At this time, there is AuditLog in Digital AI Release that allow to know when someone did something ( like skipping, removing step, etc .. ).
That's usefull, but it should be also interesting to integrate with those kind of "event" in an automated way.
I can see two mode that can be interesting :
Audit Event only
Calling a webhook (or sending the audit log into Kafka topic) to be able to filter and do something in case of action done (
Alert if skip are performed, or task deleted)
Precheck webhook :
Calling a webhook that will check if the action is allowed or not.
Ex :
Removing a task shouod be acceptable under conditions, but not in other case.
by: Gildas C. | 10 months ago | Reports
Comments
Thank you for your request. Could you please provide specific use cases that you have in mind for implementing these features with your teams?
We are interested in understanding the various use cases that other customers might have as well.
Sure
As we are using Release for delivery in our different environment, we have some environment that are less critical than others ( Ex : TST, UAT vs PROD).
Each environment is a phase in a Release.
We have some profile (Solution Architect) that have to be able to modify the release if needed to skip/add/remove step if needed.
It's not a problem in lower environment, but this should be more controlled in higher environment.
Today, it's possible to known who have done what, but only if we check explicitly to a release if someone see that there is modification.
But today the right are "on off" to edit/skip tasks, it's not scope per phase.
And we don't want to remove the Solution architect the right to do it in TST, UAT ..
So alternative are :
- Be able to sent an event/call a webhook on modification to filter when it's done on critical phase and allow to warn ( audit mode )
This should be potientially implemented with https://docs.digital.ai/bundle/devops-release-version-v.23.1/page/release/how-to/listen-events.html
But should be great if integrated in the product itself
- Be able to call a webhook on each modification to allow/disallow those modifications. (Ex : similar to AdmissionWebhook for Kubernetes ). This will allow to extends the "security" models.
This should probably need more development to integrate some hook in the process.