Any way to turn SLO breach into problem tiles?
Can't seem to find this in custom alert.
Other than possibly create a new custom metrics for it, any other possibility?
Solved! Go to Solution.
Coming soon, with calculated metrics capabilities enabling alerting on calculation results also.
Hi @wolfgang_heider , I am checking and it's not possible yet to filter Problem for SLO or release. We are integrating SLO's and releases concepts in the pipeline and fit's really important to filter Problems by releases (stage, version and component) and SLOs because thats the application of AI in the Shift-left pipeline and it will help in the decision making if release or not release in production.
Do you know when it will be available?
Thanks in advance.
We very soon release cross-linking from SLOs to related problems (based on the related entities).
To link the other way round (from Problems to affected SLOs) and filter Problems accordingly is planned to follow up with throughout the next months.
Independent of that cross-linking, the static threshold alerting on SLOs and also error budget burn rate is being worked on already as part of the metric expressions (functions and calculations with metrics - as stated in the Planned features for SLOs - Page 2 - Dynatrace Community).
Thanks @wolfgang_heider , the most important por us right now is how can we check the problem when a SLO is violated (included infra problems like CPU, etc) and the most important thing, how can u alert problems in Problem Notification section , even self-healing calling Ansible playbooks for specific issues (i.e: response time isse or error - undeploy artefact, Infra proble - resize the infra nodes).
As I see, it is coming soon in the following months (Q2 , Q3 2021).
Another important issue is to filter SLO by release which in not enabled (I don't know how long a QA test will take but so Timeframe concept in the shift-lest is not as important as filtering SLO bu releases). I don't know if this feature is scheduled or not.
Thanks in advance.
Another question, do you know if it is planned to extend SLO's to infraestructure SLI's?
Hi @Alberto-Atos , you can already do that with infra-related metrics.
More metrics are coming, e.g., host availability as direct % metric. I'm aware, that many infra-related metrics are not available as % metrics, but very soon (CQ3) you can do any math with different metrics to normalize any (also infra-related) metric for usage in SLOs.