Troubleshooting Grafana Dashboards
This section describes common errors that may occur when using Hardware Observability and Sustainability dashboards in Grafana and the basic troubleshooting steps to follow. If you require further assistance, subscribe to one of our support plans and get access to Sentry Desk, our superior customer support.
No sustainability data available
You usually get a “No data” message for sustainability metrics when:
- the query takes too long to complete. This could be due to a network issue, a lack of resources, or the Prometheus server running slow.
- Hardware Sentry OpenTelemetry Collector has been running for less than 24 hours and does not have enough data to compute the annual type of metrics (typically,
Annual Energy Usage,
Annual CO₂ Emissions)
hw.site.carbon_density_gramsoptions are not properly set in the config/hws-config.yaml file. Refer to Configure the sustainability settings for more details.
Energy usage and carbon emissions are oddly low
If the values of the Annual Energy Usage, Annual Cost and Annual CO₂ Emissions panels seem low compared to the number of sites monitored, open each monitored site and verify that values are returned for:
If no values are displayed, open the config/hws-config.yaml file dedicated to each site and configure the following metrics:
Refer to Configure the sustainability settings for more details.
No hardware metrics available for hosts
If you notice that no hardware metrics are displayed for hosts:
Connect to your Prometheus server and search for the missing metric. If the metric corresponding to the monitored host is:
- found, Hardware Sentry OpenTelemetry Collector collects data and pushes it to Prometheus. The issue is on the Grafana level. Please proceed to step 2.
- not found:
- in the
pipelines:metrics:exporterssection to enable the
pipelines: metrics: receivers: [otlp, prometheus/internal] processors: [memory_limiter, batch, resourcedetection, metricstransform] exporters: [prometheusremotewrite/your-server, prometheus]
- restart Hardware Sentry OpenTelemetry Collector
- connect to the server running the Hardware Sentry Agent and open the URL
http://<localhost>:24375/metricsto verify that Hardware Sentry OpenTelemetry Collector is collecting data. If data:
- is collected, the Prometheus exporter is not properly set. Refer to Integration with Prometheus Server for more details.
- is not collected:
- Open the config/hws-config.yaml file and verify that the monitoring configuration is correct
- Enable debug and investigate further. It might be a java, a firewall, or a configuration issue. Refer to Debugging for more details. If the debug indicates that the issue is related to the instrumentation layer, use Sentry Software's Troubleshooting tools to identify the real source of the problem. If you subscribed to one of our support plan, you can also contact our Support Team.
- in the
When data is available in Prometheus but not in Grafana:
- Connect to Grafana
- Open the Host dashboard
- Edit the panel of the hardware components for which no data is available
- In the Metrics browser field, verify that the query and metrics labels are correct and make the corrections required.