A caching proxy for Docker; allows centralised management of (multiple) registries and their authentication; caches images from any registry. Caches the potentially huge blob/layer requests (for bandwidth/time savings), and optionally caches manifest requests ("pulls") to avoid rate-limiting.
Starting November 2nd, 2020, DockerHub will
supposedly
start
rate-limiting pulls,
also known as the Docker Apocalypse.
The main symptom is Error response from daemon: toomanyrequests: Too Many Requests. Please see https://docs.docker.com/docker-hub/download-rate-limit/ during pulls.
Many unknowing Kubernetes clusters will hit the limit, and struggle to configure imagePullSecrets and imagePullPolicy.
Since version 0.6.0, this proxy can be configured with the env var ENABLE_MANIFEST_CACHE=true which provides
configurable caching of the manifest requests that DockerHub throttles. You can then fine-tune other parameters to your needs.
Together with the possibility to centrally inject authentication (since 0.3x), this is probably one of the best ways to bring relief to your distressed cluster, while at the same time saving lots of bandwidth and time.
Note: enabling manifest caching, in its default config, effectively makes some tags immutable. Use with care. The configuration ENVs are explained in the Dockerfile, relevant parts included below.
# Manifest caching tiers. Disabled by default, to mimick 0.4/0.5 behaviour.
# Setting it to true enables the processing of the ENVs below.
# Once enabled, it is valid for all registries, not only DockerHub.
# The envs *_REGEX represent a regex fragment, check entrypoint.sh to understand how they're used (nginx ~ location, PCRE syntax).
ENV ENABLE_MANIFEST_CACHE="false"
# 'Primary' tier defaults to 10m cache for frequently used/abused tags.
# - People publishing to production via :latest (argh) will want to include that in the regex
# - Heavy pullers who are being ratelimited but don't mind getting outdated manifests should (also) increase the cache time here
ENV MANIFEST_CACHE_PRIMARY_REGEX="(stable|nightly|production|test)"
ENV MANIFEST_CACHE_PRIMARY_TIME="10m"
# 'Secondary' tier defaults any tag that has 3 digits or dots, in the hopes of matching most explicitly-versioned tags.
# It caches for 60d, which is also the cache time for the large binary blobs to which the manifests refer.
# That makes them effectively immutable. Make sure you're not affected; tighten this regex or widen the primary tier.
ENV MANIFEST_CACHE_SECONDARY_REGEX="(.*)(\d|\.)+(.*)(\d|\.)+(.*)(\d|\.)+"
ENV MANIFEST_CACHE_SECONDARY_TIME="60d"
# The default cache duration for manifests that don't match either the primary or secondary tiers above.
# In the default config, :latest and other frequently-used tags will get this value.
ENV MANIFEST_CACHE_DEFAULT_TIME="1h"Essentially, it's a man in the middle: an intercepting proxy based on nginx, to which all docker traffic is directed using the HTTPS_PROXY mechanism and injected CA root certificates.
The main feature is Docker layer/image caching, including layers served from S3, Google Storage, etc.
As a bonus it allows for centralized management of Docker registry credentials, which can in itself be the main feature, eg in Kubernetes environments.
You configure the Docker clients (err... Kubernetes Nodes?) once, and then all configuration is done on the proxy -- for this to work it requires inserting a root CA certificate into system trusted root certs.
:latestand:latest-debugDocker tag is unstable, built from master, and amd64-only- Production/stable is
0.6.2, see 0.6.2 tag on Github - this image is multi-arch amd64/arm64 - The previous version is
0.5.0, without any manifest caching, see 0.5.0 tag on Github - this image is multi-arch amd64/arm64
- DockerHub image is at
rpardini/docker-registry-proxy:<version> - GitHub image is at
ghcr.io/rpardini/docker-registry-proxy:<version> - Since 0.5.x, they both carry the same images
- This can be useful if you're already hitting DockerHub's rate limits and can't pull the proxy from DockerHub
- Run the proxy on a host close (network-wise: high bandwidth, same-VPC, etc) to the Docker clients
- Expose port 3128 to the network
- Map volume
/docker_mirror_cachefor up toCACHE_MAX_SIZE(32gb by default) of cached images across all cached registries - Map volume
/ca, the proxy will store the CA certificate here across restarts. Important this is security sensitive. - Env
ALLOW_PUSH: This bypasses the proxy when pushing, default to false - if kept to false, pushing will not work. For more info see this commit. - Env
CACHE_MAX_SIZE(default32g): set the max size to be used for caching local Docker image layers. Use Nginx sizes. - Env
ENABLE_MANIFEST_CACHE, see the section on pull rate limiting. - Env
REGISTRIES: space separated list of registries to cache; no need to include DockerHub, its already done internally. - Env
AUTH_REGISTRIES: space separated list ofhostname:username:passwordauthentication info.hostnames listed here should be listed in the REGISTRIES environment as well, so they can be intercepted.
