Bug Report: APISIX Standalone API-driven Mode Workers Not Ready on Startup

[luoyucumt]

Environment

• APISIX Helm Chart Version: 2.11.1
• Deployment Mode: Standalone API-driven mode
• Platform: Kubernetes with Helm Chart
• Components: Apisix + Apisix Ingress Controller

helm install apisix
--namespace ingress-apisix
--create-namespace
--set service.type=NodePort
--set apisix.deployment.role=traditional
--set apisix.deployment.role_traditional.config_provider=yaml
--set etcd.enabled=false
--set ingress-controller.enabled=true
--set ingress-controller.config.provider.type=apisix-standalone
--set ingress-controller.apisix.adminSerivce.namespace=ingress-apisix
--set ingress-controller.gatewayProxy.createDefault=true
charts/apisix

Problem Description
When deploying APISIX in standalone API-driven mode with ingress-controller, APISIX pods fail readiness checks with the error:
[lua] init.lua:925: status_ready(): worker id: 7 has not received configuration, client:

Cause Analysis

1. APISIX startup: Workers initialize with status = false (no configuration received)
2. Readiness probe: /apisix/status/ready returns 503 because workers have no config
3. Pod marked NotReady: Kubernetes service excludes the pod from traffic routing
4. ingress-controller startup: Cannot reach APISIX admin API (service has no ready endpoints)
5. Deadlock: ingress-controller can't sync initial config, APISIX workers never become ready

[AlinsRan]

ref: Fix the issue of configuring distribution strategies incorrectly in different modes apache/apisix-ingress-controller@59af65e

ref: Fix the issue adminSerivce configuration not working apache/apisix-ingress-controller#2442

This issue has been resolved in version 2.0.0-rc2 of the apisix-ingress-controller.

You can use APISIX's 2.11.2 chart version for verification.

[luoyucumt]

thanks for your answer, it solved my problem

[SRELETI]

Hey @luoyucumt @AlinsRan. Hope you are doing good. I'm facing the same issue described here. Could you please elaborate on how the fix linked above solves the problem? I'm trying to understand how the apisix-admin service returns apisix pod endpoints if the apisix pods are not ready?

I'm using APISIX's 2.12.3 chart version.