WSL: WSL2: docker: Error response from daemon: cgroups: cannot find cgroup mount destination: unknown.
Please fill out the below information:
-
Your Windows build number: 18912
-
What you’re doing and what’s happening:
Intermittently, Docker CE quits working with a cgroup error. If I run wsl --shutdown then restart Ubuntu, it will work again.
docker run -it --rm -v /home/user/aks-engine:/go/src/github.com/Azure/aks-engine -w /go/src/github.com/Azure/aks-engine quay.io/deis/go-dev:v1.22.1 bash
Unable to find image 'quay.io/deis/go-dev:v1.22.1' locally
v1.22.1: Pulling from deis/go-dev
9ff7e2e5f967: Pull complete 59856638ac9f: Pull complete 6f317d6d954b: Pull complete a9dde5e2a643: Pull complete 692b28bf9c4d: Pull complete 5075eab28b88: Pull complete Digest: sha256:e872262e17cbc9b562a0d98d7abd75c6ee35eee782f13ba6e0ad0ec5e0b92753
Status: Downloaded newer image for quay.io/deis/go-dev:v1.22.1
docker: Error response from daemon: cgroups: cannot find cgroup mount destination: unknown.
Makefile:55: recipe for target 'dev' failed
make: *** [dev] Error 125
- What’s wrong / what should be happening instead:
This should work consistently.
- Docker Version
$ docker version
Client:
Version: 18.09.6
API version: 1.39
Go version: go1.10.8
Git commit: 481bc77
Built: Sat May 4 02:35:57 2019
OS/Arch: linux/amd64
Experimental: false
Server: Docker Engine - Community
Engine:
Version: 18.09.6
API version: 1.39 (minimum version 1.12)
Go version: go1.10.8
Git commit: 481bc77
Built: Sat May 4 01:59:36 2019
OS/Arch: linux/amd64
Experimental: false
About this issue
- Original URL
- State: open
- Created 7 years ago
- Reactions: 37
- Comments: 22
Commits related to this issue
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
- Add bootstrap-sel.sh and selrc Set up some housekeeping to run on every domain start. Only do the minimal still for a usable SEL environment * install additional packages * install the current /etc/... — committed to dtroyer-salad/wsl-ansible by dtroyer-salad 3 years ago
I encountered this too - the “fix” was running this in WSL:
Not really a viable fix, but it made my command work.
My environment:
The command I was running that caused it to error (interestingly, on my system it seemed to be a system load based error, only under high demand did I see this.
Where
cluster.yamlwas this (essentially creating 6 k8s nodes):The “fix”:
Works for me on Gentoo Linux with OpenRC insted of
systemd. Seems the problem is common for all systems withoutsystemd.For anyone else who found their way to this issue, I just thought it would be worth mentioning that Docker Desktop (or at least the beta version) has quite a nice integration with WSL. You install it, then there is an option under “Resources > WSL Integration” where you can select which distros you want to use it with. For me the default was already selected.
Thought it would be worth mentioning, as it seems like less trouble than trying than getting the daemon running properly within your WSL instance. I guess the downside is you end up running multiple VMs instead of one, but I’ll take that for a pain-free setup.
The real MVP.
Windows 2004 (19041.12) Docker version 19.03.5, build 633a0ea838 Reproduced by running
@mylesagray 's fix works for me.
this worked for me too
I also had this issue in the past. Usually the fix mentioned works fine:
However today, for some reason it didn’t. The fix for me today was to also restart docker service:
my wrong information
The solved solution
successfully。
The reason i guess that in the “apt-get install docker-ce=17.03.1~ce-0~ubuntu-xenial” process, docker-ce-cli is unistall,docker-ce-cli is have relationship with daemon(in my install process,daemon has so many problems.)
This guy is a legend
Windows 2004 (19041.153( Docker Version 19.03.8 build afacb8b7f0
Same as @chris-han and also solved with @mylesagray fix.