Elastic Agent / Endpoint no longer can connect to fleet server

[Ting_sf (Ting Yu)]

I had elastic agent / end point installed on about 10 computers, they were working for pass months, I just noted today that all pcs are showing offline status, and I check logs, all show following error:

[elastic_agent][error] Could not communicate with fleet-server Checking API will retry, error: fail to checkin to fleet-server: Post "http://192.168.0.46:8220/api/fleet/agents/b449511d-27ff-4644-b2a2-bea4104574cf/checkin?": dial tcp 192.168.0.46:8220: connectex: No connection could be made because the target machine actively refused it.

wondering someone could offer me some troubleshooting idea. I did not change anything at my elastic server (I am using single node), no sure where to start.

thanks in advance.

[Ting_sf (Ting Yu)]

Hi, All;

I resolved it today. By some reason, the elastic-agent is not running on fleet server (even though I reboot the machine, still come back not restart), I have to do "sudo service elastic-agent restart" to start the service, then communication is back.

I was not see this coming, I thought all services are auto start at reboot. What I think what happened is that elasticsearch takes long time to restart, and some of other service got time out to restart (just what I think, not sure)

Anyway, it is resolved.