Summary
Error 502 encountered when calling /jobs/cancel in a Kubernetes deployment. User is seeking information on how to determine the service the web app will request after calling cancel.
Question
Hi everyone, I tried to call /jobs/cancel, but the web app log shows a 502 error. I deployed it using Kubernetes. How can I find out which service the web app will request after calling cancel?
This topic has been created from a Slack thread to give it more visibility.
It will be on Read-Only mode here. Click here if you want
to access the original thread.
Join the conversation on Slack
["error-502", "kubernetes-deployment", "api", "service-request"]
It’ll ultimately talk to the server pod, but I’ve noticed cancels can be a bit slow—so I’d check your gateway/load balancer timeouts first. And also see how long it’s taking to fail (you’ll likely find that it’s some multiple of 30 seconds, or exactly 100 seconds which is usually good confirmation that it’s likely some default timeout).
Where to look really depends on your deployment. The docker compose version deploys a default nginx reverse proxy, and in Helm-based Kubernetes deployments you usually have something like a load balancer as an ingress. (For us on GCP/GKE, we have to set timeouts on the backend settings of the HTTP/S load balancer).