In Dynatrace Kubernetes Application Monitoring mode, native sidecars are not appearing under the containers list in the latest Explorer UI.
Background
In our setup, Istio runs envoy / istio-proxy as native sidecars (initContainers with restartPolicy: Always), which is the newer model promoted by Istio instead of the legacy injected sidecar approach.
However:
- In the latest Dynatrace Explorer UI, these istio-proxy containers are not visible.
- This gives the impression that native sidecars are ignored — at least from an infrastructure/container visibility perspective (not necessarily deep code monitoring).
Observations
- The behavior seems consistent across:
- Application Monitoring mode
- CloudNativeFullStack mode ( latest explorer window )
- In Explorer Classic, running init containers were previously listed when clicking on containers in a pod, but this is no longer the case in the latest UI.
- Legacy sidecars (regular containers) appear normally.
Questions
- Is this an expected limitation of Dynatrace with Kubernetes native sidecars?
- Do understand , there could be limitations on DT side to inject on init conatiners, but how kuberenetes platform monitoring is missing it ?
- Is there any roadmap or workaround to get visibility at least on the infra/container level?
Currently thinking about relying on external metrics sources (Prometheus / node-exporter / Istio metrics) ingested into Dynatrace to track native sidecar resource usage instead.
Would appreciate if anyone has experience dealing with native sidecars or initcontainers+ Dynatrace visibility.