Skip to content

Commit 7c9ec57

Browse files
author
Paul D.Smith
committed
Complete TSG.
1 parent 7f4f2dd commit 7c9ec57

File tree

1 file changed

+18
-5
lines changed

1 file changed

+18
-5
lines changed

articles/operator-nexus/troubleshoot-dns-issues.md

Lines changed: 18 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -17,20 +17,33 @@ cluster running on Azure VMs (Virtual Machines) and Azure, accessing Azure resou
1717
domain names. However a DNS (Domain Name System) error in NNF can mean that Azure resources
1818
can't be contacted which impacts deployment or management of Nexus resources.
1919

20+
The DNS proxy that causes this error is an [Envoy DNS Proxy](https://www.envoyproxy.io/docs/envoy/latest/)
21+
running via a Kubernetes deployment in either an Infrastructure or Tenant cluster.
22+
The precise location of the DNS proxy will have been determined when the customer
23+
has deployed ther NAKS (Nexus Azure Kubernetes Service) cluster or during some other
24+
deployment.
25+
2026
## Diagnosis
2127

2228
* Deployment or management of remote Nexus resources fails with "DeploymentFailed."
2329
* Azure portal shows no errors being generated for the Azure resources that are unreachable; there are no errors because the failing operations aren't reaching the Azure resources at all.
2430

2531
## Mitigation steps
2632

27-
Follow these steps for mitigation.
28-
2933
### Trigger a DNS cache refresh for the NNF Workload Proxy
30-
34+
35+
- Based on configuration and deployment, identify the Infrastructure or Tenant Kubernetes Cluster on which the DNS proxy is runnning
36+
- Log-in to the Kubernetes cluster
37+
- Using the Azure portal, find your cluster
38+
- From the _Overview_ blade, click the _Connect_ command (between _Refresh_ and _Delete_)
39+
- Follow the instructions from the resulting pop-up window that explains how to use Kubernetes cluster
40+
- Identify the DNS proxy deployment using this command
41+
```bash
42+
$ kubectl get deployments --all-namespaces=true | grep envoy
43+
```
44+
- Restart the deployment, which will cause the DNS caching to be reset, using this command:
3145
```bash
32-
# TBD awaiting feedback from folks who worked on the original IcMs as to
33-
# what commands should appear here.
46+
kubectl rollout restart deployment <your-envoy-deployment-name> --namespace <namespace-where-envoy-pod-exists>
3447
```
3548

3649
## Verification

0 commit comments

Comments
 (0)