Status Monitoring

Atom provides real-time uptime monitoring for all your services.

How It Works

When services are loaded on the dashboard, Atom checks each service's health status using the most appropriate method:

Method	When Used	Timeout
HTTP HEAD	External HTTP/HTTPS URLs	5 seconds
HTTP GET (fallback)	When HEAD returns 405	5 seconds
TCP Ping	Internal/private network services	2 seconds
ICMP Ping	Services with a custom `ping` field	System default

Batch Checking

Dashboard status checks are batched — up to 50 services checked simultaneously with a concurrency limit of 10 parallel requests to prevent overwhelming your network.

Status Indicators

State	Color	Condition
Up	Green	Response received, latency < 500ms
Slow	Yellow	Response received, latency ≥ 500ms
Down	Red	No response, timeout, or error
Loading	Pulse animation	Check in progress

Uptime History

Atom records every status check in the database. View historical data in the Uptime Widget or the Status Overview page (/apps).

Data tracked per check:

Timestamp
Status (up / slow / down)
HTTP response code
Response latency (ms)

Status Overview Page

The /apps page shows a summary of all services:

Total up / slow / down counts
Average latency per service
Uptime percentage (24h window)
Sortable and searchable

Notifications

The Notification Center (bell icon in header) alerts you when:

A service goes down (was previously up)
A service becomes slow (was previously up)
A service recovers (was down, now up)

Private Network Detection

Atom automatically detects internal services and uses TCP ping instead of HTTP:

localhost, 127.x.x.x
10.x.x.x, 172.16-31.x.x, 192.168.x.x
.local and .internal hostnames
Hostnames without dots

Custom Ping Host

For services behind load balancers or reverse proxies, you can set a custom ping host:

Edit the service
Set the Ping field to a specific IP or hostname
Atom uses ICMP ping instead of HTTP for status checks

This is useful when the HTTP endpoint returns a cached response but you want to check the actual server.