- Env
AUTH_REGISTRIES_DELIMITERto change the separator between authentication info. By default, a space: "AUTH_REGISTRIES_DELIMITER=";;;". In that case,AUTH_REGISTRIEScould contain something likeregistry1.com:user1:pass1;;;registry2.com:user2:pass2. - Env
AUTH_REGISTRY_DELIMITERto change the separator between authentication info parts. By default, a colon: ":". If you use keys that contain single colons, you should update this variable, e.g. setting it toAUTH_REGISTRIES_DELIMITER=":::". In that case,AUTH_REGISTRIEScould contain something likeregistry1.com:::user1:::pass1 registry2.com:::user2:::pass2. - Env
PROXY_REQUEST_BUFFERING: If push is allowed, buffering requests can cause issues on slow upstreams. If you have trouble pushing, set this tofalsefirst, then fix remainig timeouts. Default istrueto not change default behavior. ENV PROXY_REQUEST_BUFFERING="true" - Timeouts ENVS - all of them can pe specified to control different timeouts, and if not set, the defaults will be the ones from
Dockerfile. The directives will be added intohttpblock.:- SEND_TIMEOUT : see send_timeout
- CLIENT_BODY_TIMEOUT : see client_body_timeout
- CLIENT_HEADER_TIMEOUT : see client_header_timeout
- KEEPALIVE_TIMEOUT : see [keepalive_timeout](http://nginx.org/en/docs/http/ngx_http_core_module.html#keepalive_timeout
- PROXY_READ_TIMEOUT : see proxy_read_timeout
- PROXY_CONNECT_TIMEOUT : see proxy_connect_timeout
- PROXY_SEND_TIMEOUT : see proxy_send_timeout
- PROXY_CONNECT_READ_TIMEOUT : see proxy_connect_read_timeout
- PROXY_CONNECT_CONNECT_TIMEOUT : see proxy_connect_connect_timeout
- PROXY_CONNECT_SEND_TIMEOUT : see proxy_connect_send_timeout)
docker run --rm --name docker_registry_proxy -it \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
rpardini/docker-registry-proxy:0.6.2For Docker Hub authentication:
hostnameshould beauth.docker.iousernameshould NOT be an email, use the regular username
docker run --rm --name docker_registry_proxy -it \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
-e REGISTRIES="k8s.gcr.io gcr.io quay.io your.own.registry another.public.registry" \
-e AUTH_REGISTRIES="auth.docker.io:dockerhub_username:dockerhub_password your.own.registry:username:password" \
rpardini/docker-registry-proxy:0.6.2For regular registry auth (HTTP Basic), the hostname should be the registry itself... unless your registry uses a different auth server.
See the example above for DockerHub, adapt the your.own.registry parts (in both ENVs).
This should work for quay.io also, but I have no way to test.
GitLab may use a different/separate domain to handle the authentication procedure.
Just like DockerHub uses auth.docker.io, GitLab uses its primary (git) domain for the authentication.
If you run GitLab on git.example.com and its registry on reg.example.com, you need to include both in REGISTRIES and use the primary domain for AUTH_REGISTRIES.
For GitLab.com itself the authentication domain should be gitlab.com.
docker run --rm --name docker_registry_proxy -it \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
-e REGISTRIES="reg.example.com git.example.com" \
-e AUTH_REGISTRIES="git.example.com:USER:PASSWORD" \
rpardini/docker-registry-proxy:0.6.2For Google Container Registry (GCR), username should be _json_key and the password should be the contents of the service account JSON.
