Metrics

Shiplon records how every app is doing, with nothing to install or configure. Measurements are taken every 30 seconds and kept for 30 days.

See an app’s charts

1. Open the app from Applications.
2. Click the Metrics tab.
3. Pick a period at the top right: 1 hour, 24 hours, 7 days or 30 days.

Hover over (or tap) a chart to see the exact value at that time. With a keyboard, focus the chart and use the arrow keys.

The summary

Tile	What it means
Uptime	The share of time the app passed its health check while it was deployed. Time while the app was stopped doesn’t count against it.
Requests	HTTP requests that reached the app in the period.
Server errors	The share of those requests answered with a 5xx status.
Bandwidth	Data sent to visitors (the size of responses).
Restarts	How many times the app was restarted after a crash or failed health checks.
Out of memory	How many times the app went over its 512 MB limit and was restarted. If this isn’t 0, a notice tells you so.

The charts

Chart	What it shows
CPU	Processor use, in cores. The dashed line is the app’s limit (0.5 core). An app at the limit is slowed down, not stopped.
Memory	Memory in use. The dashed line is the limit (512 MB). Going over it restarts the app.
Requests	Requests per minute.
Server errors (5xx)	5xx responses per minute.
Response time (p95)	95% of requests were answered within this time.
Bandwidth to visitors	Data sent per second.
Available	Whether the app passed its health check: 100% is ready, 0% is not.

A gap in a chart means nothing was recorded then, for example because the app wasn’t deployed.

See all your apps at once

Click Metrics in the sidebar. You’ll see:

• Deployments in the period: how many, how many succeeded or failed, and the typical time from start to live.
• Applications: a row per app with CPU and memory right now, and requests, server errors, uptime and bandwidth for the period. Click an app to open its charts.

What’s measured, and where

• CPU and memory are read from the server the app runs on.
• Requests, errors, response time and bandwidth are counted where traffic enters Shiplon, before it reaches your app. They include requests to your shiplon.app address and your custom domains.
• Uptime is recorded once a minute from the app’s health check.

Everyone in the organization, including viewers, can see metrics. Each organization only ever sees its own apps.

When there are no charts

• “Metrics aren’t set up on this server yet.” Metrics haven’t been turned on for this Shiplon installation. Deployments, logs and everything else still work.
• “The metrics store didn’t answer.” A temporary problem. Try again in a minute.
• A new app shows little or nothing: charts fill in as the app runs and receives traffic.

For what an app prints, see Logs.