IBMC opensearch

Centralized dashboard-based solution to analysis the teutholgoy runs. See https://github.com/vamahaja/teuthology-metrics

The service running in container mode. Scheduler is the main service running the cron based solution to fetch the result from paddle and send the data to Opensearch.

1. Podman
   1. A daemon-less, rootless container engine that lets you build and run containers without requiring a central service. It provides strong compatibility with Docker image formats while improving security by not having a long-running daemon process.
2. Podman Compose
   1. A tool (or wrapper) that allows multi-container application definitions (via YAML) to be spun up using Podman. It enables starting the full stack of containers/services (as defined in container-compose.yaml) in one go.
3. Opensearch
   1. A distributed, open-source search and analytics engine (forked from Elasticsearch) that allows you to ingest, store, search and analyse large volumes of data (such as test run metrics) in near real-time.
4. OpenSearch Dashboard
   1. A browser-based visualisation and analytics interface for OpenSearch data — allowing you to explore, query, visualise and build dashboards over the ingested metrics.
5. Scheduler
   1. A component that periodically or on-trigger gathers test run results and schedules ingestion into OpenSearch. It orchestrates when metrics are pulled, processed and forwarded to the Opensearch.

Deployment Steps - https://github.com/vamahaja/teuthology-metrics/tree/main?tab=readme-ov-file#deployment-steps

1. Failure Report
2. Test Runs Result Trend
3. Test Runs Analysis by Suite and Branch

TODO