Description of problem: interface cni0 using subnet 10.88.0.0/16 that is the same network as in our organisation. There is no way to configure cni subnet during installation. 1. First we can check where all the configuration for cni0 bridge is present: ``` # grep -ril cni0 /etc /etc/cni/net.d/100-crio-bridge.conf /etc/cni/net.d/87-podman-bridge.conflist # # cat /etc/cni/net.d/100-crio-bridge.conf { "cniVersion": "0.3.0", "name": "crio-bridge", "type": "bridge", "bridge": "cni0", "isGateway": true, "ipMasq": true, "ipam": { "type": "host-local", "subnet": "10.88.0.0/16", "routes": [ { "dst": "0.0.0.0/0" } ] } } # # cat /etc/cni/net.d/87-podman-bridge.conflist { "cniVersion": "0.3.0", "name": "podman", "plugins": [ { "type": "bridge", "bridge": "cni0", "isGateway": true, "ipMasq": true, "ipam": { "type": "host-local", "subnet": "10.88.0.0/16", "routes": [ { "dst": "0.0.0.0/0" } ] } }, { "type": "portmap", "capabilities": { "portMappings": true } } ] } # ``` 2. Next, we can check the source of these files: ``` # grep -ril cni0 /etc | xargs rpm -qf cri-o-1.13.9-1.rhaos4.1.gitd70609a.el8.x86_64 podman-1.0.2-1.dev.git96ccc2e.el8.x86_64 # ``` It looks like, this default configuration comes directly from the podman and cri-o rpms installed on the VM. This means that there is no direct control of this subnet through the ignition-config or any other installation file of Openshift. Version-Release number of the following components: Openshift 4.1 How reproducible: Steps to Reproduce: 1. Install Openshift. 2. 3. Actual results: No way to configure it in ignition file. Expected results: Should be able to change it via ignition file. Additional info: Please attach logs from ansible-playbook with the -vvv flag
Hmm. This isn't a SDN question per se, but it's definitely interesting to think about. We only directly execute containers via podman for bootstrap; everything else uses the configured cidrs as part of kubernetes. Honestly there will probably *always* be IP conflicts, no matter which range you pick. So maybe we should just execute the bootstrap process host-network and be done with it. cc Abhinav. thoughts?
Filed https://github.com/openshift/installer/pull/2001 to fix this.
Verified this bug on 4.2.0-0.nightly-2019-08-08-103722 Check the boostrap containers are using hostnetwork #ps -ef | grep podman root 5439 1542 4 02:13 ? 00:00:01 podman run --quiet --net=host --rm --volume /var/opt/openshift:/assets:z --volume /etc/kubernetes:/etc/kubernetes:z --network=host quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:a4c2814e36368b7b9df3ebc13e655d98bce51472ace21fccba6d9b6225c64b28 start --tear-down-early=false --asset-dir=/assets --required-pods=openshift-kube-apiserver/kube-apiserver,openshift-kube-scheduler/openshift-kube-scheduler,openshift-kube-controller-manager/kube-controller-manager,openshift-cluster-version/cluster-version-operator
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:2922