travis-ci里是不是空间不足挂了？ about gcr.io_mirror

[anjia0532]

https://cloudstatus.eu/status/travisci
13小时前发生故障，11小时前已经转移恢复了。现在正常。不过，目前剩下几个单镜像贼大的（2000多个，每个过G，导致mirror有点慢）

[anjia0532]

待同步的

    704 cloud-datalab/datalab
    692 cloud-datalab/datalab-gateway
    719 cloud-datalab/datalab-gpu
     75 kubeflow-images-public/tensorflow-1.4.1-notebook-cpu
    124 kubeflow-images-public/tensorflow-1.4.1-notebook-gpu
      4 kubeflow-images-public/tensorflow-1.5.1-notebook-cpu
    124 kubeflow-images-public/tensorflow-1.5.1-notebook-gpu
     76 kubeflow-images-public/tensorflow-1.6.0-notebook-cpu
     45 kubeflow-images-public/tensorflow-1.6.0-notebook-gpu
     28 kubeflow-images-public/tensorflow-1.7.0-notebook-gpu
     57 kubeflow-images-public/tensorflow-1.8.0-notebook-cpu
     35 kubeflow-images-public/tensorflow-1.8.0-notebook-gpu
    115 kubeflow-images-public/tf-model-server-gpu

[zavatar]

k8s.gcr.io/kube-apiserver:v1.12.2
这个就没同步啊

[anjia0532]

我的同步逻辑是，如果repo中，有待同步的，等同步完后，再去比最新的，我手动调整了一下优先级，现在可以了

docker pull anjia0532/google-containers.kube-apiserver:v1.12.2
# k8s.gcr.io/{image}/{tag} <==> gcr.io/google-containers/{image}/{tag} <==> anjia0532/google-containers.{image}/{tag}

https://github.com/anjia0532/gcr.io_mirror#syntax