habsi/authentik
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Revert "website: latest migration to new structure" (#11634)

Browse Source 
Revert "website: latest migration to new structure (#11522)"

This reverts commit 9a89a5f94b.
This commit is contained in:
Tana M Berry
2024-10-08 17:30:50 -05:00
committed by GitHub
parent 9a89a5f94b
commit 6b2fced1b9
336 changed files with 883 additions and 2636 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
website/docs/sys-mgmt/ops/monitoring.md
View File

@ -1,29 +0,0 @@
---
title: Monitoring
---
authentik can be easily monitored in multiple ways.
## Server monitoring
Configure your monitoring software to send requests to `/-/health/live/`, which will return a `HTTP 200` response as long as authentik is running. You can also send HTTP requests to `/-/health/ready/`, which will return `HTTP 200` if both PostgreSQL and Redis connections can be/have been established correctly.
## Worker monitoring
The worker container can be monitored by running `ak healthcheck` in the worker container. This will ping the worker and ensure it can communicate with redis as required.
## Outpost monitoring
Both kinds of outpost (proxy and LDAP) listen on a separate port (9300) and can be monitored by sending HTTP requests to `/outpost.goauthentik.io/ping`.
---
Both Docker Compose and Kubernetes deployments use these methods by default to determine when authentik is ready after starting, and to only route traffic to healthy instances; unhealthy instances are restarted.
## Metrics
Both the core authentik server and any outposts expose Prometheus metrics on a separate port (9300), which can be scraped to gather further insight into authentik's state. The metrics require no authentication, as they are hosted on a separate, non-exposed port by default.
You can find an example dashboard here: [grafana.com](https://grafana.com/grafana/dashboards/14837-authentik/)
![](../../install-config/dashboard.png)