this should be reviewed by an approver with an access to a DT tenant and some familiarity with DQL to verify this is working correctly.
Hi @Vickysomtee we will for sure refine this issue next Wednesday community meeting, we will assign it to you as soon as it is refined! If you like you can join in if you would like to have more info on the ticket. We post a reminder on our CNCF slack channel, or you can find the zoom link here
Hi @Vickysomtee we will for sure refine this issue next Wednesday community meeting, we will assign it to you as soon as it is refined! If you like you can join in if you would like to have more info on the ticket. We post a reminder on our CNCF slack channel, and you can find the zoom link here
@RealAnna Thank You! I will be there no mater what