I am cloning this issue under the OVN Kubernetes component to see if it's possible changing https://github.com/openshift/machine-config-operator/blob/master/templates/common/_base/files/configure-ovs-network.yaml to use different name for OVS bridge/port/interface according to https://bugzilla.redhat.com/show_bug.cgi?id=1913248#c4 +++ This bug was initially created as a clone of Bug #1913248 +++ Description of problem: I am trying to create the following NodeNetworkConfigurationPolicy: --- apiVersion: nmstate.io/v1alpha1 kind: NodeNetworkConfigurationPolicy metadata: name: bigip-bridges spec: nodeSelector: node-role.kubernetes.io/worker: "" desiredState: interfaces: - name: bond0.375 type: vlan state: up vlan: base-iface: bond0 id: 375 - name: bigip-mgmt description: Linux bridge with bond0 vlan375 as a port! type: linux-bridge state: up bridge: options: stp: enabled: false port: - name: bond0.375 - name: bond0.376 type: vlan state: up vlan: base-iface: bond0 id: 376 - name: bigip-ha description: Linux bridge with bond0 vlan376 as a port! type: linux-bridge state: up bridge: options: stp: enabled: false port: - name: bond0.376 which results in the following error message: {"level":"error","ts":1609931516.2395382,"logger":"controllers.NodeNetworkConfigurationPolicy","msg":"Rolling back network configuration, manual intervention needed: ","nodenetworkconfigurationpolicy":"/bigip-bridges","error":"error reconciling NodeNetworkConfigurationPolicy at desired state apply: , failed to execute nmstatectl set --no-commit --timeout 480: 'exit status 1' '' 'Traceback (most recent call last):\n File \"/usr/bin/nmstatectl\", line 11, in <module>\n load_entry_point('nmstate==0.3.4', 'console_scripts', 'nmstatectl')()\n File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 67, in main\n return args.func(args)\n File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 267, in apply\n args.save_to_disk,\n File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 289, in apply_state\n save_to_disk=save_to_disk,\n File \"/usr/lib/python3.6/site-packages/libnmstate/netapplier.py\", line 69, in apply\n net_state = NetState(desired_state, current_state, save_to_disk)\n File \"/usr/lib/python3.6/site-packages/libnmstate/net_state.py\", line 40, in __init__\n save_to_disk,\n File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 106, in __init__\n self._pre_edit_validation_and_cleanup()\n File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 128, in _pre_edit_validation_and_cleanup\n self._validate_over_booked_slaves()\n File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 423, in _validate_over_booked_slaves\n f\"Interface {iface.name} slave {slave_name} is \"\nlibnmstate.error.NmstateValueError: Interface br-ex slave enp5s0 is already enslaved by interface bond0\n'","stacktrace":"github.com/go-logr/zapr.(*zapLogger).Error\n\t/remote-source/app/vendor/github.com/go-logr/zapr/zapr.go:128\ngithub.com/nmstate/kubernetes-nmstate/controllers.(*NodeNetworkConfigurationPolicyReconciler).Reconcile\n\t/remote-source/app/controllers/nodenetworkconfigurationpolicy_controller.go:277\nsigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).reconcileHandler\n\t/remote-source/app/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:244\nsigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).processNextWorkItem\n\t/remote-source/app/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:218\nsigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).worker\n\t/remote-source/app/vendor/sigs.k8s.io/controller-runtime/pkg/internal/controller/controller.go:197\nk8s.io/apimachinery/pkg/util/wait.BackoffUntil.func1\n\t/remote-source/app/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:155\nk8s.io/apimachinery/pkg/util/wait.BackoffUntil\n\t/remote-source/app/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:156\nk8s.io/apimachinery/pkg/util/wait.JitterUntil\n\t/remote-source/app/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:133\nk8s.io/apimachinery/pkg/util/wait.Until\n\t/remote-source/app/vendor/k8s.io/apimachinery/pkg/util/wait/wait.go:90"} This is the networking layout on the nodes: enp4s0 bond0: enp5s0,enp6s0 br-ex bridge created during deployment includes bond0 interface nmcli con NAME UUID TYPE DEVICE ovs-if-br-ex e0b6fe95-b4c6-4bbb-8f15-50ad1bd6b718 ovs-interface br-ex Wired connection 1 9ca44b7c-265d-3fe3-bc51-7e52d84ab74c ethernet enp4s0 br-ex 49a80196-d3df-42d9-ac1b-33282d94ae8d ovs-bridge br-ex ovs-if-phys0 22fb5643-4768-4fff-839a-122a0868a6c5 bond bond0 ovs-port-br-ex 3c390181-35c4-4f6b-9fe6-464f10210121 ovs-port br-ex ovs-port-phys0 69aaa7d4-450d-40f6-80f9-696ebbc6bc72 ovs-port bond0 System enp5s0 9310e179-14b6-430a-6843-6491c047d532 ethernet enp5s0 System enp6s0 b43fa2aa-5a85-7b0a-9a20-469067dba6d6 ethernet enp6s0 bond0 ad33d8b0-1f7b-cab9-9447-ba07f855b143 bond -- [core@worker-0-0 ~]$ ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: enp4s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 52:54:00:35:d4:fc brd ff:ff:ff:ff:ff:ff inet6 fd00:1101::552a:b19:e27a:4e9/128 scope global dynamic noprefixroute valid_lft 2045sec preferred_lft 2045sec inet6 fe80::83db:b124:3a5d:20fd/64 scope link noprefixroute valid_lft forever preferred_lft forever 3: enp5s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000 link/ether 52:54:00:a5:49:43 brd ff:ff:ff:ff:ff:ff 4: enp6s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000 