Q & A Online Meetup: Deploy, Configure, and Monitor Traefik with Prometheus and Grafana with Brian Christner, 56K.Cloud
Check out the YouTube video: https://youtu.be/3q-K4JDcH6I
Repo: A Docker Swarm Stack for monitoring Traefik with Promethues and Grafana https://github.com/vegasbrianc/docker-traefik-prometheus
Question: You mentioned your Prometheus/Grafana would normally be node==worker deployments, for the demo it's just on the manager. Why is it necessary to add the constraint for them, doesn't Swarm do that automatically?
Answer: Yes, Swarm without a constraint automatically schedules services on the first node available. Typically, I add a constraint to monitoring services to run it separate from my workload. I try to dedicate nodes just for monitoring.
Question: How do you push to Slack from there? With Prometheus?
Answer: I detail this out in a separate Repo how to enable alerting to Slack. https://github.com/vegasbrianc/prometheus#alerting
Question: When do you use alerts through Prometheus vs. in Grafana? I actually prefer doing both.
Answer: Prometheus has a really powerful rules engine which then you can visualize and control the Alerts within Grafana. Grafana alerts are more visual and you drag sliders up and down to the desired number which triggers the alarm.
Question: We currently use Stackdriver in my team, will you advise to switch to Prometheus-Grafana?
Answer: I am a fan of Prometheus and Grafana as it has a huge community and great integrations.