Open paleksic opened 5 years ago
I am also seeing a lot of the same error. But the network seems to work ... It is very confusing ...
I am using Version: 2.5.0 (version check update disabled)
Service: router
Protocol: weave 1..2
Name: 7e:98:57:a0:b6:d5(mykc-dev-blue-k8s-master-nf-3)
Encryption: enabled
PeerDiscovery: enabled Targets: 4 Connections: 4 (3 established, 1 failed) Peers: 4 (with 12 established connections) TrustedSubnets: none
Service: ipam
Status: ready
Range: 10.233.64.0/18
DefaultSubnet: 10.233.64.0/18
My error log looks similar. The more nodes the more errors ... . Very, very annoying ...
ERRO: 2019/02/02 23:08:37.833752 Captured frame from MAC (3e:89:96:d5:fb:01) to (ba:18:12:53:40:44) associated with another peer 26:5f:ee:b4:31:b6(mykc-dev-blue-k8s-master-nf-2) ERRO: 2019/02/02 23:08:38.464359 Captured frame from MAC (3e:89:96:d5:fb:01) to (7e:9d:8c:fb:20:5c) associated with another peer 26:5f:ee:b4:31:b6(mykc-dev-blue-k8s-master-nf-2) INFO: 2019/02/02 23:08:43.215362 Discovered remote MAC 92:eb:f3:f0:13:6e at 7e:98:57:a0:b6:d5(mykc-dev-blue-k8s-master-nf-3) ERRO: 2019/02/02 23:08:43.215411 Captured frame from MAC (92:eb:f3:f0:13:6e) to (ba:18:12:53:40:44) associated with another peer 7e:98:57:a0:b6:d5(mykc-dev-blue-k8s-master-nf-3) ERRO: 2019/02/02 23:08:43.904890 Captured frame from MAC (92:eb:f3:f0:13:6e) to (7e:9d:8c:fb:20:5c) associated with another peer 7e:98:57:a0:b6:d5(mykc-dev-blue-k8s-master-nf-3)
Same here I'm getting tons of them but network works, only thing I found is that hairpin is off even though kubelet and weave are configured to enable it.
my yaml setup for the weave pods look like this (created using kubespray)
scheduler.alpha.kubernetes.io/critical-pod: ''
spec:
priorityClassName: system-node-critical
containers:
- name: weave
command:
- /home/weave/launch.sh
env:
- name: HOSTNAME
valueFrom:
fieldRef:
apiVersion: v1
fieldPath: spec.nodeName
- name: WEAVE_PASSWORD
valueFrom:
secretKeyRef:
name: weave-net
key: WEAVE_PASSWORD
- name: CHECKPOINT_DISABLE
value: "0"
- name: CONN_LIMIT
value: "100"
- name: HAIRPIN_MODE
value: "true"
- name: IPALLOC_RANGE
value: "10.233.64.0/18"
- name: EXPECT_NPC
value: "1"
- name: WEAVE_MTU
value: "1376"
- name: NO_MASQ_LOCAL
value: "1"
image: docker.io/weaveworks/weave-kube:2.5.0
We're both using kubespray and weave 2.5.0 configuration is the same. I'm on k8s 1.12 No idea how to troubleshoot this further and why my veth bridge hairpin is off and if that has anything to do with error.
how do you check "veth bridge hairpin is off" ? are you looking into journalctl ?
This is how my journalctl log looks like:
root@mykc-dev-blue-k8s-master-nf-1:~# journalctl | grep hairpin Feb 02 23:39:09 mykc-dev-blue-k8s-master-nf-1 kubelet[14641]: I0202 23:39:09.905985 14641 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:39:20 mykc-dev-blue-k8s-master-nf-1 kubelet[14746]: I0202 23:39:20.598608 14746 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:39:31 mykc-dev-blue-k8s-master-nf-1 kubelet[14825]: I0202 23:39:31.068918 14825 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:39:41 mykc-dev-blue-k8s-master-nf-1 kubelet[14971]: I0202 23:39:41.315733 14971 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:39:51 mykc-dev-blue-k8s-master-nf-1 kubelet[15074]: I0202 23:39:51.509612 15074 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:40:01 mykc-dev-blue-k8s-master-nf-1 kubelet[15227]: I0202 23:40:01.229541 15227 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:40:11 mykc-dev-blue-k8s-master-nf-1 kubelet[15241]: I0202 23:40:11.525293 15241 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:40:12 mykc-dev-blue-k8s-master-nf-1 kubelet[15241]: W0202 23:40:12.846739 15241 docker_service.go:540] Hairpin mode set to "promiscuous-bridge" but kubenet is not enabled, falling back to "hairpin-veth" Feb 02 23:40:12 mykc-dev-blue-k8s-master-nf-1 kubelet[15241]: I0202 23:40:12.846765 15241 docker_service.go:236] Hairpin mode set to "hairpin-veth" Feb 02 23:42:10 mykc-dev-blue-k8s-master-nf-1 kubelet[15935]: I0202 23:42:10.739280 15935 flags.go:33] FLAG: --hairpin-mode="promiscuous-bridge" Feb 02 23:42:12 mykc-dev-blue-k8s-master-nf-1 kubelet[15935]: W0202 23:42:12.238495 15935 docker_service.go:540] Hairpin mode set to "promiscuous-bridge" but kubenet is not enabled, falling back to "hairpin-veth" Feb 02 23:42:12 mykc-dev-blue-k8s-master-nf-1 kubelet[15935]: I0202 23:42:12.238875 15935 docker_service.go:236] Hairpin mode set to "hairpin-veth"
ip -d link show vethwe-bridge
link/ether de:7d:50:50:dc:33 brd ff:ff:ff:ff:ff:ff promiscuity 1
veth
bridge_slave state forwarding priority 32 cost 2 hairpin off guard off root_block off fastleave off learning on flood on addrgenmode eui64```
Some discussion at #2877 of how this can happen and be innocuous.
I'm seeing bunch of these error messages. Is it safe to ignore them as I've read somewhere it can be logging issue.