bug: Specified ectd.host from command line does not work

[stricklandye]

Current Behavior

Hi there.
I just follow the doc to install apisix ingress controller with a specified etcd host , here is :

helm install apisix \
	--version 2.10.1 \
  --namespace ingress-apisix \
  --create-namespace \
  --set apisix.deployment.role=traditional \
  --set apisix.deployment.role_traditional.config_provider=yaml \
  --set etcd.enabled=false \
  --set etcd.host=http://172.16.150.18:2379 \ 
  --set ingress-controller.enabled=true \
  --set ingress-controller.config.provider.type=apisix-standalone \
  --set ingress-controller.apisix.adminService.namespace=ingress-apisix \
  --set ingress-controller.gatewayProxy.createDefault=true \
  apisix/apisix

but the apisix stucks in CrashLoopBackOff status and the log is:

 /usr/local/openresty//luajit/bin/luajit ./apisix/cli/apisix.lua init                                                                                               
/usr/local/openresty//luajit/bin/luajit ./apisix/cli/apisix.lua init_etcd                                                                                         
request etcd endpoint 'http://etcd.host:2379/version' error, closed                                                                                                
 all etcd nodes are unavailable                                                                                                                                     │
 Warning! Request etcd endpoint 'http://etcd.host:2379/version' error, closed, retry time=1                                                                          Warning! Request etcd endpoint 'http://etcd.host:2379/version' error, closed, retry time=2

So it seems the command args not works as expected. How ever the ectd pod works fine :)

curl http://172.16.150.18:2379/version
{"etcdserver":"3.5.12","etcdcluster":"3.5.0"}

Expected Behavior

No response

Error Logs

No response

Steps to Reproduce

1. Install etcd through helm chart
2. setting etcd.host manually

Environment

version info:

apisix  ingress-apisix  1               2025-11-02 19:29:40.084123006 +0800 CST deployed        apisix-2.10.1   3.11.0

[kayx23]

try externalEtcd.host
https://github.com/apache/apisix-helm-chart/tree/master/charts/apisix

[stricklandye]

Yep. After checking the helm chart in the github repo I also found similar solution and hope the community will fix it someday :D
One more question ,according to the doc , create a ApisixRoute CRD the apisix ingress controller will then create the corresponding route rule but I list all the routes from admin api I got empty list. why?

# svc and pods
k get svc -A         
NAMESPACE        NAME                                    TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)                  AGE
default          kubernetes                              ClusterIP   10.96.0.1        <none>        443/TCP                  6h33m
default          nginx-svc                               ClusterIP   10.98.82.24      <none>        80/TCP                   65m

k get pods -A
NAMESPACE            NAME                                         READY   STATUS    RESTARTS         AGE
default              nginx-deployment-5599d6c467-px29j            1/1     Running   0                65m
default              nginx-deployment-5599d6c467-xksss            1/1     Running   0                65m

k get apisixroutes -A
NAMESPACE   NAME                 HOSTS                          URIS     AGE
default     nginx-route-backup   ["nginx-backup.example.com"]   ["/*"]   7m11s

# show all routes
curl "http://172.16.150.10:9180/apisix/admin/routes" -X GET -H "X-API-KEY: <key>"                                                       
{"list":[],"total":0}

[kayx23]

--set ingress-controller.config.provider.type=apisix-standalone

If you're using the standalone mode, you should use /apisix/admin/configs endpoint to see configured resources.

[kayx23]

Actually what mode are you using? standalone mode does not use etcd

[stricklandye]

I see. Finally, I didnt use the official helm repo and downloaded charts from github repo then I made some modification to the helm chart which just replaces ectd host with a external etcd service (ClusterIP) , I didnt modify any rest of configuration. So from the helm chart:

config:
  logLevel: "info"
  controllerName: apisix.apache.org/apisix-ingress-controller
  leaderElection:
    id: "apisix-ingress-controller-leader"
    leaseDuration: "15s"
    renewDeadline: "10s"
    retryPeriod: "2s"
    disable: false
  metricsAddr: ":8080"
  enableHTTP2: false
  probeAddr: ":8081"
  secureMetrics: false
  execADCTimeout: "15s"
  provider:
    type: "apisix"
    syncPeriod: "1m"
    initSyncDelay: "20m"
  kubernetes:
    ingressClass: apisix
    defaultIngressClass: false

Is it standalone mode ?

[stricklandye]

@kayx23 And sorry to bother you again. Could you plz also checkout this issue (I posted it in apisix repo)? . Thanks in advance 🙏