Docker quick start (2020 notes)
Just my old cheatsheet for running containers

docker run -d -p 80:80 nginx
docker ps
docker exec -it container bash
docker network inspect bridge
docker network create my_network
docker volume inspect my-volume
docker build -t my-image .
docker container stop $(docker container ls -a -q) && docker container rm $(docker container ls -a -q)
docker service create --name vote -p 8080:80 instavote/vote
docker run -v /host-directory:/container-directory example-image:latest
Pretty basic stuff, but useful when you forget.

Kubernetes basics reminder
From when I played with minikube ~2021

kubectl get pods
kubectl apply -f deployment.yaml
kubectl port-forward svc/myapp 8080:80
kubectl get pods -A --field-selector spec.nodeName=ip-10-254-1-20
kubectl describe node ip-10-254-1-20| grep Taints
kubectl set image deployment/hostname www=image:v3
kubectl logs -f drone-agent-557ddc7bb4-crjns
Still works for local testing.

Prometheus + Grafana setup
Old homelab config

Run prometheus.yml with scrape_configs for node_exporter.
Grafana dashboards: import 1860 for node stats.
node_memory_MemAvailable_bytes/1024/1024 offset 10m
sum(rate(container_network_receive_bytes_total{cluster_name=~"$cluster_name", container_label_com_amazonaws_ecs_container_name=~"$service_name"}[$interval])) by (name, container_label_com_amazonaws_ecs_container_name, container_label_com_amazonaws_ecs_cluster)
sum(rate(container_network_transmit_bytes_total{cluster_name=~"$cluster_name", container_label_com_amazonaws_ecs_container_name=~"$service_name"}[$interval])) by (name, container_label_com_amazonaws_ecs_container_name, container_label_com_amazonaws_ecs_cluster)
( (1 - rate(node_cpu_seconds_total{job="node-exporter", mode="idle", instance="$instance"}[$__interval])) / ignoring(cpu) group_left count without (cpu)( node_cpu_seconds_total{job="node-exporter", mode="idle", instance="$instance"}) )
sum(rate(http_server_requests_seconds_count{status=~"5[0-9][0-9]"}[1m])) by (service, pod, uri) * 60
Nothing fancy.