Check out GCR docs.
The service account key is in JSON format, it contains spaces (" ") and colons (":").
To be able to use GCR you should set AUTH_REGISTRIES_DELIMITER to something different than space (e.g. AUTH_REGISTRIES_DELIMITER=";;;") and AUTH_REGISTRY_DELIMITER to something different than a single colon (e.g. AUTH_REGISTRY_DELIMITER=":::").
Example with GCR using credentials from a service account from a key file servicekey.json:
docker run --rm --name docker_registry_proxy -it \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
-e REGISTRIES="k8s.gcr.io gcr.io quay.io your.own.registry another.public.registry" \
-e AUTH_REGISTRIES_DELIMITER=";;;" \
-e AUTH_REGISTRY_DELIMITER=":::" \
-e AUTH_REGISTRIES="gcr.io:::_json_key:::$(cat servicekey.json);;;auth.docker.io:::dockerhub_username:::dockerhub_password" \
rpardini/docker-registry-proxy:0.6.2Kind is a tool for running local Kubernetes clusters using Docker container “nodes”.
Because cluster nodes are Docker containers, docker-registry-proxy needs to be in the same docker network.
Example joining the kind docker network and using hostname docker-registry-proxy as hostname :
docker run --rm --name docker_registry_proxy -it \
--net kind --hostname docker-registry-proxy \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
rpardini/docker-registry-proxy:0.6.2Now deploy your Kind cluster and then automatically configure the nodes with the following script :
#!/bin/sh
KIND_NAME=${1-kind}
SETUP_URL=http://docker-registry-proxy:3128/setup/systemd
pids=""
for NODE in $(kind get nodes --name "$KIND_NAME"); do
docker exec "$NODE" sh -c "\
curl $SETUP_URL \
| sed s/docker\.service/containerd\.service/g \
| sed '/Environment/ s/$/ \"NO_PROXY=127.0.0.0\/8,10.0.0.0\/8,172.16.0.0\/12,192.168.0.0\/16\"/' \
| bash" & pids="$pids $!" # Configure every node in background
done
wait $pids # Wait for all configurations to endK3d is similar to Kind but is based on k3s. In order to run with its registry you need to setup settings like shown below.
# docker-registry-proxy
docker run -d --name registry-proxy --restart=always \
-v /tmp/registry-proxy/mirror_cache:/docker_mirror_cache \
-v /tmp/registry-proxy/certs:/ca \
rpardini/docker-registry-proxy:0.6.4
export PROXY_HOST=registry-proxy
export PROXY_PORT=3128
export NOPROXY_LIST="localhost,127.0.0.1,0.0.0.0,10.0.0.0/8,172.16.0.0/12,192.168.0.0/16,.local,.svc"
cat <<EOF > /etc/k3d-proxy-config.yaml
apiVersion: k3d.io/v1alpha3
kind: Simple
name: mycluster
servers: 1
agents: 0
options:
k3d:
wait: true
timeout: "60s"
kubeconfig:
updateDefaultKubeconfig: true
switchCurrentContext: true
env:
- envVar: HTTP_PROXY=http://$PROXY_HOST:$PROXY_PORT
nodeFilters:
- all
- envVar: HTTPS_PROXY=http://$PROXY_HOST:$PROXY_PORT
nodeFilters:
- all
- envVar: NO_PROXY='$NOPROXY_LIST'
nodeFilters:
- all
volumes:
- volume: $REGISTRY_DIR/docker_mirror_certs/ca.crt:/etc/ssl/certs/registry-proxy-ca.pem
nodeFilters:
- all
EOF
k3d cluster create --config /etc/k3d-proxy-config.yamlSeparate instructions for Mac clients available in this dedicated Doc Desktop for Mac document.
Let's say you setup the proxy on host 192.168.66.72, you can then curl http://192.168.66.72:3128/ca.crt and get the proxy CA certificate.
On each Docker host that is to use the cache:
- Configure Docker proxy pointing to the caching server
- Add the caching server CA certificate to the list of system trusted roots.
