You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would be also good if talos would automatically create a larger RAM allocation when creating it with talosctl cluster create
Also edit that Cgroupv2 should be used (as one could go into not so nice errors if one for example installs coder.com:
process: failed to write "400000": write /sys/fs/cgroup/cpu/kubepods/burstable/podf174f9b0-c1ad-4a26-8254-af003f321206/dev/cpu.cfs_quota_us: invalid argument: unknown
To solve this do:
1. Go to %USERPROFILE%
2. Create .wslconfig
3. Add to .wslconfig
[wsl2]
kernelCommandLine = cgroup_no_v1=all systemd.unified_cgroup_hierarchy=1
memory=10GB
localhostForwarding=true
4. Showdown wsl
Cmd: wsl --shutdown
5. Start ubuntu on win 11
Cmd: mount | grep cgroup
- Should only have one entry with cgroupv2
6. Start docker
Add to Docker Engine:
"exec-opts": [
"native.cgroupdriver=cgroupfs"
],
"experimental": false,
"features": {
"buildkit": true
}
Run in cmd: docker info
Has to be cgroupv2
Logs
Environment
Talos version: [talosctl version --nodes <problematic nodes>]
Kubernetes version: [kubectl version --short]
Platform:
The text was updated successfully, but these errors were encountered:
Bug Report
Description
Update https://www.talos.dev/v1.7/talos-guides/install/local-platforms/docker/ documentation.
As docker on win 11 is only having default 2GB of RAM allocated many people will get OOM error.
This is not a good start for people who try talos for the first time.
Add to the documentation that if Out of memory OOM node failure happens, one has to increase the RAM in docker:
docker update --memory 10g --memory-swap 12g talos-default-worker-1
It would be also good if talos would automatically create a larger RAM allocation when creating it with talosctl cluster create
Also edit that Cgroupv2 should be used (as one could go into not so nice errors if one for example installs coder.com:
process: failed to write "400000": write /sys/fs/cgroup/cpu/kubepods/burstable/podf174f9b0-c1ad-4a26-8254-af003f321206/dev/cpu.cfs_quota_us: invalid argument: unknown
To solve this do:
1. Go to %USERPROFILE%
2. Create .wslconfig
3. Add to .wslconfig
[wsl2]
kernelCommandLine = cgroup_no_v1=all systemd.unified_cgroup_hierarchy=1
memory=10GB
localhostForwarding=true
4. Showdown wsl
Cmd: wsl --shutdown
5. Start ubuntu on win 11
Cmd: mount | grep cgroup
- Should only have one entry with cgroupv2
6. Start docker
Add to Docker Engine:
"exec-opts": [
"native.cgroupdriver=cgroupfs"
],
"experimental": false,
"features": {
"buildkit": true
}
Has to be cgroupv2
Logs
Environment
talosctl version --nodes <problematic nodes>
]kubectl version --short
]The text was updated successfully, but these errors were encountered: