Setting workdir as subdirectory of mounted directory fails

[ewanbarr]

Description

The documentation for the "-w" flag on docker create/run states that the directory will be created if it does not already exist. This does not appear to work when the requested workdir is a subdirectory of a mounted volume.

The error returned is always:

Starting container process caused "chdir to cwd (\"some directory") set in config.json failed: no such file or directory

This looks like an order of operations issue, where chdir is called ahead of the volume creation. It is not clear if this is a bug/feature/limitation of the system.

Steps to reproduce the issue:

1. Make sure "/tmp/random-directory-name" does not exist on your system
2. docker run -v /tmp/:/tmp/ -w /tmp/random-directory-name/ ubuntu:16.04

Describe the results you received:

This returns the following:

container_linux.go:247: starting container process caused "chdir to cwd (\"/tmp/random-directory-name\") set in config.json failed: no such file or directory"
docker: Error response from daemon: oci runtime error: container_linux.go:247: starting container process caused "chdir to cwd (\"/tmp/random-directory-name\") set in config.json failed: no such file or directory".

Describe the results you expected:

I would expect Docker to first mount the volume, create the subdirectory then set the working directory.

Additional information you deem important (e.g. issue happens only occasionally):
N/A

Output of docker version:

Client:
 Version:      17.03.1-ce
 API version:  1.27
 Go version:   go1.7.5
 Git commit:   c6d412e
 Built:        Tue Mar 28 00:40:02 2017
 OS/Arch:      darwin/amd64

Server:
 Version:      17.03.1-ce
 API version:  1.27 (minimum version 1.12)
 Go version:   go1.7.5
 Git commit:   c6d412e
 Built:        Fri Mar 24 00:00:50 2017
 OS/Arch:      linux/amd64
 Experimental: true

Output of docker info:

Containers: 7
 Running: 0
 Paused: 0
 Stopped: 7
Images: 630
Server Version: 17.03.1-ce
Storage Driver: aufs
 Root Dir: /var/lib/docker/aufs
 Backing Filesystem: extfs
 Dirs: 241
 Dirperm1 Supported: true
Logging Driver: json-file
Cgroup Driver: cgroupfs
Plugins:
 Volume: local
 Network: bridge host ipvlan macvlan null overlay
Swarm: inactive
Runtimes: runc
Default Runtime: runc
Init Binary: docker-init
containerd version: 4ab9917febca54791c5f071a9d1f404867857fcc
runc version: 54296cf40ad8143b62dbcaa1d90e520a2136ddfe
init version: N/A (expected: 949e6facb77383876aeff8a6944dde66b3089574)
Security Options:
 seccomp
  Profile: default
Kernel Version: 4.9.13-moby
Operating System: Alpine Linux v3.5
OSType: linux
Architecture: x86_64
CPUs: 2
Total Memory: 1.952 GiB
Name: moby
ID: 6RTM:GORH:OK2Y:5NR6:UBS6:PRRO:BCQP:SUFM:QUZW:XLV5:I46E:7RBV
Docker Root Dir: /var/lib/docker
Debug Mode (client): false
Debug Mode (server): true
 File Descriptors: 17
 Goroutines: 28
 System Time: 2017-04-15T16:09:06.80608598Z
 EventsListeners: 1
No Proxy: *.local, 169.254/16
Username: mpifrpsr
Registry: https://index.docker.io/v1/
Experimental: true
Insecure Registries:
 127.0.0.0/8
Live Restore Enabled: false

Additional environment details (AWS, VirtualBox, physical, etc.):

Same behaviour also seen on native Linux box running:

Distributor ID:	Debian
Description:	Debian GNU/Linux 8.7 (jessie)
Release:	8.7
Codename:	jessie

[cpuguy83]

While the UX is kind of ugly on this... I'm not sure I agree with the use-case.
The mounted directory is something you passed in and should be your responsibility to make sure everything exists.

[ewanbarr]

That's reasonable. However, whatever the intended behaviour is, it should be captured in the docs.

[ClayAndMore]

i meet the question, as same as your say.

[sorenmat]

I'm experiencing this issue, even if the directory exists.
So if i mount /mydir and then set the workingdir to /mydir/A and A does exist, I still get the error