- The following error is observed in Clarity UI when scoped to a Kubernetes region.
The cluster <cluster_name> seems to be taking too long to respond.
- The following errors are subsequently observed in /var/log/nginx/error.log on the DU.
[error] 5576#5576: *101638 upstream prematurely closed connection while reading response header from upstream, client: 127.0.0.1, server: , request: "GET /qbert/v1/clusters/<CLUSTER_UUID>/k8sapi/api/v1/nodes HTTP/1.1", upstream: "http://127.0.0.1:3000/v1/clusters/<CLUSTER_UUID>/k8sapi/api/v1/nodes", host: "<DU_FQDN>"
- Clarity UI is failing to load these resources as the Qbert Kubernetes API endpoint is returning an HTTP 502 Bad Gateway response code for the associated requests.
- Platform9 Managed Kubernetes - All Version
- Clarity UI
Qbert is not able to create a socket hence no session is created for the forwarder.
- Restart the Qbert service on the DU.
# systemctl restart pf9-qbert
An internal JIRA is opened for a detailed analysis of this issue.