link/ether 52:54:00:a5:49:43 brd ff:ff:ff:ff:ff:ff 6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue master ovs-system state UP group default qlen 1000 link/ether 52:54:00:a5:49:43 brd ff:ff:ff:ff:ff:ff 7: ovs-system: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether 42:0e:c2:af:fe:9d brd ff:ff:ff:ff:ff:ff 8: br-ex: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000 link/ether 52:54:00:a5:49:43 brd ff:ff:ff:ff:ff:ff inet 192.168.123.123/24 brd 192.168.123.255 scope global dynamic noprefixroute br-ex valid_lft 2229sec preferred_lft 2229sec inet 192.168.123.10/32 scope global br-ex valid_lft forever preferred_lft forever inet6 fe80::acb7:e581:b640:6c69/64 scope link noprefixroute valid_lft forever preferred_lft forever [...] Version-Release number of selected component (if applicable): registry-proxy.engineering.redhat.com/rh-osbs/iib:35589 python3-libnmstate-0.3.4-17.el8_3.noarch nmstate-0.3.4-17.el8_3.noarc How reproducible: Steps to Reproduce: 1. Deploy OCP 4.7 via baremetal IPI flow. Nodes have the following network layout: one nic used for provisioning network and 2 nics grouped in a bond used for the control plane network 2. Deploy CNV 2.6 3. Create the following NNCP --- apiVersion: nmstate.io/v1alpha1 kind: NodeNetworkConfigurationPolicy metadata: name: bigip-bridges spec: nodeSelector: node-role.kubernetes.io/worker: "" desiredState: interfaces: - name: bond0.375 type: vlan state: up vlan: base-iface: bond0 id: 375 - name: bigip-mgmt description: Linux bridge with bond0 vlan375 as a port! type: linux-bridge state: up bridge: options: stp: enabled: false port: - name: bond0.375 - name: bond0.376 type: vlan state: up vlan: base-iface: bond0 id: 376 - name: bigip-ha description: Linux bridge with bond0 vlan376 as a port! type: linux-bridge state: up bridge: options: stp: enabled: false port: - name: bond0.376 Actual results: NNCP fails to get configured Expected results: NNCP is configured correctly Additional info: Attaching nmstate pods logs. --- Additional comment from Petr Horáček on 2021-01-06 11:35:20 UTC --- The traceback split to multiple lines: File \"/usr/bin/nmstatectl\", line 11, in <module> load_entry_point('nmstate==0.3.4', 'console_scripts', 'nmstatectl')() File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 67, in main return args.func(args) File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 267, in apply args.save_to_disk, File \"/usr/lib/python3.6/site-packages/nmstatectl/nmstatectl.py\", line 289, in apply_state save_to_disk=save_to_disk, File \"/usr/lib/python3.6/site-packages/libnmstate/netapplier.py\", line 69, in apply net_state = NetState(desired_state, current_state, save_to_disk) File \"/usr/lib/python3.6/site-packages/libnmstate/net_state.py\", line 40, in __init__ save_to_disk, File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 106, in __init__ self._pre_edit_validation_and_cleanup() File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 128, in _pre_edit_validation_and_cleanup self._validate_over_booked_slaves() File \"/usr/lib/python3.6/site-packages/libnmstate/ifaces/ifaces.py\", line 423, in _validate_over_booked_slaves f\"Interface {iface.name} slave {slave_name} is \" libnmstate.error.NmstateValueError: Interface br-ex slave enp5s0 is already enslaved by interface bond0 @fge I don't see anything wrong with the configuration, do you? Could it be a bug in nmstate? Note that br-ex (to which bond0 is connected) is an OVS bridge. --- Additional comment from Gris Ge on 2021-01-08 05:21:29 UTC --- nmstate-0.3 does not support applying state with OVS interface holding the same name with OVS bridge. In my test, when br_ex does not hold DNS or routes, the state apply goes well. Please confirm my guess on the network setup: ## Before nmstatectl run: * enp4s0 no idea. * enp5s0 and enp6s0 be bond0 slave/port. * bond0 be system interface of OVS bridge br_ex which also the OVS internal interface named as br_ex. * br0_ex OVS internal interface holding the DNS and default gateway. ## Nmstatectl run: * Create two vlan 375 and 376 out of bond0. And assign each vlan to a new bridge. --- Additional comment from Gris Ge on 2021-01-08 07:40:44 UTC --- Found the root cause. Base on the output of above `nmcli con`, you are using `br-ex` and `bond0` for both OVS bridge, OVS port and OVS interface. Which cause nmstate-0.3 think `bond0, eth1 and eth2` are port/slave of OVS bridge br_ex which the triggered above exception. I will continue my investigation on this. --- Additional comment from Gris Ge on 2021-01-08 08:26:56 UTC --- Even nmstate somehow supported OVS bridge/port/interface using the same name, you will still stopped by NetworkManager bug: https://bugzilla.redhat.com/show_bug.cgi?id=1914139 So my suggestion is change this `baremetal IPI flow` to use different name for OVS bridge/port/interface.
To clarify, we are unable to setup a VLAN iface on top of the bonding attached to br-ex. The cause is a NetworkManager bug https://bugzilla.redhat.com/show_bug.cgi?id=1914139. A workaround would be to use a unique name for the OVS port connecting to the bonding (currently it has the same name as the bonding).
Hi Petr, I am working from nmstate-0.3(RHEL 8.3) and nmstate-1.0(RHEL 8.4) to workaround the NetworkManager bug.
Closing this one since according to https://bugzilla.redhat.com/show_bug.cgi?id=1913248#c9 the issue will be addressed in nmstate.