Egress self-check
Stop recording incidents when the problem is your own outbound connectivity.
If the Checkmate instance itself loses outbound connectivity, every monitor fails at once and every one of them raises an incident, even though nothing being monitored is down.
The egress self-check prevents that. Checkmate periodically probes a set of reliability targets. While none of them can be reached, outbound connectivity is treated as down and monitor failures are not recorded as incidents.
It is configured under Settings > Monitoring > Egress self-check.
Settings
| Field | Notes |
|---|---|
| Enable egress self-check | Off by default |
| Reliability targets | One per line, up to 10, each 253 characters or fewer |
Leave the targets empty to use the defaults:
https://www.google.com/generate_204
https://www.cloudflare.com/cdn-cgi/traceTarget formats
The format of each line decides how it is probed:
| Format | Probe |
|---|---|
1.1.1.1 or example.com | ICMP ping |
dns.google:53 | TCP connection |
https://example.com | HTTP request |
Pick targets that are reliable and independent of each other and of whatever you are monitoring. Two targets behind the same provider will fail together for reasons that have nothing to do with your own egress.
Degraded checks
A check taken while egress is down is recorded with a degraded status instead of a plain failure and the Checks page has an Egress degraded filter so you can separate "the target was down" from "we could not get out".