Question

我正在尝试使用Circle Ci构建持续集成流程。当我将代码推送到github时，流程开始，但它会因错误而停止：

    Build-agent version 0.0.4071-34fe4dc (2017-09-05T01:33:40+0000)
Starting container micheleminno/elasticsearch:latest
  image cache not found on this host, downloading micheleminno/elasticsearch:latest
latest: Pulling from micheleminno/elasticsearch
42a816b9ad19: Pull complete
5dabff016c33: Pull complete
e22082c539e8: Pull complete
57845aa0da23: Pull complete
3da62a3688f0: Pull complete
bce992ca3d80: Pull complete
2734c785c0d8: Pull complete
08cbf6dbba97: Pull complete
750b60032fcc: Pull complete
Digest: sha256:f07025f1da82ea87e70cac194929953a245ae4560086471c90b1756693246ec4
Status: Downloaded newer image for micheleminno/elasticsearch:latest
  using image micheleminno/elasticsearch@sha256:f07025f1da82ea87e70cac194929953a245ae4560086471c90b1756693246ec4
Starting container mysql:5.5
  using image mysql@sha256:7674d74d9f010b1500c1ae38f7af90014145ee3c92ddd65dd2e1edfd8c61270f
Starting container micheleminno/db-migrations
  image cache not found on this host, downloading micheleminno/db-migrations
latest: Pulling from micheleminno/db-migrations
c89e59f636b6: Pull complete
cfb692fa93ee: Pull complete
d607b85c45a1: Pull complete
0150b9c56763: Pull complete

CircleCI was unable to start the container because of a userns remapping failure in Docker.

This typically means the image contains files with UID/GID values that are higher than Docr and CircleCI can use.

Checkout our docs https://circleci.com/docs/2.0/high-uid-error/ to learn how to fix this problem.

Original error: failed to register layer: Error processing tar file(exit status 1): Container ID 5924152 cannot be mapped to a host ID

我跟着link并按照它的建议行事，但我找不到任何带有该ID的内容。

这是我的配置圈yml文件：

version: 2

jobs:
  build:
    docker:
      - image: micheleminno/elasticsearch:latest
      - image: mysql:5.5
        environment:
          MYSQL_ROOT_PASSWORD: password
          MYSQL_DATABASE: real-affinities-test
          MYSQL_USER: development
          MYSQL_PASSWORD: development
      - image: micheleminno/db-migrations
        environment:
          - NODE_ENV: development

    working_directory: ~/

    steps:
      - checkout
      - run:
          name: Build real-affinities and run tests
          command: docker-compose -f docker-compose.test.yml -p ci up --build

更新09/09/2017：

我遵循@FelicianoTech建议直接在Dockerfile中更改文件的所有权，更新导致高uids问题的项目的Dockerfile：

FROM node:8.1.4-alpine

WORKDIR /app

# add package.json and run npm install before adding the rest of the files
# this way, you only run npm install when package.json changes
ADD package.json /app
RUN npm install

# add the rest of the files
ADD . /app

# change ownership recursively of all files
RUN chown root:root . -R

CMD ["./node_modules/.bin/knex", "migrate:latest"]

但问题仍然存在。

Answer 1

问题在于第三个Docker镜像micheleminno/db-migrations。如果在本地运行映像，请运行以下命令：

ls -lah /app/node_modules/

您将看到多个目录的用户ID高于65534.这就是问题所在。将这些ID更改为更低的值以解决此问题。

Docker和CircleCI 2.0的高uid错误问题

1 个答案: