docker
building prometheus inside a container
Don’t install go, instead build it inside the go container;
docker run --rm -v %CD%:/go/src/github.com/prometheus/prometheus -w /go/src/github.com/prometheus/prometheus golang:1.10 make build
Run it inside that docker container too!
docker run --rm -v %CD%:/go/src/github.com/prometheus/prometheus -w /go/src/github.com/prometheus/prometheus -p 9090:9090 golang:1.10 ./prometheus
Note: I executed the following to have it use a test config;
copy ./config/testdata/conf.good.yml ./prometheus.yml
bringing up containers in azure
Create a resource group
az group create --name myGroup --location eastus
Create the container
az container create --resource-group myGroup --name myName --image tensorflow/tensorflow --dns-name-label myDnsName --ports 8888
Show status
az container show --resource-group myGroup --name myName --query "{FQDN:ipAddress.fqdn,ProvisioningState:provisioningState}" --out table
set container timezone
Set the TZ
environment variable inside linux containers to set the timezone, for example docker run -e TZ={timezone}
or, for example inside your docker-compose
;
environment:
TZ: "Australia/Sydney"
using docker without sudo
sudo gpasswd -a $USER docker
azure container repository login
az login
az acr login --name repoName
accessing windows event viewer
updating running containers
# set to always restart
docker update --restart=always <container>
# rename
docker rename <container> <new name>
run container indefinitely
Set the command to tail -f /dev/null
Example:
docker run -d node tail -f /dev/null
install docker via cloudformation
docker swarm ports
docker disk usage
Find large directories
du -ah / | sort -h -r | head -n 20
Find containers using a given volume
docker ps -a --filter volume=<volume>
Remove container and its associated volume
docker rm -v container
See total volume, image, container space
docker system df
See space used per volume
docker system df -v
Remove dangling volumes from disk
docker volume prune
wsl uses all the ram
yolo
Force linux to drop the page cache
echo 1 | sudo tee /proc/sys/vm/drop_caches
limit memory assigned to WSL2 VM
Create a %UserProfile%\.wslconfig
file
[wsl2]
memory=6GB
swap=0
localhostForwarding=true
add container to another container's network
For example, adding the container some-postgis
to the same network as the container postgres_adminer_1
.
Find the NetworkID
of the network:
docker inspect postgres_adminer_1 | grep NetworkID
Add the container to the network:
docker network connect f51952f2eb6ea17fbf1917b3609b99f2a70cce8b4e1e1c35177bcb4388ff1124 some-postgis
remove containers by image
For the image verdaccio/verdaccio
:
docker rm $(docker ps -q -a --filter ancestor=verdaccio/verdaccio)
posts
- Running Containers for Free on fly.io
- Continuous Deployment to Kubernetes - Part 2
- Continuous Deployment to Kubernetes - Part 1
- Minimal Elasticsearch Resources in Kubernetes
- Static Code Analysis for .NET with SonarQube
- This Social Image was Generated
- Load testing ASP.NET Core SignalR
- Docker containers aren't just for long running tasks
- Noob tips for Traefik
- ASP.NET 5 on the Amazon (Linux) Cloud
- Optimising WordPress and MySQL docker images for Amazon EC2 micro instances
- Gnashing of Teeth on the Bleeding Edge
- .NET Development on Linux
links
The next version of docker (v4.40) will add native llm capability to the docker cli. Docker Model Runner is not yet publicly released, but adds commands like docker model run
that will run LLM models outside of containers. Initial reports look promising and may be nice replacement for running llama.cpp, koboldcpp or ollama locally.