Frequently Asked Questions
Common questions and troubleshooting tips for the Minimal Status Panel plugin.
š Installation & Setup
Q: The plugin doesnāt appear in my Grafana installation
A: Check these common issues:
- Plugin directory: Ensure files are in the correct location:
ls /var/lib/grafana/plugins/minimal-status-panel/ # Should show: plugin.json, module.js, README.md, etc. - Grafana restart: Always restart Grafana after plugin installation:
sudo systemctl restart grafana-server - File permissions: Ensure Grafana can read the plugin files:
sudo chown -R grafana:grafana /var/lib/grafana/plugins/minimal-status-panel - Grafana logs: Check for plugin loading errors:
sudo journalctl -u grafana-server -f
Q: I get āPlugin not foundā errors
A: This usually indicates:
- Plugin files are in the wrong directory
plugin.jsonis missing or corrupted- Grafana version compatibility issues
- Plugin signing issues (check Grafanaās plugin security settings)
š Data & Queries
Q: The panel shows āNo status data availableā
A: This is the most common issue. Check:
- Query returns data: Use Query Inspector to verify your query returns results
- Required labels: Ensure your metrics have an
instancelabel - Metric name: The plugin looks for
probe_successby default - Data source: Verify your Prometheus data source is working
Debug steps:
# Test this query in Grafana Explore:
probe_success
# Should return metrics with instance labels
Q: Why do my services show as āUnknownā?
A: The plugin determines status from the probe_success metric value:
- Value
1= Up (green) - Value
0= Down (red) - Missing/invalid values = Unknown (gray)
Check your Blackbox Exporter configuration and ensure itās returning proper 1/0 values.
Q: Heartbeat bars are not showing real data
A: The plugin uses time-series data from your query. If you see uniform patterns, it means:
- Not enough historical data in the selected time range
- Blackbox Exporter hasnāt been running long enough
- Time range is too short (try āLast 24 hoursā)
Q: How do I monitor non-HTTP services?
A: The plugin works with any Prometheus metric that has:
- An
instancelabel for service identification - Binary values (1 for up, 0 for down)
Examples:
# TCP port monitoring
probe_success{job="blackbox",module="tcp_connect"}
# Custom health checks
up{job="my-service"}
# Database connectivity
mysql_up{instance="db-server"}
šØ Display & Configuration
Q: Service names are too long/ugly
A: Use the Custom Service Names feature:
{
"https://very-long-api-endpoint.company.com:8080/health": "API Service",
"10.0.1.100:3306": "Database",
"monitoring-system.internal.company.com": "Monitoring"
}
Q: How do I hide URLs from public displays?
A: In the panel options:
- Set Show URLs to
false - Use Custom Service Names for user-friendly names
- Consider Ultra-minimal display level
Q: The panel takes up too much space
A: Try these space-saving options:
- Display Mode: Switch to
GridorCompact - Display Level: Use
Ultra-minimal - Max Items: Reduce the number
- Show Labels: Disable for compact mode
Q: Colors donāt match my theme
A: The plugin automatically adapts to Grafanaās light/dark themes. For custom colors:
- Use Grafanaās custom theme features
- Check your organizationās theme settings
- Consider dashboard-level styling options
ā” Performance & Optimization
Q: The panel is slow with many services
A: Optimize performance with:
- Reduce Max Items: Lower to 10-20 services per panel
- Increase Refresh Interval: Use 60+ seconds for overview dashboards
- Limit Time Range: Shorter ranges load faster
- Split Panels: Use multiple panels instead of one large panel
- Optimize Queries: Use specific label selectors
Q: Auto-refresh is too frequent/not working
A:
- Panel refresh: Set in panel options (5-300 seconds)
- Dashboard refresh: Set in dashboard settings
- Browser performance: Very short intervals may impact performance
- Data source caching: May affect refresh behavior
Q: Memory usage is high
A: The plugin limits heartbeat data to 50 points per service. High memory usage usually indicates:
- Too many services in one panel
- Very frequent refresh intervals
- Large time ranges with high-resolution data
š§ Advanced Configuration
Q: Can I use custom metrics instead of probe_success?
A: Yes! The plugin works with any metric that:
- Has an
instancelabel - Contains binary values (1/0 or true/false)
Example with custom metrics:
my_service_healthy{instance="service-name"}
Q: How do I group services by environment?
A: Use Prometheus label selectors:
# Production only
probe_success{environment="production"}
# Multiple environments
probe_success{environment=~"production|staging"}
Create separate panels for each environment group.
Q: Can I change the heartbeat bar size/color?
A: The heartbeat bars automatically:
- Resize based on available space
- Use theme-appropriate colors
- Show 50 data points maximum
Currently, these are not user-configurable, but may be added in future versions.
Q: How do I set up alerts with the status panel?
A: The panel is for visualization only. Set up Grafana alerts separately:
# Alert when service is down
probe_success == 0
# Alert on slow response
probe_duration_seconds > 5
The panel and alerts complement each other perfectly.
š Troubleshooting
Q: Data shows in Query Inspector but not in panel
A: Check:
- Instance labels: Ensure metrics have
instancelabels - Custom names JSON: Verify JSON syntax is valid
- Max items setting: You might have hit the limit
- Time range: Ensure data exists in the selected range
Q: Tooltips on heartbeat bars donāt work
A: This usually indicates:
- Browser compatibility issues (try a different browser)
- Missing timestamp data
- CSS conflicts with dashboard themes
Q: Panel shows old data after updates
A: Try:
- Manual refresh: Click dashboard refresh button
- Clear browser cache: Force refresh (Ctrl+F5)
- Check data source cache: May need to be cleared
- Verify time range: Ensure it includes recent data
Q: Services appear and disappear randomly
A: This typically means:
- Inconsistent metrics: Services reporting intermittently
- Label changes: Instance labels changing over time
- Query issues: Query returning different results
- Data retention: Old data being pruned
š Known Issues & Limitations
Q: What are the current limitations?
A: Known limitations include:
- Maximum services: Performance degrades with 100+ services
- Heartbeat resolution: Limited to 50 data points
- Custom colors: Limited theme customization options
- Mobile optimization: Some display modes work better than others
- Historical data: Depends on Prometheus retention settings
Q: Are there any browser compatibility issues?
A: The plugin works best with:
- Chrome/Edge: Full feature support
- Firefox: Full feature support
- Safari: Full feature support
- Mobile browsers: Basic functionality
Q: What about Grafana version compatibility?
A:
- Minimum: Grafana 9.0.0
- Recommended: Grafana 10.0+
- Latest: Tested with current Grafana versions
š Getting Help
Q: Where can I report bugs or request features?
A:
- GitHub Issues: Report bugs
- Feature Requests: Use GitHub Issues with āenhancementā label
- Documentation: This documentation covers most use cases
Q: How can I contribute?
A: Contributions welcome!
- Code: Submit pull requests on GitHub
- Documentation: Help improve these docs
- Bug Reports: Detailed bug reports help everyone
- Testing: Try new features and provide feedback
Q: Is commercial support available?
A: This is an open-source project. Community support is available through:
- GitHub Issues
- Documentation
- Community forums
Q: Can I modify the plugin for my needs?
A: Absolutely! The plugin is MIT licensed:
- Fork the repository
- Make your modifications
- Consider contributing improvements back
š Additional Resources
- Getting Started Guide - Installation and setup
- Configuration Guide - Detailed option explanations
- Display Modes - Visual examples and use cases
- Advanced Usage - Complex configurations and tips
- API Reference - Technical documentation
- GitHub Repository - Source code and issues
Still need help? Create an issue on GitHub with:
- Grafana version
- Plugin version
- Your configuration (sanitized)
- Expected vs actual behavior
- Any error messages