Feature Request
Currently connmon allows monitoring only a single IP/host to determine internet connectivity.
This can lead to false positives when the monitored IP becomes temporarily unreachable (e.g., remote server issue, routing issue, ICMP blocked), even though the internet connection is still working.
Proposed Enhancement
Add support for multiple monitoring IPs (e.g., 2 or 3), with configurable behavior such as:
-
Failover mode
- If primary IP fails, test secondary IP(s) before declaring connection down.
-
Quorum mode
- Consider connection DOWN only if all configured IPs fail.
- Consider connection UP if at least one IP responds.
-
Optional:
- Configurable threshold (e.g., "minimum X successful responses out of Y IPs").
- Separate timeout per host.
- Round-robin testing instead of sequential.
Example Use Case
For example:
MONITOR_IPS="1.1.1.1 8.8.8.8 9.9.9.9"
If 1.1.1.1 fails but 8.8.8.8 responds, connection should still be considered UP.
This would significantly reduce false internet-down triggers caused by single endpoint failures.
Why This Matters
Many users rely on connmon for:
- WAN failover
- Auto-reconnect logic
- Monitoring stability
Single-endpoint monitoring is inherently fragile. Multi-endpoint support would make connmon more robust and production-ready.
Feature Request
Currently connmon allows monitoring only a single IP/host to determine internet connectivity.
This can lead to false positives when the monitored IP becomes temporarily unreachable (e.g., remote server issue, routing issue, ICMP blocked), even though the internet connection is still working.
Proposed Enhancement
Add support for multiple monitoring IPs (e.g., 2 or 3), with configurable behavior such as:
Failover mode
Quorum mode
Optional:
Example Use Case
For example:
MONITOR_IPS="1.1.1.1 8.8.8.8 9.9.9.9"
If 1.1.1.1 fails but 8.8.8.8 responds, connection should still be considered UP.
This would significantly reduce false internet-down triggers caused by single endpoint failures.
Why This Matters
Many users rely on connmon for:
Single-endpoint monitoring is inherently fragile. Multi-endpoint support would make connmon more robust and production-ready.