mohsenasm
commented
1 year ago Hi @mbseid, Could you send some logs?
Please ssh into the manager (newman) and send the output of these commands:
docker service ps homelab_cadvisordocker exec -it homelab_cadvisor<tab> sh- Then,
curl -sS http://localhost:8080/metrics | grep memory_
- Then,
The above log will contain your services' name & image. If you don't want to send those, use these commands instead:
docker service ps homelab_cadvisor- Copy the
IDof the current task running in thenewmanserver.
- Copy the
docker exec -it homelab_cadvisor<tab> sh- Then,
curl -sS http://localhost:8080/metrics | grep -E 'memory_.*<the_copied_id>'
- Then,
trajano
mbseid
schenklklopfer
First, great project. Thank you!
I set up the dashboard on a new swarm cluster and it's not showing the memory usage. CPU usage is being captured. Picture attached:
I checked the websocket information and it's sending 0 on the wire. It may not be collecting it properly.
Additional context:
There is a two node cluster. It's running a few levels of virtualization (which shouldn't be the cause). Proxmox => Ubuntu => Docker Swarm
Thanks for the help!