- Restart
dockerd
Do it all at once, tested on Ubuntu Xenial, Bionic, and Focal, all systemd based:
# Add environment vars pointing Docker to use the proxy
mkdir -p /etc/systemd/system/docker.service.d
cat << EOD > /etc/systemd/system/docker.service.d/http-proxy.conf
[Service]
Environment="HTTP_PROXY=http://192.168.66.72:3128/"
Environment="HTTPS_PROXY=http://192.168.66.72:3128/"
EOD
### UBUNTU
# Get the CA certificate from the proxy and make it a trusted root.
curl http://192.168.66.72:3128/ca.crt > /usr/share/ca-certificates/docker_registry_proxy.crt
echo "docker_registry_proxy.crt" >> /etc/ca-certificates.conf
update-ca-certificates --fresh
###
### CENTOS
# Get the CA certificate from the proxy and make it a trusted root.
curl http://192.168.66.72:3128/ca.crt > /etc/pki/ca-trust/source/anchors/docker_registry_proxy.crt
update-ca-trust
###
# Reload systemd
systemctl daemon-reload
# Restart dockerd
systemctl restart docker.serviceClear dockerd of everything not currently running: docker system prune -a -f beware
Then do, for example, docker pull k8s.gcr.io/kube-proxy-amd64:v1.10.4 and watch the logs on the caching proxy, it should list a lot of MISSes.
Then, clean again, and pull again. You should see HITs! Success.
Do the same for docker pull ubuntu and rejoice.
Test your own registry caching and authentication the same way; you don't need docker login, or .docker/config.json anymore.
Since 0.4 there is a separate -debug version of the image, which includes nginx-debug, and (since 0.5.x) has a mitmproxy (actually mitmweb) inserted after the CONNECT proxy but before the caching logic, and a second mitmweb between the caching layer and DockerHub.
This allows very in-depth debugging. Use sparingly, and definitely not in production.
docker run --rm --name docker_registry_proxy -it
-e DEBUG_NGINX=true -e DEBUG=true -e DEBUG_HUB=true -p 0.0.0.0:8081:8081 -p 0.0.0.0:8082:8082 \
-p 0.0.0.0:3128:3128 -e ENABLE_MANIFEST_CACHE=true \
-v $(pwd)/docker_mirror_cache:/docker_mirror_cache \
-v $(pwd)/docker_mirror_certs:/ca \
rpardini/docker-registry-proxy:0.6.2-debugDEBUG=trueenables the mitmweb proxy between Docker clients and the caching layer, accessible on port 8081DEBUG_HUB=trueenables the mitmweb proxy between the caching layer and DockerHub, accessible on port 8082 (since 0.5.x)DEBUG_NGINX=trueenables nginx-debug and debug logging, which probably is too much. Seriously.
- If you authenticate to a private registry and pull through the proxy, those images will be served to any client that can reach the proxy, even without authentication. beware
- Repeat, this will make your private images very public if you're not careful.
Currently you cannot push images while using the proxy which is a shame. PRs welcome.SEEALLOW_PUSHENV FROM USAGE SECTION.- Setting this on Linux is relatively easy.
Yes, Docker offers Registry as a pull through cache, unfortunately
it only covers the DockerHub case. It won't cache images from quay.io, k8s.gcr.io, gcr.io, or any such, including any private registries.
That means that your shiny new Kubernetes cluster is now a bandwidth hog, since every image will be pulled from the Internet on every Node it runs on, with no reuse.
This is due to the way the Docker "client" implements --registry-mirror, it only ever contacts mirrors for images
with no repository reference (eg, from DockerHub).
When a repository is specified dockerd goes directly there, via HTTPS (and also via HTTP if included in a
--insecure-registry list), thus completely ignoring the configured mirror.
Yeah. Docker Inc should do it. So should NPM, Inc. Wonder why they don't. 😼
- Basic Docker-for-Mac set-up instructions
- Basic Docker-for-Windows set-up instructions.
- Test and make auth work with quay.io, unfortunately I don't have access to it (hint, hint, quay)
- Hide the mitmproxy building code under a Docker build ARG.
- "Developer Office" proxy scenario, where many developers on a fast LAN share a proxy for bandwidth and speed savings (already works for pulls, but messes up pushes, which developers tend to use a lot)
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent