You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What happened:
UDP connection from edge to cloud will cause "stream reset", then UDP connection will not available again, only restart edgemesh on edge node will recover.
I1011 16:55:27.030478 1 proxysocket.go:180] "New UDP connection from client" address="x.x.x.x:50811"
I1011 16:55:27.030514 1 loadbalancer.go:618] "NextEndpoint for service" servicePortName="kube-system/kube-dns-edge:dns" address="x.x.x.x:50811" endpoints=[cloudnode1:coredns-548878875f-77cws:a.a.a.a:53 cloudnode1:coredns-548878875f-zt5v7:b.b.b.b:53]
I1011 16:55:27.030534 1 loadbalancer.go:679] "Mapped service to endpoint" service="kube-system/kube-dns-edge:dns" endpoint="cloudnode1:coredns-548878875f-77cws:b.b.b.b:53"
I1011 16:55:27.030628 1 tunnel.go:278] New stream between peer cloudnode1: {12D3KooWBr91n5RCjPuyxvscZjRDvYXpJQjkkNuDVm9B9BKrewNg: [/ip4/x.x.x.x/tcp/20006/p2p/12D3KooWBr91n5RCjPuyxvscZjRDvYXpJQjkkNuDVm9B9BKrewNg/p2p-circuit /ip4/x.x.x.x/tcp/20006/p2p/12D3KooWKLgcEsqdBpJaT1dcqvLqv8NDwVjkDJMRUBoALwf71CeJ/p2p-circuit /ip4/x.x.x.x/tcp/20006/p2p/12D3KooWDHKkUym46jSG2G5QKNrTpQioqRQeTDz4oiWxAQymi5yp/p2p-circuit /ip4/x.x.x.x/tcp/20006 /ip4/x.x.x.x/tcp/20006 /ip4/127.0.0.1/tcp/20006 /ip4/x.x.x.x/tcp/20006/p2p/12D3KooWDHKkUym46jSG2G5QKNrTpQioqRQeTDz4oiWxAQymi5yp/p2p-circuit /ip4/x.x.x.x/tcp/20006/p2p/12D3KooWBr91n5RCjPuyxvscZjRDvYXpJQjkkNuDVm9B9BKrewNg/p2p-circuit /ip4/x.x.x.x/tcp/20006/p2p/12D3KooWKLgcEsqdBpJaT1dcqvLqv8NDwVjkDJMRUBoALwf71CeJ/p2p-circuit]} success
E1011 16:55:27.031395 1 loadbalancer.go:686] "Dial failed" err="get proxy stream from cloudnode1 error: stream between paas241 reset err: stream reset"
What you expected to happen:
Do not "stream reset"
How to reproduce it (as minimally and precisely as possible):
expose a UDP service to edge, for example, coredns on cloud. Then run this on edge node.
for((i=0;;i++)); do echo $i; nslookup kubernetes.default.svc z.z.z.z; done
z.z.z.z is the ClusterIP of coredns service.
about 230 times you will see.
Anything else we need to know?:
Environment:
EdgeMesh version:1.13
Kubernetes version (use kubectl version):1.2.6
KubeEdge version(e.g. cloudcore --version and edgecore --version):1.13
Cloud nodes Environment:
Hardware configuration (e.g. lscpu):
OS (e.g. cat /etc/os-release):
Kernel (e.g. uname -a):
Go version (e.g. go version):
Others:
Edge nodes Environment:
edgecore version (e.g. edgecore --version):
Hardware configuration (e.g. lscpu):
OS (e.g. cat /etc/os-release):
Kernel (e.g. uname -a):
Go version (e.g. go version):
Others:
The text was updated successfully, but these errors were encountered:
What happened:
UDP connection from edge to cloud will cause "stream reset", then UDP connection will not available again, only restart edgemesh on edge node will recover.
What you expected to happen:
Do not "stream reset"
How to reproduce it (as minimally and precisely as possible):
expose a UDP service to edge, for example, coredns on cloud. Then run this on edge node.
z.z.z.z is the ClusterIP of coredns service.
about 230 times you will see.
Anything else we need to know?:
Environment:
EdgeMesh version:1.13
Kubernetes version (use
kubectl version
):1.2.6KubeEdge version(e.g.
cloudcore --version
andedgecore --version
):1.13Cloud nodes Environment:
lscpu
):cat /etc/os-release
):uname -a
):go version
):Edge nodes Environment:
edgecore --version
):lscpu
):cat /etc/os-release
):uname -a
):go version
):The text was updated successfully, but these errors were encountered: