tokers edited a comment on issue #338:
URL:
https://github.com/apache/apisix-ingress-controller/issues/338#issuecomment-816340652
> @tokers I created the CRDs using the `samples/deploy` directory
> This worked and the `httpbin-route.yaml` was created successfully with
`apisix.apache.org/v2alpha1`
>
> Looking at the pod `apisix-ingress-controller-847df99d4c-v5g2p`, initially
it is up and running fine. But then it crashes and the following are the logs:
>
> ```
> 2021-04-09T00:41:08+08:00 info ingress/ingress.go:113 use configuration
> {
> "log_level": "info",
> "log_output": "stderr",
> "http_listen": ":8080",
> "enable_profiling": true,
> "kubernetes": {
> "kubeconfig": "",
> "resync_interval": "1m0s",
> "app_namespaces": [
> ""
> ],
> "election_id": "ingress-apisix-leader",
> "ingress_class": "apisix",
> "ingress_version": "networking/v1",
> "apisix_route_version": "apisix.apache.org/v2alpha1"
> },
> "apisix": {
> "base_url": "http://apisix-admin:9180/apisix/admin";,
> "admin_key": "edd1c9f034335f136f87ad84b625c8f1"
> }
> }
> I0409 00:41:08.128038 1 leaderelection.go:243] attempting to acquire
leader lease ingress-apisix/ingress-apisix-leader...
> 2021-04-09T00:41:08+08:00 info controller/controller.go:198 LeaderElection
{"message": "apisix-ingress-controller-847df99d4c-v5g2p became leader",
"event_type": "Normal"}
> I0409 00:41:08.146784 1 leaderelection.go:253] successfully acquired lease
ingress-apisix/ingress-apisix-leader
> 2021-04-09T00:41:08+08:00 info controller/controller.go:273 controller now
is running as leader {"namespace": "ingress-apisix", "pod":
"apisix-ingress-controller-847df99d4c-v5g2p"}
> 2021-04-09T00:41:08+08:00 warn apisix/cluster.go:235 waiting cluster to
ready, it may takes a while
> 2021-04-09T00:41:08+08:00 info apisix/cluster.go:113 syncing cache
{"cluster": ""}
> 2021-04-09T00:41:08+08:00 warn controller/controller.go:236 found a new
leader apisix-ingress-controller-847df99d4c-v5g2p
> 2021-04-09T00:41:08+08:00 info apisix/cluster.go:120 cache synced
{"cost_time": "14.051871ms", "cluster": ""}
> 2021-04-09T00:41:08+08:00 warn apisix/cluster.go:241 cluster now is ready,
cost time 14.277115ms
> 2021-04-09T00:41:08+08:00 info controller/endpoint.go:64 endpoints
controller started
> 2021-04-09T00:41:08+08:00 info controller/apisix_upstream.go:56
ApisixUpstream controller started
> 2021-04-09T00:41:08+08:00 info controller/ingress.go:59 ingress controller
started
> 2021-04-09T00:41:08+08:00 info controller/apisix_route.go:57 ApisixRoute
controller started
> 2021-04-09T00:41:08+08:00 info controller/secret.go:62 secret controller
started
> 2021-04-09T00:41:08+08:00 warn apisix/upstream.go:113 upstream not found
{"fullname": "kube-system_metrics-server_443", "url":
"http://apisix-admin:9180/apisix/admin/upstreams/4a344633";, "cluster": ""}
> 2021-04-09T00:41:08+08:00 warn controller/endpoint.go:140 upstream is not
referenced {"cluster": "name=; base_url=http://apisix-admin:9180/apisix/admin";,
"upstream": "kube-system_metrics-server_443"}
> 2021-04-09T00:41:08+08:00 warn apisix/upstream.go:113 upstream not found
{"fullname": "ingress-apisix_apisix-etcd-headless_2380", "url":
"http://apisix-admin:9180/apisix/admin/upstreams/82c8442b";, "cluster": ""}
> 2021-04-09T00:41:08+08:00 warn controller/endpoint.go:140 upstream is not
referenced {"cluster": "name=; base_url=http://apisix-admin:9180/apisix/admin";,
"upstream": "ingress-apisix_apisix-etcd-headless_2380"}
> panic: runtime error: invalid memory address or nil pointer dereference
> [signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0x15d2e7d]
>
> goroutine 177 [running]:
>
github.com/apache/apisix-ingress-controller/pkg/kube/translation.(*translator).TranslateRouteV2alpha1(0xc0000c3110,
0xc0003bb680, 0x7, 0xc00060dee8, 0x10, 0x1c95760, 0xc00032f640, 0x0, 0x0, 0x0)
> /build/controller/pkg/kube/translation/apisix_route.go:110 +0x74d
>
github.com/apache/apisix-ingress-controller/pkg/ingress/controller.(*apisixRouteController).sync(0xc00032e000,
0x1c95560, 0xc00053b080, 0xc00063fe00, 0x1a70df8, 0x0)
> /build/controller/pkg/ingress/controller/apisix_route.go:142 +0x39e
>
github.com/apache/apisix-ingress-controller/pkg/ingress/controller.(*apisixRouteController).runWorker(0xc00032e000,
0x1c95560, 0xc00053b080)
> /build/controller/pkg/ingress/controller/apisix_route.go:78 +0x57
> created by
github.com/apache/apisix-ingress-controller/pkg/ingress/controller.(*apisixRouteController).run
> /build/controller/pkg/ingress/controller/apisix_route.go:66 +0x1ef
> ```
>
> The tests fail
(https://github.com/apache/apisix-ingress-controller/blob/master/docs/en/latest/practices/proxy-the-httpbin-service.md#test)
and I assume the pod failure is the issue
I'm sorry because in `0.4.0` we don't have a field called `backends` but
only `backend`, so you can use the following snippet instead:
```yaml
# httpbin-route.yaml
apiVersion: apisix.apache.org/v2alpha1
kind: ApisixRoute
metadata:
name: httpserver-route
spec:
http:
- name: rule1
match:
hosts:
- local.httpbin.org
paths:
- /*
backends:
- serviceName: httpbin
servicePort: 80
```
We'll soon relaese the apisix-ingress-controller `0.5.0`, which fixed
several bugs and is more robust.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
