无法ping通flannel.1网卡

[zhutong196]

Expected Behavior

I use six servers to create k8s clusters, and there is no network segment between each two servers. Now I find that there are only four flannels 1 is connected, and the other two are only connected by themselves, which is not connected with the other four

10.57.2.129 flannel.1 10.244.0.0 10.57.2.130 flannel.1 10.244.1.0

10.57.154.18 flannel.1 10.244.2.0 10.57.154.19 flannel.1 10.244.3.0

10.57.157.48 flannel.1 10.244.7.0 10.57.157.49 flannel.1 10.244.8.0

Current Behavior

My node information: 10.244.0.0-10.244.3.0 not ping 10.244.7.0/10.244.8.0

10.244.7.0 can ping 10.244.8.0 each

Your Environment

• Flannel version: v0.11.0
• Backend used (e.g. vxlan or udp): vxlan
• Etcd version: 3.4.3-0
• Kubernetes version (if used): v1.17.0
• Operating System and version:
• Link to your project (optional):

[zhutong196]

the udp 8472 port of every host have open,So is this the problem caused by the host network or where I configure it? How do I check it

[manuelbuil]

Could you use a newer k8s and flannel version please? Perhaps your problem is fixed in newer versions

[zhangguanzhang]

docker info
iptables -S

[yuzhaopeng]

检查一下7.0 和 8.0两台机器上，flannel 启动时候绑定的网卡，有可能是flannel 错误的识别了网卡，可以手动指定一下

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.