Monitoring component for Cloud-native benchmarking tool bexhoma
7.8K
The image is based on prom/prometheus.
This folder contains the Dockerfile for a Prometheus monitoring container used by Bexhoma.
At startup the container writes $BEXHOMA_WORKERS to /etc/prometheus/prometheus-bexhoma.yml
and launches Prometheus with that configuration.
Prometheus listens on port 9090 (web UI and HTTP API).
The time-series data directory is /prometheus (mounted as a volume).
A default prometheus.yml is bundled in the image for reference but is not used at
runtime - the live configuration always comes from BEXHOMA_WORKERS.
BEXHOMA_WORKERS: Full YAML content of the Prometheus scrape configuration.
Written to /etc/prometheus/prometheus-bexhoma.yml at container startup.
When empty the container starts with an empty configuration and Prometheus will
report a parse error; always supply a valid YAML string.The bundled prometheus.yml defines two scrape jobs as a reference template:
| Job | Target | Interval |
|---|---|---|
monitor-node | localhost:9300 | 3 s |
monitor-gpu | localhost:9400 | 3 s |
These targets are not active unless BEXHOMA_WORKERS contains equivalent
scrape-config blocks.
Content type
Image
Digest
sha256:fcbb73bd6…
Size
137 MB
Last updated
7 days ago
docker pull bexhoma/monitoring