Hello, We have a corporative Jitsi server installed by a colleague of mine that we use to provide streaming services for the company. This is custom jitsi docker installation with some changes. We also have 3 Jibri servers in the test environment and 20 Jibri servers in production.
I was assigned a task of creating some sort of monitoring system especially for our Jibri servers. The goal is to check if any of the jibri instances has stopped and restart it in this case.
As I said this installation wasn’t done by me and I’m only starting to know Jitsi’s architecture. So initially I wanted to ask if this was possible, but I missed this community initially and created an issue in Jibri’s github page. This one: Control Jibri via API · Issue #492 · jitsi/jibri · GitHub
Where user Damencho said I could use curl
http://localhost:8888/debug to list Jibri’s instances.
So my initial thought was to check running Jibri instances with this command and then use Jibri’s http API to check if they were running.
The problem is that when I entered the server where Jitsi is running I saw no exposed port on Jicofo container. I entered the container and there was a running service listening in port 8888. So we exposed port 8888 in the docker-composer.yml file and tried the command
http://localhost:8888/debug, but I returned an HTML with a body of 404.
We also discovered this post: Jicofo health checks through rest API - #8 by shawn
And enabled it, but I’m getting an empty response.
Is there something else we need to enable in order to run
http://localhost:8888/about/health correctly ?