docker run --rm -it --privileged \
--network host \
-v /var/run/docker.sock:/var/run/docker.sock \
--pid="host" \
-w $(pwd) \
-v $(pwd):$(pwd) \
ghcr.io/srl-labs/clab bash
Once started this way clab deploy works perfectly with the following clab topology:
---
name: test
prefix: ""
mgmt:
ipv4-subnet: 10.0.0.0/16
topology:
kinds:
ceos:
image: arista/ceos:latest
defaults:
kind: ceos
nodes:
ar01:
mgmt-ipv4: 10.0.1.1
ar02:
mgmt-ipv4: 10.0.1.2
links:
- endpoints: ["ar01:eth1_1", "ar02:eth1_1"]
However when replaced with podman run it fails with:
INFO[0000] Containerlab v0.57.2 started
INFO[0000] Parsing & checking topology file: topology.clab.yml
WARN[0000] failed gleaning v4 and/or v6 addresses from bridge via netlink, falling back to docker network inspect data
Error: failed to lookup link "br-2f72e818ccc3": Link not found
Curious if anyone have seen this before and found any solution? I might dig into that myself later, however usually I run cLab on our own devcontainer images and this works just fine. As it works via VSCode, there should be some knob in place...
This is rootful podman with Docker compatibility enabled:
ls -la /var/run | grep docker
lrwxr-xr-x 1 root daemon 60 Sep 20 14:07 docker.sock -> /Users/pa/.local/share/containers/podman/machine/podman.sock
PS: this is on ARM MacBook =wink=
Once started this way
clab deployworks perfectly with the following clab topology:However when replaced with
podman runit fails with:Curious if anyone have seen this before and found any solution? I might dig into that myself later, however usually I run cLab on our own devcontainer images and this works just fine. As it works via VSCode, there should be some knob in place...
This is rootful podman with Docker compatibility enabled:
PS: this is on ARM MacBook =wink=