We request IBM to develop a dedicated Instana sensor for IBM Cloud Pak for AIOps (CP4AIOps) running on Red Hat OpenShift.
At present, we use Instana K8Sensor together with the Instana agent to monitor OpenShift and workloads running on it, including CP4AIOps. This provides good cluster- and Kubernetes-level visibility and cp4aiops tracing performace some overall metrics related to technologies it usages such as cassandra, Kafka, postgresssql, call rate etc., but it does not deliver the product-specific operational detail needed for effective monitoring of CP4AIOps itself.
CP4AIOps already includes a self-monitoring capability through Prometheus and Grafana, and this exposes many critical metrics for the platform. Those metrics are useful and give much deeper visibility into the health and behavior of CP4AIOps components than what is currently available in Instana. However, having to rely on a separate Prometheus/Grafana view creates operational fragmentation. If we have to rely on Prometheus/Grafa then Instana becomes irrelevant.
We would like IBM to provide a native Instana sensor that can discover CP4AIOps components on OpenShift and surface the most important CP4AIOps metrics directly in Instana.
Current challenge
Today, our teams can use Instana to monitor:
OpenShift cluster health
Nodes, namespaces, pods, and containers
General workload performance
However, for detailed CP4AIOps monitoring, we still need to rely on the CP4AIOps self-monitoring stack. This leads to several challenges:
Monitoring is split across multiple tools, Teams must switch between Instana and Grafana, Root cause analysis takes longer, Correlating platform symptoms with CP4AIOps internal issues is more difficult and Instana cannot act as the single observability interface for the full IBM stack
Requested capability:
The requested Instana sensor should:
1. Automatically detect CP4AIOps services and components deployed on OpenShift
2. Ingest key CP4AIOps metrics already exposed through self-monitoring
3. Provide prebuilt dashboards for CP4AIOps health and performance
4. Show service/component dependencies in Instana topology
5. Enable alerting on critical CP4AIOps conditions
6. Correlate CP4AIOps issues with OpenShift and Kubernetes resource conditions
Metrics and visibility desired
Examples of useful coverage include:
1. Event and alert ingestion rates
2. Processing latency and backlog indicators
3. Connector and integration health
4. AI/ML and analytics service health
5. API/service response health
6. Built-in health indicators and recommended thresholds for major CP4AIOps components
Business and operational value
This enhancement would provide important value to customers who use Instana as their strategic observability platform:
1. Creates a single pane of glass for OpenShift and CP4AIOps
2. Reduces operational complexity and tool switching
3. Improves troubleshooting and mean time to resolution
4. Increases value and adoption of Instana in IBM software environments
5. Strengthens integration across the IBM observability and automation portfolio
Why this matters
Since CP4AIOps already exposes rich metrics through Prometheus and Grafana, it is clear that the necessary observability data exists. A dedicated Instana sensor would allow customers to consume that operational insight directly within Instana, which would significantly improve usability, visibility, and operational efficiency.
Requested outcome
Please consider adding native Instana support for CP4AIOps on OpenShift through a dedicated sensor with dashboards, topology, health signatures, and alerts for critical CP4AIOps metrics.
