Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Linkerd still routes to deleted Kubernetes endpoints

See original GitHub issue

Issue Type:

[x ] Bug report
Feature request

What happened:

I deploy an simple app in k8s,

kubectl get endpoints world-v2-service
world-v2-service   192.168.10.141:7778,192.168.10.144:7778

The endpoints ip are ‘192.168.10.141’ and ‘192.168.10.144’; I can visit by

# curl worldv2.test.com
earth (192.168.10.141)!

Then I delete the app and redeploy it, and new endpoints are

kubectl get endpoints world-v2-service
world-v2-service   192.168.10.145:7778,192.168.10.146:7778

but when I visit the ingress url , the log show

E 0907 08:45:51.882 UTC THREAD33: service failure: Failure(Invalid argument: /192.168.10.141:7778 at remote address: /192.168.10.141:7778. Remote Info: Not Available, flags=0x09) with RemoteInfo -> Upstream Address: /192.168.12.192:55967, Upstream Client Id: Not Available, Downstream Address: /192.168.10.141:7778, Downstream Client Id: #/io.l5d.k8s/default/http/world-v2-service, Trace Id: cd10d485ed2ff694.cd10d485ed2ff694<:cd10d485ed2ff694

the linkerd is still search for old endpoints.

Issue Analytics

State:
Created 6 years ago
Comments:22 (12 by maintainers)

Top GitHub Comments

1reaction

hawkwcommented, Sep 13, 2017

Hi @wjwinner, @weitzj, and @ihac, just wanted to let you all know I have a pull request #1638 that fixes this issue. We’ll try to get a hotfix out for this ASAP.

1reaction

weitzjcommented, Sep 8, 2017

I am using kubernetes 1.7.5 on AWS with flannel. RBAC is enabled. Service has ClusterIP: None Kubernetes was deployed as multi AZ master, multi AZ node using kops

Top Results From Across the Web

Linkerd continues to talk to old endpoint after Kubernetes ...

We are running a Kubernetes cluster, and this Linkerd receives traffic from the Internet via an ELB and routes to the appropriate services ......

Troubleshooting

This section provides resolution steps for common problems reported with the linkerd check command. The “pre-kubernetes-cluster-setup” checks.

Uninstalling Linkerd

Removing Linkerd from a Kubernetes cluster requires a few steps: removing any data plane proxies, removing all the extensions and then removing the...

Configuring Per-Route Policy

This command creates a namespace for the demo, downloads its Kubernetes resource ... Meanwhile, the webapp service may also send DELETE and PUT...

Topology-Aware Service Routing on Kubernetes with Linkerd

It subscribes to a service in the EndpointsWatcher and translates any create, delete, or update operations on the cached resources into updates ...