Need more information about where's the master and volume running

[neo3matrix]

Hello all,
Not sure if this is "an issue" but wanted to ask the following:

I am newbie to seaweedfs overall. So far, I have been setting up seaweedfs using docker and used to run master and volume containers separately. I also have seen this example where they run seaweedfs master+ volume on Minikube.

But in case of seaweeds-operator, when I deployed it on my local k8s cluster, I could see only 1 pod is running "seaweedfs-operator-controller-manager" which contains 2 containers - 1 is manager & 1 is kube-rbac-proxy.
The documentation says "Automatically deploy a SeaweedFS cluster with 3 masters, N volume servers, and M filers with customizable filer store managed by other operators."

I would like to know where/how can I see the 3 masters , N volumes running in this setup?
Any guide, documentation or something of that sort would really help.
Please help.

[neo3matrix]

@chrislusf Please help.

[neo3matrix]

I have a k8s cluster with 1 master & 2 worker nodes.

I Just followed the README instructions as follows:

1. git clone https://github.com/seaweedfs/seaweedfs-operator --depth=1
2. In my first attempt, I didn't want to use WEBHOOKS as I was just playing around - so didn't do ENABLE_WEBHOOKS & cert-manager steps.
3. make deploy

In my 2nd attempt, on the same k8s cluster, I did everything:

1. git clone https://github.com/seaweedfs/seaweedfs-operator --depth=1
2. Installed cert-manager using helm chart.
3. ENABLE_WEBHOOKS=true in config/manager/manager.yaml
4. make deploy

In any of those approaches, I could only see only one pod "seaweedfs-operator-controller-manager" which contains 2 containers - one is manager & another one is kube-rbac-proxy.
I tried to exec into those containers to check if they are running master or volume services but couldn't run any commands on them like ps, top, ls, etc.

Am I missing something?

[neo3matrix]

@chrislusf Any update/pointers?

[chrislusf]

related to #39
the recent kubernetes seem have some changes.

@thiscantbeserious is working on this.

[thiscantbeserious]

@neo3matrix yes - that's the fault of the readme :) - all you did right now is deploy the operator itself.

Now you need to configurate it - I'll push the readme changes to help understanding that better I guess I can do that now ...

Currently getting more firm on the basics and fixing a bug within cert-manager before I continue on the rebase.

[chrislusf]

@neo3matrix PR is merged. Thanks a lot!!!