InsightX
InsightX is the tab in your service that explains incidents. When something goes wrong, such as a crash, a failing health check, a full disk, or a certificate that is about to expire, InsightX records it as an incident and tells you:
- What happened, in plain language.
- Why it happened, with the root cause.
- What it affected, so you know how serious it is.
- What to do next, as concrete steps.
You do not set anything up. Incidents appear in InsightX on their own.
Where to find it
Open your service and select InsightX in the service sidebar, between Activities and Usage Logs.
On the Resources tab, a resource that has incidents shows a View in InsightX link under its status. The link opens InsightX for that resource.
What InsightX explains
InsightX explains 14 kinds of incident. They fall into two groups:
| Group | Incidents | How the explanation is made |
|---|---|---|
| Crashes and failed health checks | Out of memory, a crash that recovered, a crash that stayed down, a failing health check | InsightX reads the evidence from your app and investigates. The result shows a confidence figure. |
| Conditions Kuberns measures | SSL certificate expiring or expired, disk getting full or almost full, high CPU, high memory, critical memory, a server out of memory, an app restarted to protect its server, an app that keeps crashing and recovering | Kuberns already knows the cause because it measured it, so the explanation is fixed and has no confidence figure. |
Incident types describes each one: when it appears, what you see, and what to do.
Which resources it covers
InsightX explains incidents on the app resources of your service, which are web services and workers.