compose: Error when trying to run docker-compose up. "oci runtime error: container_linux.go:247..."
When trying to launch a built container with docker-compose up I’m getting an error:
ERROR: for app Cannot start service app: invalid header field value "oci runtime error: container_linux.go:247: starting container process caused \"exec: \\\"script/docker-entrypoint.sh\\\": stat script/docker-entrypoint.sh: no such file or directory\"\n"
ERROR: compose.cli.main.main: Encountered errors while bringing up the project.
I’ve tried reinstalling docker, docker-compose, virtualbox, rebuilding containers, recreating machines. Downgrading almost all of those things to previous versions. Literally retrying the whole docker instalation flow.
The problem occurs only in docker-compose up. Everything works fine when I use docker run ....
The important thing is the whole setup works correctly on my OSX installation. Unfortunately my Xenial Ubuntu doesn’t want to cooperate.
I’ve checked the issues archive and couldn’t find an answer that would help me fix the problem.
Here are my whole setup configs:
docker-compose.yml:
version: "2"
services:
app:
build: .
volumes:
- .:/app
depends_on:
- cache-redis
links:
- cache-redis
nginx:
image: nginx:1.11.1-alpine
depends_on:
- app
links:
- app
ports:
- "80:80"
volumes:
- ./config/nginx/app.conf:/etc/nginx/conf.d/default.conf
cache-redis:
image: redis:3.2.1-alpine
volumes:
- cache-redis:/var/lib/cache-redis/data
volumes:
cache-redis:
driver: local
Dockerfile
FROM jruby:9.1.2.0-jre-alpine
RUN apk --update --no-cache add build-base less git openssh
RUN mkdir -p /app
WORKDIR /app
COPY Gemfile Gemfile.lock ./
RUN gem install bundler && bundle install
COPY . ./
CMD ["script/docker-entrypoint.sh"]
script/docker-entrypoint.sh
#!/bin/sh
rm -f tmp/pids/server.pid
padrino s -h 0.0.0.0
docker-compose -v:
docker-compose version 1.8.1, build 878cff1
docker version:
Client:
Version: 1.12.2
API version: 1.24
Go version: go1.6.3
Git commit: bb80604
Built: Tue Oct 11 18:29:41 2016
OS/Arch: linux/amd64
Server:
Version: 1.12.2
API version: 1.24
Go version: go1.6.3
Git commit: bb80604
Built: Tue Oct 11 17:00:50 2016
OS/Arch: linux/amd64
docker info:
Containers: 6
Running: 1
Paused: 0
Stopped: 5
Images: 15
Server Version: 1.12.2
Storage Driver: aufs
Root Dir: /mnt/sda1/var/lib/docker/aufs
Backing Filesystem: extfs
Dirs: 37
Dirperm1 Supported: true
Logging Driver: json-file
Cgroup Driver: cgroupfs
Plugins:
Volume: local
Network: bridge host null overlay
Swarm: inactive
Runtimes: runc
Default Runtime: runc
Security Options: seccomp
Kernel Version: 4.4.24-boot2docker
Operating System: Boot2Docker 1.12.2 (TCL 7.2); HEAD : 9d8e41b - Tue Oct 11 23:40:08 UTC 2016
OSType: linux
Architecture: x86_64
CPUs: 1
Total Memory: 995.9 MiB
Name: default
ID: G442:OWMQ:BPXD:7MK5:HM7J:R7PO:DNBP:ZSKI:HJH4:OCE4:UX36:M2PO
Docker Root Dir: /mnt/sda1/var/lib/docker
Debug Mode (client): false
Debug Mode (server): true
File Descriptors: 21
Goroutines: 38
System Time: 2016-10-18T09:18:30.024046419Z
EventsListeners: 0
Registry: https://index.docker.io/v1/
Labels:
provider=virtualbox
Insecure Registries:
127.0.0.0/8
About this issue
- Original URL
- State: closed
- Created 8 years ago
- Reactions: 23
- Comments: 92
you gotta make the docker-entrypoint.sh an executable before building the image:
otherwise it cant be executed and you get the permission error.
I’ve managed to fix it. I had wrong permissions and needed to run
docker-composeassudo.I am getting this error too, and
sudofixed it. However, there should be an alternative solution to this issue.I’m getting a similar error and unfortunately sudo doesn’t fix this one.
click on icon tray Docker > Settings… > Shared Drivers and select checkbox Shared: C and D Maybe it`s help you!
I had similar issue with my images and found this thread. In my case, It turned out that mounting the volumes is the source of confusion.
In your case, please make sure that app.conf is an existing file.
./config/nginx/app.conf:/etc/nginx/conf.d/default.confThis will mount your local app.conf inside of the docker container. If it does not exist on your local machine, docker will create locally an “app.conf” directory, which is not what you would usually expect in this particular case.
I hope it will help you.
This issue shouldn’t be closed - it still happens, and sudo isn’t really a good solution when needing to point docker-compose at a remote machine for deployment.
This is with docker-compose 1.9.0.
Interestingly, if I ssh into the machine, docker-compose up -d completes without an error. So, for me, this only happens when running remotely.
Odd thing is, the mount in the error message is for a local path (in bold below) - referred to relatively in my docker-compose, so perhaps this is a relative/absolute path problem with docker-compose.
