Summary
Introduce a built-in alerting system that notifies administrators when critical events or abnormal operating conditions occur in NethVoice.
Problem
Administrators often become aware of issues only after users report them. While many system metrics can already be monitored using external tools, NethVoice currently lacks a native mechanism to generate meaningful alerts based on the status of its core services and telephony components.
Providing built-in alerts would enable administrators to react more quickly to service disruptions and reduce downtime.
Proposed Solution
Add a configurable alerting framework that can generate notifications when predefined events or thresholds are reached.
Examples of useful alerts include:
Telephony
- SIP trunk availability and registration
- A significant number of device becoming unregistered
- A significant number of failed registration attempts on Asterisk
- A significant number of users are logging out of NethVoice CTI
- High rate of failed calls
- RTP jitter
- RTP packet loss
- DNS Failure
Core Services
- Asterisk service down
- Mariadb service down and response time
- NethVoice CTI server service down
- NethVoice CTI WebSocket unavailable
- Janus gateway service down
- Kamailio service down
- RTP engine service down
- Phonebook DB service down
- Tancredi/provisioning service down
- Satellite service down, when enabled
- Satellite MQTT/Postgres/RTP service down, when enabled
- Reports API/Redis/UI down, when installed
AI Services
- Error from the API keys of Deepgram or OpenAI, like credit exhausted or other errors
Recording
- Recording storage nearing capacity
Infrastructure
- Low disk space
- High CPU or memory usage
- Load average
- TLS certificate approaching expiration
- Backup failure
Benefits
- Faster awareness of service disruptions
- Reduced mean time to detect (MTTD)
- Improved system reliability
- Better operational visibility
- Reduced dependency on external monitoring platforms for basic operational alerts
Summary
Introduce a built-in alerting system that notifies administrators when critical events or abnormal operating conditions occur in NethVoice.
Problem
Administrators often become aware of issues only after users report them. While many system metrics can already be monitored using external tools, NethVoice currently lacks a native mechanism to generate meaningful alerts based on the status of its core services and telephony components.
Providing built-in alerts would enable administrators to react more quickly to service disruptions and reduce downtime.
Proposed Solution
Add a configurable alerting framework that can generate notifications when predefined events or thresholds are reached.
Examples of useful alerts include:
Telephony
Core Services
AI Services
Recording
Infrastructure
Benefits