RA489
commented
3 years ago /kind support
Closed jakoberpf closed 2 years ago
RA489
commented
3 years ago /kind support
jakoberpf
commented
2 years ago Does anyone have an idea what could be the issue?
tspearconquest
commented
2 years ago I started seeing this recently on my Mac (not an M1) but I am using the flag --kubernetes-version=v1.18.19 and it works fine if I use the flag --kubernetes-version=v1.19.13
It was working fine, I took a week off for Christmas, come back and it's not working now. I did some googling and found that kubernetes v1.19 was the first to support cgroups v2. This seems to be related to the error but I'm not sure what has changed in minikube that would cause this. Interestingly, your logs show using kubernetes v1.22.3 which is the latest k8s release, and you are still having this issue, but yet when I change the kubernetes version from the older release to the newer (but not latest) release, it works fine for me. That tells me that my root cause might be different from yours.
jakoberpf
commented
2 years ago @tspearconquest Interesting that you saw the same issue and where able to fix it with and older kubernetes version. I tried the same, but am still seeing the same error.
tspearconquest
commented
2 years ago Which MacOS version are you running? Is there anything in your networking configuration that you manually configured on the Mac?
Does the cluster still work, for example do the pods start up and can you still access them over the network in any way?
taeuuoo
commented
2 years ago If you just wanna run anyway, make docker network manually(I used subnet 192.168.9.0/24).
Then minikube start might work.
jakoberpf
commented
2 years ago @tspearconquest I am on 12.0.1 and my network configuration is quite basic, as just DHCP on private WIFI. I can access the cluster ressources via portforwarding, but not via ingresses
@taeuuoo Your previous commen actually did the trick. Running docker network create --subnet 192.168.9.0/24 --driver bridge minikube before minikube start work fixed my error.
Is this in general a problem of minikube (not beeing able to create this network by itself) or could it be related to the M1 arch or my (default) network conifguration?
klaases
commented
2 years ago From what I can tell, the the network was already being used by docker, which would be related to your network configuration. For minikube, subnet creation should iterate up to a point, approx. 20 tries.
For additional debugging purposes, provide additional details from docker.
Please share the output of docker network ls
And also share the output of docker inspect [name of the network]
That way we can see what networks docker is using and troubleshoot further.
klaases
commented
2 years ago Hi @jakoberpf – is this issue still occurring? Are additional details available? If so, please feel free to re-open the issue by commenting with /reopen. This issue will be closed as additional information was unavailable and some time has passed.
Additional information that may be helpful:
Whether the issue occurs with the latest minikube release
The exact minikube start command line used
Attach the full output of minikube logs, run minikube logs --file=logs.txt to create a log file
Thank you for sharing your experience!
yama6a
commented
2 weeks ago I now this was closed a while ago, but since I just ran into this, I thought I might share my solution to posterity:
check your networks: docker network ls
Minikube runs by default with the bridge driver (at least when running as a 1-node-cluster), so find the network ID for the bridge network. My output looks like this:
$ docker network ls
NETWORK ID NAME DRIVER SCOPE
a14a99c24fff bridge bridge local # <-- THIS ONE
dcec7ebba4dd host host local
1de7b2d366aa none null localinspect the bridge network to see the subnet: docker inspect a14a99c24fff | grep subnet -iC3
My output looks like this:
$ docker inspect a14a99c24fff | grep subnet -iC3
"Options": null,
"Config": [
{
"Subnet": "172.17.0.0/16", # <-- THIS ONE
"Gateway": "172.17.0.1"
}
]Use a portion of that subnet as your minikube subnet when starting minikube (I'm using the second half of the 172.17.0.0/16 CIDR block, i.e. : 172.17.128.0/17 which holds 32768 IP addresses (surely ought to be enough).
minikube start \
--cpus=8 \
--delete-on-failure=true \
--driver=docker \
--kubernetes-version=v1.31.0 \
--memory=8192 \
--subnet=172.17.128.0/17 # <-- THIS ONE
Steps to reproduce the issue: On M1 Pro Macbook Pro 2021 and docker installed, run
minikube start.I did some googling but came up empty for this issue.
Full output of failed command if not
minikube start: