Link monitoring
An office or ISP connection watched from both ends, so a filtered port and a dead line never read the same.
Included on every plan.
| What breaks | hostnag |
|---|---|
| An ISP link dropsthe public address from several locations, confirmed by a ping from inside the site | Yes |
| hostnag: outside and inside | |
How it works
- 01
Add the public address
The public IP or hostname of the site's router or gateway, plus the port to reach it on (443 by default).
- 02
hostnag dials it from outside
From the group's locations, the same way uptime monitoring does.
- 03
Turn on the inside ping
A heartbeat the device on the far end calls on its own, so a filtered port and a truly dead line stop looking alike.
- 04
Five verdicts, one word each
Up, degraded, link down, inbound blocked, inside silent. Only link down raises an alert that pages anyone.
What you'll be told
What you get told
- Link
[Critical] Office link is down
Down from US East. Last error: connect_failed: ECONNREFUSED.
- Link
[Warning] Inbound blocked on Office
The site is online (a device inside is pinging) but port 443 is filtered from outside. Check the firewall or the ISP's CGNAT.
- Link
[Warning] Public IP changed for Office
Office now pings from 198.51.100.20 (was 203.0.113.8). A new lease from the ISP, or failover to a backup line.
Sample alerts, rendered by the same code that writes your email and Slack messages.
Settings
- Target
- public IP or hostname, plus a port
- Default port
- 443
- Inside ping
- optional heartbeat, its own expected interval and grace
- Public IP change
- raised as a Changed event
- Interval
- 5 min on Free, 1 min on Pro and Business
Questions
What is the difference between inbound blocked and link down?
Inbound blocked means the outside test is refused while the inside device keeps pinging; link down means neither side is heard from.
Do I need to install anything on the inside device?
No. It calls the ping URL on its own schedule, the same as a heartbeat.
Related
Add a host. Hear about it first.
Start free