Hello World with Fission (Error getting logs)

[endersonmenezes]

I'm testing Fission, following the tutorial present here: https://fission.io/docs/installation/

I'm currently using an installation without Helm, through this file: https://github.com/fission/fission/releases/download/v1.17.0/fission-all-v1.17.0.yaml

1. I can create the env.
2. I can create a function.

The following command: fission function test --name hello-py -v=2, returns me the following error with verbosity:

Setting up port forward to application=fission-api in namespace 
Waiting for local port 36491
Starting port forward from local port 36491
Using kubeconfig from environment "/home/end/Code/kubernetes/configs/tis-automation-b86046.yaml"
Connected to Kubernetes API
Connecting to port 8888 on pod fission/controller-68f4f6fd57-fm6qj
Starting port forwarder
Forwarding from 127.0.0.1:36491 -> 8888
Forwarding from [::1]:36491 -> 8888
Waiting for port forward 36491 to start...
Port forward from local port 36491 started
Setting up port forward to application=fission-router in namespace 
Handling connection for 36491
Waiting for local port 37377
Starting port forward from local port 37377
Using kubeconfig from environment "/home/end/Code/kubernetes/configs/tis-automation-b86046.yaml"
Connected to Kubernetes API
Connecting to port 8888 on pod fission/router-585fd48697-k68fw
Starting port forwarder
Forwarding from 127.0.0.1:37377 -> 8888
Forwarding from [::1]:37377 -> 8888
Waiting for port forward 37377 to start...
Port forward from local port 37377 started
Function test url: http://127.0.0.1:37377/fission-function/hello-py
Handling connection for 37377
GET /fission-function/hello-py HTTP/1.1
Host: 127.0.0.1:37377
User-Agent: Go-http-client/1.1
Accept-Encoding: gzip


Handling connection for 37377
HTTP/1.1 500 Internal Server Error
Content-Length: 33
Content-Type: text/plain; charset=utf-8
Date: Wed, 09 Nov 2022 14:19:22 GMT


Error: Error calling function hello-py: 500; Please try again or fix the error: error sending request to function
Try to get pod logs from controller 'logs/hello-py'
Handling connection for 36491
Error: Error getting function logs from controller: error getting logs from controller, status code: '500'. Try to get logs from log database.
Setting up port forward to application=fission-api in namespace 
Waiting for local port 42327
Starting port forward from local port 42327
Using kubeconfig from environment "/home/end/Code/kubernetes/configs/tis-automation-b86046.yaml"
Connected to Kubernetes API
Connecting to port 8888 on pod fission/controller-68f4f6fd57-fm6qj
Starting port forwarder
Forwarding from 127.0.0.1:42327 -> 8888
Forwarding from [::1]:42327 -> 8888
Waiting for port forward 42327 to start...
Handling connection for 42327
Port forward from local port 42327 started
Handling connection for 42327
Error querying logs: Internal error - 502 Bad Gateway

The controller log

{"level":"error","ts":"2022-11-09T14:19:22.862Z","logger":"controller.api","caller":"controller/api.go:125","msg":"no active pods found","code":500,"stacktrace":"github.com/fission/fission/pkg/controller.(*API).respondWithError\n\tpkg/controller/api.go:125\ngithub.com/fission/fission/pkg/controller.(*API).FunctionPodLogs\n\tpkg/controller/functionApi.go:330\nnet/http.HandlerFunc.ServeHTTP\n\t/opt/hostedtoolcache/go/1.19.0/x64/src/net/http/server.go:2109\ngithub.com/fission/fission/pkg/utils/metrics.HTTPMetricMiddleware.func1\n\tpkg/utils/metrics/http_metrics.go:88\nnet/http.HandlerFunc.ServeHTTP\n\t/opt/hostedtoolcache/go/1.19.0/x64/src/net/http/server.go:2109\ngithub.com/gorilla/mux.(*Router).ServeHTTP\n\t/home/runner/go/pkg/mod/github.com/gorilla/mux@v1.8.0/mux.go:210\ngo.opentelemetry.io/contrib/instrumentation/net/http/otelhttp.(*Handler).ServeHTTP\n\t/home/runner/go/pkg/mod/go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp@v0.34.0/handler.go:204\nnet/http.serverHandler.ServeHTTP\n\t/opt/hostedtoolcache/go/1.19.0/x64/src/net/http/server.go:2947\nnet/http.(*conn).serve\n\t/opt/hostedtoolcache/go/1.19.0/x64/src/net/http/server.go:1991"}
2022/11/09 14:19:25 http: proxy error: dial tcp: lookup influxdb on 10.43.0.10:53: server misbehaving

[endersonmenezes]

I managed to investigate further, and when we have an internal communication problem between the pods, the errors are kind of hazy, to fix my case I had to add this: *.fission to the noproxy variable. I believe that communication errors between pods could be better handled, I'll be reading the code more calmly and maybe I can contribute to this in the future.

[sanketsudake]

@kanuahs can you please take look here ?