ERROR: for nginx Cannot start service nginx: invalid header field value “oci runtime error: container_linux.go:247: starting container process caused "process_linux.go:359: container init caused \"rootfs_linux.go:53: mounting \\\\ "/data/www/aq/docker/nginx/default.conf \\\" to rootfs \\\"/var /lib/docker/aufs/mnt/8e3b80c25d85b70b78ea479a5f68373e236fdd3dafd62b241495907ace62da2f\\\" at \\\"/var/lib/docker/aufs/mnt /8e3b80c25d85b70b78ea479a5f68373e236fdd3dafd62b241495907ace62da2f/etc/nginx/conf.d /default.conf\\\" caused \\\"not a directory\\\"\""\n”
This is due to a line in docker-compose.yml trying to point nginx’s configuration back to the local filesystem:
This should resolve to /home/root/docker on the remote, and /data/www/aq/docker locally. Instead, docker-compose seems to be trying to use the local path on the remote.
Happy to provide any other details/logs.
I’ve noticed having a CMD line in a dockerfile and a -command parameter in a docker-compose file will also trigger this error.
issue for me was that I recently changed windows password, Going into Docker settings -> shared drives, un-selecting my drive, apply, and then re-selecting (apply) fixed it.
I am so sorry to add to this issue after it has been opened:
But I found a way around this. Simply rebuild the image by running:
And then start a container using the newly built image:
That’s all.
I hope this helps
On Windows 10 I was encountering the same issue with Docker for Windows. This started happening and I couldn’t figure out why until I realized that I changed my OS credentials earlier in the day! I went into Docker Settings --> shared drives and clicked ‘Reset my credentials’. I was then asked for my new credentials and everything started up fine after that.
On windows 10 and just encountered this bug. Going into Docker settings -> shared drives and then unselecting my drive, apply, and then re-selecting (apply) seems to have fixed it. Thanks @artem-gavrylenko
If you’re using MacOS, you should make your host’s docker-entrypoint.sh executable too, not just on your Dockerfile. Got this from a tip, https://github.com/composer/docker/issues/7
Found it, I was not assigning options in the right order:
fails while:
works.
Those parameter order restrictions are quite of annoying. Never has so many issues with any Unix executable…
Worked for me, thank you! 😄
@michalwarda @jakerobers I am experiencing this issue too. I think you should reopen the issue.
I am using a pretty vanilla install of Docker for Mac, and I can’t remember doing anything that would require this permission level on the
docker-composecommand.@chielsen Seems like it may happen at least after Docker update. I solved this problem for me with restarting Docker daemon.
I had a similar error today. And after almost an hour of debugging and testing several of the provided solutions, the only thing that solved the problem was, a system Restart. True story.
Tu é pica mesmo
For what it is worth:
sudo docker container prunefixed the issue for me.It helped me. Thank you.
“volumes” binds directories… not files.
@imarchuang bind the “config” directory => elk/elasticsearch/config with /usr/share/elasticsearch/config
@chielsen bind “php-fpm” direcotry with /etc/php/7.1/fpm/conf.d directory
Instead of running as sudo simply run the command
chmod a+x /path/to/your/script.shto give permission to execute. This is much better than using sudo each time.Iam also getting a similar but little different, Please help
docker-compose up Starting wapdocker_wordpress_1 … error
ERROR: for wapdocker_wordpress_1 Cannot start service wordpress: b’OCI runtime create failed: container_linux.go:345: starting container process caused “exec: \”/bin/sh\“: stat /bin/sh: no such file or directory”: unknown’
ERROR: for wordpress Cannot start service wordpress: b’OCI runtime create failed: container_linux.go:345: starting container process caused “exec: \”/bin/sh\“: stat /bin/sh: no such file or directory”: unknown’ ERROR: Encountered errors while bringing up the project.
Another issue can be when you bind a volume attached to the local directory and your entry point is located it the directory that was replaced in the container. Thus, docker tries to execute the entry point and doesn’t find it or gets a permissions denial.
Just to add in to the other comments here by several people, this happened to me recently and I can confirm. I had recently changed the password, computer was restarted and
docker-composewould not bring up the containers. Unchecking and rechecking the Shared Drive, followed by entering the updated password got everything up and running.I was using alpine-based images and changed /bin/bash to /bin/sh and it worked ok 😃
In my case ,
yum install libseccomp libseccomp-devel.I had this error a very stupid thing:
I was doing this
Insetead of this:
I had this happen when trying to run a docker through docker toolbox in windows on WSL. For what it’s worth to anyone here, my problem was solved by making sure that i was accessing things through the windows rather than the linux file-system:
docker run -it --rm --name lispy -v “$PWD”:/Lis_Py python:3 python Lis_Py/lis.py || docker start -ia lispy
# on WSL with docker running in windows, $PWD will not work and must be set to the same as $PWD from docker quick-start, # viz. docker run -it --rm --name lispy -v /c/Users/firez/Documents/Development/lis.py:/Lis_Py python:3 python Lis_Py/lis.py || docker start -ia lispy
You either mount a specific file, or you mount a whole folder. You cannot mount a single file into a folder!
As taken from the link provided by @chielsen you can do this:
which maps the users local
.bash_historyfile into the container with the specified filename.This issue happens on my native docker on Mac too… no idea how to resolve it!
below are some info about my docker: