The Elastic integration "NGINX" for elastic agent, when deployed over fleet does not fully respect the "Hosts" setting for the integration even though the tooltip says "The following settings are applicable to all inputs below."
With the default setting "http://127.0.0.1:80" the elastic agent tries to crawl the nginx metrics at https://127.0.0.1/ - i.e. it changes the protocol from http to https.
For self-managed-servers this results in a "certificate signed by unknown authority error".
The Elastic Agent NGINX integration should fully respect and utilize whatever is put into the host field and should not change the protocol from http to https.
Elastic agent log:
Error fetching data for metricset nginx.stubstatus: error fetching status: error making http request: Get "https://127.0.0.1/nginx_status": x509: certificate signed by unknown authority
Screenshot Elastic agent NGINX integration
