0

I am new to Kubernetes.

So I have this .NET Core console application Docker image that I built:

FROM mcr.microsoft.com/dotnet/core/runtime:3.0-buster-slim AS base
WORKDIR /app

FROM mcr.microsoft.com/dotnet/core/sdk:3.0-buster AS build
WORKDIR /src
COPY ["DataSync.Entry/DataSync.Entry.csproj", "DataSync.Entry/"]
RUN dotnet restore "DataSync.Entry/DataSync.Entry.csproj"
COPY . .
WORKDIR "/src/DataSync.Entry"
RUN dotnet build "DataSync.Entry.csproj" -c Release -o /app/build

FROM build AS publish
RUN dotnet publish "DataSync.Entry.csproj" -c Release -o /app/publish

FROM base AS final
WORKDIR /app
COPY --from=publish /app/publish .
ENTRYPOINT ["dotnet", "DataSync.Entry.dll"]

This image works perfectly on my machine. However when I try to deploy it as a CronJob in Kubernetes it fails.

Here's the YAML:

apiVersion: batch/v1beta1
kind: CronJob
metadata:
  name: netcore-cronjob
  namespace: aim-ns
  labels:
    app: netcore-cronjob
spec:
  schedule: "*/2 * * * *"
  jobTemplate:
    spec: #JOB
      backoffLimit: 1
      template: 
        spec: #POD
          containers:
            - name: netcore-job
              image: <ACR>.azurecr.io/aim/test-data-sync-netcore:v1 #my docker image above pushed to ACR
              command: ["dotnet", "DataSync.Entry.dll"]
              args: []
              resources:
                requests:
                  memory: "64Mi"
                  cpu: "100m"
                limits:
                  memory: "128Mi"
                  cpu: "150m"
          restartPolicy: Never

And the error message I got from viewing the logs on the Pod:

NAME                                        READY   STATUS    RESTARTS   AGE
netcore-cronjob-1582854120-scwfb            0/1     Error     0          19s

And when I run kubectl logs -n aim-ns netcore-cronjob-1582854120-scwfb I get:

  It was not possible to find any installed .NET Core SDKs
  Did you mean to run .NET Core SDK commands? Install a .NET Core SDK from:
      https://aka.ms/dotnet-download

What am I missing here?

Arshad Ahmad
  • 1
  • It seems that the image you think you're running on AWS is not the image that you were running locally. Are you able to `docker run .azurecr.io/aim/test-data-sync-netcore:v1 -- dotnet DataSync.Entry.dll`? This should prove that, what Kubernetes is running on AWS is what you're running locally. – DazWilkin Feb 28 '20 at 02:18
  • 1
    One gotcha with Docker is that `v1` is not unique. What you host uses as `v1` is not necessarily what another host (i.e. a remote Kubernetes node) thinks it `v1`. – DazWilkin Feb 28 '20 at 02:19
  • @DazWilkin thanks for your suggestion. I tried running that command and got a `C:\Program Files\Docker\Docker\resources\bin\docker.exe: Error response from daemon: OCI runtime create failed: container_linux.go:346: starting container process caused "exec: \"--\": executable file not found in $PATH": unknown.` error. Maybe we're on to something here. – Arshad Ahmad Feb 28 '20 at 02:39
  • 1
    I'd have preferred to see the same error that Kubernetes reported (`It was not possible to find any installed .NET Core SDKs`). Now we have a third problem :-( – DazWilkin Feb 28 '20 at 02:55
  • do a force docker pull and do a `ls` on the container, I think you are running on the wrong path `docker pull .azurecr.io/aim/test-data-sync-netcore:v1` `docker run --entrypoint /bin/bash .azurecr.io/aim/test-data-sync-netcore:v1` – Tummala Dhanvi Feb 28 '20 at 07:00

1 Answers1

0

I had the same problem and I needed to change the "FROM mcr.microsoft.com/dotnet/core/runtime:3.0-buster-slim AS base" (which is used for run the application) change to "FROM mcr.microsoft.com/dotnet/core/sdk:3.0-buster AS base". I thought the runtime has to be enough but...

bye Markus

screamingworld
  • 93
  • 1
  • 11