Sync: Synchronize the synchronization rate of image tags in the list regularly. As long as the tags exist, they are synchronized

Due to the recent sharp increase in usage, in order to ensure the availability of most

• Whitelist&flow restriction&degraded public information #2328
• If you have any questions, please consult #4183

Many images are abroad. For example, gcr. Domestic downloads are slow and need to be accelerated.

• A simple and effective method can speed up these packages. Concise name mapping
• Easy to add, add new packages, no need to modify the code.
• Stable and reliable, updated in real time. Check the synchronization every day.
• This item is only a mirror of the source image warehouse (Registry)
  • All hashes (sha256) are consistent with the source (lazy loading mechanism).
  • Due to the existence of cache, there may be a delay of 1 hour.
  • If it has not been updated for more than one hour, it is estimated that the international bandwidth is suspended.
• The blob will be cached on third-party object storage
  • Currently, no detection has been performed on the content. It is planned to add detection.

 docker run -d -P m.daocloud.io/docker.io/library/nginx

Add prefix (Recommended method). For example:

 docker.io/library/busybox | V m.daocloud.io/docker.io/library/busybox

Or the supported image warehouse Prefix replacement Can be used. For example:

 docker.io/library/busybox | V docker.m.daocloud.io/library/busybox

!!!!!!!!!!!!!!!!!!!!!!!! Strongly recommended!!!!!!!!!!!!!!!!!!!!!!!!

You can Mirror Synchronization Issue Template Create an Issue, and a robot will help you proactively synchronize the specified image

The original image under will continue to be the same. You need to restart the docker and then pull again to access the synchronized cache. Therefore, it is recommended that you synchronize once before trying to pull The latest tag, which often changes content, needs to be resynchronized after change

Only recommended for images smaller than 20MiB This channel is mainly used to keep the image up to date. Each pull will check whether it is synchronized with the upstream

If you see that there is no progress, it is due to limited bandwidth. As long as there are several people under the command, large files will block subsequent downloads. You can try Single mirror synchronization

Even without synchronization Direct pull , the first pull will be slower than the synchronized one

All lazy loads bandwidth

• International bandwidth 6 * 30 Mbps

It is strongly recommended to use Single mirror synchronization

mirror.txt

If you want to add a new one, just ask PR to modify it. for example PR#1 , and please sort before PR submission: ./hack/fmt.sh mirror.txt

It is recommended to add prefix

Prefix replacement registry rules, which are manually configured, need to provide Issue

• Install kubeadm by accelerating

 kubeadm config images pull --image-repository k8s-gcr.m.daocloud.io

• By accelerating the installation of kind

 kind create cluster --name kind --image m.daocloud.io/docker.io/kindest/node:v1.22.1

• By accelerating the deployment of applications (Ingress is taken as an example here)

 wget -O image-filter.sh  https://github.com/DaoCloud/public-image-mirror/raw/main/hack/image-filter.sh && chmod +x image-filter.sh wget -O deploy.yaml  https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.1.0/deploy/static/provider/baremetal/deploy.yaml cat ./deploy.yaml | ./image-filter.sh | kubectl apply -f -

• Docker acceleration

Add to /etc/docker/daemon.json

 { "registry-mirrors" : [ " https://docker.m.daocloud.io " ] }

• Image acceleration: https://github.com/DaoCloud/public-image-mirror
• Binary file acceleration: https://github.com/DaoCloud/public-binary-files-mirror
• Helm acceleration: https://github.com/DaoCloud/public-helm-charts-mirror

Made with contrib.rocks .