Creating a Probe
Set up continuous SIP monitoring with scheduled calls, threshold-based health evaluation, and automated alerting for your VoIP infrastructure.
Probes provide continuous, automated monitoring of your SIP infrastructure. Unlike tests which run on demand, probes execute on a schedule -- placing a single SIP call at regular intervals and evaluating quality metrics against configurable thresholds. When metrics breach thresholds, the probe's health status changes and configured webhooks fire to alert your team.
Probes vs. Tests
Before creating a probe, understand how probes differ from tests:
| Aspect | Test | Probe |
|---|---|---|
| Execution | On-demand, manual trigger | Scheduled, automatic |
| Scale | 1 to thousands of endpoints | Single endpoint per execution |
| Duration | Configurable (seconds to hours) | Short, fixed-duration calls |
| Purpose | Load testing, stress testing, capacity planning | Continuous quality monitoring |
| Output | Detailed per-run metrics and analytics | Health status (Healthy/Degraded/Unhealthy) with metric snapshots |
| Alerting | None (manual review) | Webhooks on status changes |
| Status pages | No | Yes -- probes power public status pages |
Use tests when you need to measure capacity or compare configurations. Use probes when you need ongoing visibility into the health of your SIP infrastructure.
When to Use Probes
Probes are ideal for:
- 24/7 production monitoring -- continuously verify that your SIP server is reachable and call quality meets expectations
- SIP trunk health checks -- monitor the quality of your SIP trunk provider over time
- SLA verification -- collect objective quality data to validate (or dispute) SIP provider SLAs
- Early warning system -- detect quality degradation before users start complaining
- Regression detection -- catch quality regressions after infrastructure changes (firmware updates, route changes, capacity additions)
- Multi-site monitoring -- monitor call quality from multiple geographic regions to your central SIP infrastructure
Creating a Probe
- Open your project from the sidebar
- Click Probes in the project navigation
- Click New Probe
- Configure the probe settings (detailed below)
- Click Create Probe
The probe begins executing on the configured schedule immediately after creation.
Probe Configuration
Basic Settings
| Field | Required | Description |
|---|---|---|
| Name | Yes | A descriptive label (e.g., "Production PBX Health", "SIP Trunk A - US East", "IVR System Check") |
| Registrar | Yes | The SIP server this probe monitors. Select from your project's configured registrars |
| Mode | Yes | Caller or Callee -- determines the probe's behavior (see below) |
Probe Modes
Caller Mode
The probe registers with your SIP server and actively places a call to a configured target URI.
How it works:
- Endpoint registers with the registrar using assigned SIP account
- Endpoint sends a SIP INVITE to the target URI
- If the call is answered, media exchange begins
- Quality metrics are collected during the call
- The call terminates after the configured duration
- Metrics are evaluated against thresholds to determine health status
Use Caller mode for:
- Testing outbound call quality from your SIP server
- Monitoring SIP trunk availability and quality
- Verifying PBX call routing and media quality
- Testing IVR or auto-attendant destinations
Callee target configuration: Enter the SIP URI that the probe should call (e.g., sip:echo@sip.example.com, sip:1000@your-pbx.com).
Echo test targets
Many SIP servers offer an echo test extension that answers calls and plays back the received audio. This is ideal for probe targets because it verifies the complete round-trip media path. Check your PBX documentation for echo test configuration.
Callee Mode
The probe registers with your SIP server and waits for incoming calls from an external system.
How it works:
- Endpoint registers with the registrar using assigned SIP account
- Endpoint waits for an incoming SIP INVITE
- When a call arrives, it answers automatically
- Media exchange and metric collection proceed as normal
- Health status is evaluated after the call
Use Callee mode for:
- Monitoring inbound call paths to your SIP server
- Validating media quality from external call sources
- Testing incoming call routing through your PBX
- Verifying that external systems can reach your infrastructure
External caller required
In Callee mode, something external to CallMeter must call the probe's registered endpoint. If no call arrives within the timeout window, the probe reports UNKNOWN health status. Ensure your external call source is configured and running.
Scheduling
Probes support two scheduling modes:
Interval Scheduling
Select a fixed interval between executions:
| Interval | Executions per Day | Best For |
|---|---|---|
| Every 5 minutes | 288 | Critical production systems requiring near-real-time monitoring |
| Every 15 minutes | 96 | Standard production monitoring |
| Every 30 minutes | 48 | Non-critical systems or cost-conscious monitoring |
| Every 60 minutes | 24 | Baseline monitoring, development environments |
Cron Scheduling
Define a cron expression for precise scheduling control:
| Expression | Schedule |
|---|---|
*/15 * * * * | Every 15 minutes, 24/7 |
0 9-17 * * 1-5 | Every hour during business hours (9 AM - 5 PM), Monday through Friday |
0 */2 * * * | Every 2 hours, 24/7 |
0 8,12,17 * * * | Three times per day at 8 AM, 12 PM, and 5 PM |
Cron scheduling is useful when you want to monitor only during business hours, avoid weekend executions, or align monitoring with known traffic patterns.
Media Configuration
Configure the audio codec (and optionally video) for probe calls. The media configuration is identical to test group configuration:
- Audio codec -- select from PCMA, PCMU, G.722, or Opus
- Video -- enable or disable video, with codec and resolution settings
Match your production codec
Configure the probe to use the same codec as your production environment. This ensures the quality metrics reflect real-world conditions. If your PBX uses PCMA for most calls, use PCMA for the probe.
Threshold Configuration
Set warning and critical thresholds for quality metrics. These thresholds determine when the probe's health status changes. See Threshold Configuration for detailed guidance.
Webhook Configuration
Set up webhook URLs to receive notifications when the probe's health status changes. See Webhooks for setup instructions and payload format.
Viewing Probe Results
After the first execution, the probe dashboard shows:
Current Health Status
A prominent badge showing the probe's current health:
- HEALTHY (green) -- all metrics within acceptable ranges
- DEGRADED (yellow) -- one or more metrics exceed warning thresholds
- UNHEALTHY (red) -- one or more metrics exceed critical thresholds
- UNKNOWN (gray) -- no data collected or evaluation error
Health History Timeline
A chronological view of health status changes, showing when the probe transitioned between states. This timeline helps identify patterns -- for example, regular degradation during peak hours or intermittent issues at specific times.
Latest Metric Snapshot
The most recent execution's metric values:
- MOS score
- Jitter
- Packet loss
- RTT
- Setup time
- Any metrics that triggered threshold alerts
Execution History
A list of past probe executions with their individual results. Click any execution to see detailed metrics and the SIP message trace for that specific call.
Plan Limits on Probes
Your subscription plan defines the maximum number of probes you can create per project and per organization. Check your current probe count and limits in the organization billing settings.
Higher-tier plans support more probes with shorter intervals, enabling comprehensive monitoring of large SIP deployments.
Best Practices
- One probe per critical path -- create separate probes for each SIP trunk, PBX, or call route you need to monitor
- Match production configuration -- use the same codec, transport, and credentials that production traffic uses
- Set meaningful thresholds -- start with industry-standard values (MOS warning at 3.5, critical at 2.5) and tune based on your environment
- Configure webhooks -- probes without alerting are dashboards without visibility. Always set up at least one webhook
- Monitor from multiple regions -- if your SIP server handles calls from different geographies, create probes in each relevant region
- Use descriptive names -- "Prod PBX US East" is more useful than "Probe 1" when scanning a list of probes
Next Steps
- Threshold Configuration -- Set warning and critical thresholds
- Webhooks -- Configure alert notifications
- Status Pages -- Create public health dashboards
- Probe Health States -- Detailed health state reference