configuring flannel to use a non default interface in kubernetes
Solution 1
I've the same problem, trying to use k8s and Vagrant. I've found this note in the documentation of flannel:
Vagrant typically assigns two interfaces to all VMs. The first, for which all hosts are assigned the IP address 10.0.2.15, is for external traffic that gets NATed.
This may lead to problems with flannel. By default, flannel selects the first interface on a host. This leads to all hosts thinking they have the same public IP address. To prevent this issue, pass the --iface eth1 flag to flannel so that the second interface is chosen.
So I look for it in the flannel's pod configuration. If you download the kube-flannel.yml file, you should look at DaemonSet spec, specifically at the "kube-flannel" container. There, you should add the required "--iface=enp0s8" argument (Don't forget the "="). Part of the code I've used.
containers:
- name: kube-flannel
image: quay.io/coreos/flannel:v0.10.0-amd64
command:
- /opt/bin/flanneld
args:
- --ip-masq
- --kube-subnet-mgr
- --iface=enp0s8
Then run kubectl apply -f kube-flannel.yml
Hope helps.
Solution 2
don't know directly running "kubectl apply -f kube-flannel.yml" does not work at my side, it still show using interface with name eth0.
after running kubectl delete -f kube-flannel.yml
then kubectl apply -f kube-flannel.yml
, it shows using the interface with eth1:
I1122 11:31:44.405982 1 main.go:488] Using interface with name eth1 and address 192.168.0.24
I1122 11:31:44.406153 1 main.go:505] Defaulting external address to interface address (192.168.0.24)
I1122 11:31:44.428414 1 kube.go:131] Waiting 10m0s for node controller to sync
I1122 11:31:44.428552 1 kube.go:294] Starting kube subnet manager
I1122 11:31:45.429349 1 kube.go:138] Node controller sync successful
clvx
Updated on June 13, 2022Comments
-
clvx almost 2 years
Is there a way to define in which interface Flannel should be listening? According to his documentation adding FLANNEL_OPTIONS="--iface=enp0s8" in /etc/sysconfig/flanneld should work, but it isn't.
My master node configuration is running in a xenial(ubuntu 16.04) vagrant:
$ sudo kubeadm init --pod-network-cidr 10.244.0.0/16 --apiserver-advertise-address 10.0.0.10 $ kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml clusterrole "flannel" created clusterrolebinding "flannel" created serviceaccount "flannel" created configmap "kube-flannel-cfg" created daemonset "kube-flannel-ds" created ubuntu@master:~$ kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/k8s-manifests/kube-flannel-rbac.yml clusterrole "flannel" configured clusterrolebinding "flannel" configured
Host ip addresses:
$ ip addr 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1 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: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 02:63:8e:2c:ef:cd brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global enp0s3 valid_lft forever preferred_lft forever inet6 fe80::63:8eff:fe2c:efcd/64 scope link valid_lft forever preferred_lft forever 3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 08:00:27:fb:ad:bb brd ff:ff:ff:ff:ff:ff inet 10.0.0.10/24 brd 10.0.0.255 scope global enp0s8 valid_lft forever preferred_lft forever 4: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default link/ether 02:42:da:aa:6e:13 brd ff:ff:ff:ff:ff:ff inet 172.17.0.1/16 scope global docker0 valid_lft forever preferred_lft forever 5: flannel.1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc noqueue state UNKNOWN group default link/ether 5e:07:a1:7f:97:53 brd ff:ff:ff:ff:ff:ff inet 10.244.0.0/32 scope global flannel.1 valid_lft forever preferred_lft forever inet6 fe80::5c07:a1ff:fe7f:9753/64 scope link valid_lft forever preferred_lft forever 6: cni0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000 link/ether 0a:58:0a:f4:00:01 brd ff:ff:ff:ff:ff:ff inet 10.244.0.1/24 scope global cni0 valid_lft forever preferred_lft forever inet6 fe80::7037:fcff:fe41:b1fb/64 scope link valid_lft forever preferred_lft forever
Pods names:
$ kubectl get pods --all-namespaces NAMESPACE NAME READY STATUS RESTARTS AGE kube-system etcd-master 1/1 Running 0 2m kube-system kube-apiserver-master 1/1 Running 0 2m kube-system kube-controller-manager-master 1/1 Running 0 2m kube-system kube-dns-545bc4bfd4-gjjth 0/3 ContainerCreating 0 3m kube-system kube-flannel-ds-gdz8f 1/1 Running 0 1m kube-system kube-flannel-ds-h4fd2 1/1 Running 0 33s kube-system kube-flannel-ds-rnlsz 1/1 Running 1 33s kube-system kube-proxy-d4wv9 1/1 Running 0 33s kube-system kube-proxy-fdkqn 1/1 Running 0 3m kube-system kube-proxy-kj7tn 1/1 Running 0 33s kube-system kube-scheduler-master 1/1 Running 0 2m
Flannel Logs:
$ kubectl logs -n kube-system kube-flannel-ds-gdz8f -c kube-flannel I1216 12:00:35.817207 1 main.go:474] Determining IP address of default interface I1216 12:00:35.822082 1 main.go:487] Using interface with name enp0s3 and address 10.0.2.15 I1216 12:00:35.822335 1 main.go:504] Defaulting external address to interface address (10.0.2.15) I1216 12:00:35.909906 1 kube.go:130] Waiting 10m0s for node controller to sync I1216 12:00:35.909950 1 kube.go:283] Starting kube subnet manager I1216 12:00:36.987719 1 kube.go:137] Node controller sync successful I1216 12:00:37.087300 1 main.go:234] Created subnet manager: Kubernetes Subnet Manager - master I1216 12:00:37.087433 1 main.go:237] Installing signal handlers I1216 12:00:37.088836 1 main.go:352] Found network config - Backend type: vxlan I1216 12:00:37.089018 1 vxlan.go:119] VXLAN config: VNI=1 Port=0 GBP=false DirectRouting=false I1216 12:00:37.295988 1 main.go:299] Wrote subnet file to /run/flannel/subnet.env I1216 12:00:37.296025 1 main.go:303] Running backend. I1216 12:00:37.296048 1 main.go:321] Waiting for all goroutines to exit I1216 12:00:37.296084 1 vxlan_network.go:56] watching for new subnet leases
How do I do to configure flannel in kubernetes to listen in enp0s8 instead of enp0s3?
-
CTodea almost 6 yearsI was suspecting that the problem was around the iface flannel was using for the vxlan, but was trying to solve it changing the routing table. Your solution worked like a charm.
-
jonashackt almost 6 yearsAnd also be sure to add the following to kubelet.service:
--node-ip={{ hostvars[inventory_hostname]['ansible_enp0s8']['ipv4']['address'] }}
, this way the correct host IPs will be matching the Pod CIDRs -
Hao over 5 yearsmy iface is eth0
-
mlazarov over 5 yearsHey lalo, your solution will work only if all nodes are the same! If the interface on some of the nodes isn't enp0s8 this won't work.
-
Bathz about 5 yearsThanks for your comment, @Libo-zhu, we experienced exactly the same behaviour.
-
Alex G almost 5 years--iface-regex=10\.0\.*\.* is the answer
-
tftd over 2 years@AlexG thank you sir for this wonderful hint! This was EXACTLY what I was looking for. In my case we were building a hybrid cluster and some nodes where in different subnet. The
iface-regex
idea works absolutely brilliant! :)