meamidos / dcind Goto Github PK

After the upgrade we're running into the following issue:

11:11:17
Starting Docker...
11:11:17
test.sh: line 101: declare: not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found
11:11:17
bash: try_start: command not found

Have you tried use task cache?

Was thinking it may be useful when building images.

For anyone looking for an updated version of this image I started publishing a newer version over at https://github.com/taylorsilva/dcind

Hi, I have a Concourse build task that looks like

#!/bin/bash

source /docker-lib.sh
start_docker

docker load -i image1.tar
docker load -i image2.tar
docker load -i image3.tar

docker-compose up

The problem is that some of the images are quite large and this makes the task slow to run.

I thought that it might be possible to cache /scratch/docker in my build configuration:

caches: [{path: /scratch/docker}]

But it seems that this doesn't have any effect - docker images returns an empty list before I run docker load.

How can I cache the images loaded in previous builds to speed up my build task?

The current build is breaking for me. Please add tags for new releases on dockerhub so I can revert to an older version while the issue is fixed. (I'm using digest to revert to an older cached version for now)

run source /docker-lib.sh && start_docker

but always show waiting for docker to come up

where can I find some error logs and get why?

when running my integration testsuite through either with preconfigured integration test job in my integration.yml file via docker-compose -f code-repository/integration.yml run tests or docker-compose -f code-repository/integration.yml up -d and directly after docker-compose -f code-repository/integration.yml run --rm -w /var/www/html/public php php scripts/run-tests.sh --php /usr/local/bin/php --url http://nginx --verbose --color --class DownloadRestrictionsTestcase I can see the docker-compose output "starting containers" with their done flags but concourse won't wait for further output and directly proceeds to the next step, in this case docker-compose -f code-repository/integration.yml down

Could you attach a license file to this repo?

sh: declare: not found

Hello,

I'm setting up a pipeline in concourse that tests ansible roles. One of my jobs spins up a full (systemd enabled) ubuntu1804 image. However, one of the ansible steps for systemd fails with the following error:

Cannot determine cgroup we are running in: No such file or directory
Failed to initialize manager: No such file or director

I added the following to my job config:

$ source /docker-lib.sh
$ start_docker
$ ls /sys/fs/cgroup
blkio             cpu               cpu,cpuacct       cpuacct           cpuset            devices           freezer           hugetlb           memory            net_cls           net_cls,net_prio  net_prio          perf_event        pids              rdma

as we can see, /sys/fs/cgroup/systemd does not exist, so it makes sense that the "/sys/fs/cgroup:/sys/fs/cgroup:ro" mount from ansible-role-tester doesn't pull in systemd support.

I found a suggestion here that adds a manual systemd folder creation, and this seems to resolve the issue. My build now works, and systemd is supported.

mkdir /sys/fs/cgroup/systemd
mount -t cgroup -o none,name=systemd cgroup /sys/fs/cgroup/systemd

My question to @meAmidos is thus: is this the best workaround, and if can we add it to docker-lib.sh? I'm still learning about cgroups, so I'm not sure what that script does.

Hi Team,

I am trying to use the amidos/dcind image on arm64 platform but it seems it is not available for arm64.

I have successfully built the image using the command docker build -t image_name . on the arm64 platform without making any changes in the Dockerfile.

I have used Github actions to build and push the image for both the platforms.

Commit Link - odidev@b071913

Github actions link - https://github.com/odidev/dcind/runs/2360093417?check_suite_focus=true

Docker Hub Link - https://hub.docker.com/r/odidev/dcind/tags?page=1&ordering=last_updated

Do you have any plans on releasing arm64 images?

It will be very helpful if arm64 image is available. If interested, I will raise a PR.

I'm trying to use dcind to run tests against a golang project that pulls and pushes docker images. The current version of dcind (2.1.0), uses alpine 3.10. That version of alpine supports golang 1.12. However, my project requires golang 1.16.

Would it be possible to update to the latest alpine? Or possibly support multiple versions (I'm not sure how big of an ask this is).

Any chance you could update to a newer version of Docker? This (now resolved) Docker bug is currently blocking my use of dcind: docker/docker-py#1842

These two lines were removed from the example in the README:

source /docker-lib.sh
start_docker

But without them, any calls to docker functions cause the following error:

Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?

Since the entrypoint.sh script isn't used when a concourse task is run, these lines need to be in the task script.

Ubuntu perhaps?