Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create initial data for virtualization models #1491

Closed
candlerb opened this issue Sep 15, 2017 · 20 comments
Closed

Create initial data for virtualization models #1491

candlerb opened this issue Sep 15, 2017 · 20 comments
Labels
type: feature Introduction of new functionality to the application

Comments

@candlerb
Copy link
Contributor

candlerb commented Sep 15, 2017

Issue type

[X] Feature request
[ ] Bug report
[ ] Documentation

Environment

  • Python version: 3.5.2
  • NetBox version: v2.2-beta1

Description

Simple enhancement: loaddata initial_data could include

  • Initial cluster types: "Public Cloud", "vSphere", "oVirt", "libvirt", "lxd", "docker"
  • Initial clusters: "Amazon EC2", "Google GCE" (of type "Public Cloud")
  • New Device Role "VM Host"

Use case: makes it quicker to add your first VM when trying out Netbox. Then the minimal work is to create one VM in "Amazon EC2", or to create a cluster + VM.

@jeremystretch jeremystretch added beta type: feature Introduction of new functionality to the application labels Sep 15, 2017
@jeremystretch jeremystretch changed the title Virtualization: default data Create initial data for virtualization models Sep 15, 2017
@RyanBreaker
Copy link
Contributor

Just submitted the above pull request adding these items. I can easily add more if anyone has any additional suggestions before @jeremystretch accepts it.

@jeremystretch
Copy link
Member

I'll leave this open for a couple days in case anyone has suggestions.

@darkstar
Copy link

hmmm, how about CEPH & Kubernetes as cluster types? And MS Azure as additional initial cluster?

@candlerb
Copy link
Contributor Author

I wouldn't include Ceph because it's not a virtualization environment, it's a storage service (*). The others are fine.

(*) Maybe some people might want to model non-virtualization clusters as Netbox "cluster", but that wasn't what it was intended for

@Zorlin
Copy link

Zorlin commented Oct 17, 2017 via email

@darkstar
Copy link

In my opinion clusters should be limited to Compute workloads in Netbox.

That seems like an arbitrary decision and is not the way I'm using NetBox right now. Any reason behind your opinion?
We have virtualized storage systems that look and act like regular Virtual Machines.

@candlerb
Copy link
Contributor Author

We have virtualized storage systems that look and act like regular Virtual Machines.

Something that acts and looks like a Virtual Machine can be modelled as "Virtual Machine" - no problem with that.

However, a netbox "Cluster" is not a grouping of virtual machines; it's a grouping of virtualization hosts on which VMs can run. The database models the relation Virtual Machine : (runs-on) : Cluster

The issue raised was whether "Ceph" should be one of the options for "Cluster Type". I'd say no, because Ceph is not a virtualization platform; the database would let you mark VMs as executing on the Ceph cluster, which is not the case.

@darkstar
Copy link

@candlerb I agree with that. I think I was confused and actually meant something else but it doesn't matter ;)

@RyanBreaker
Copy link
Contributor

So do we want to add Azure as another cluster option? Otherwise, if someone uses one or the other, it would be a simple rename.

@ktims
Copy link

ktims commented Oct 20, 2017

Azure should be added, as should HyperV as a cluster type.

There are too many possible types of clusters to list here. I would again advocate for this to be added to the Admin panel and let the user decide how to use it.

@RyanBreaker
Copy link
Contributor

Added those two.

@candlerb
Copy link
Contributor Author

There are too many possible types of clusters to list here. I would again advocate for this to be added to the Admin panel and let the user decide how to use it

The cluster types are already user-editable. This ticket is just about creating sample data that can be preloaded at install time, like the pre-loaded set of Manufacturers and Platforms for example.

@jeremystretch
Copy link
Member

If we're going to list AWS and Azure as cloud providers, I feel somewhat obligated to list DigitalOcean too, given that they are NetBox's sole sponsor. However, I'm also very sensitive to potential conflicts of interest, so I wanted to propose it here. Any objections?

@rdegez
Copy link

rdegez commented Oct 26, 2017

Seems 100% fair to me and it's very nice of you to even ask, I would say!

@RyanBreaker
Copy link
Contributor

Added.

@cimnine
Copy link
Contributor

cimnine commented Oct 27, 2017

Since libvirt and vSphere are there, may I suggest Xen and maybe kvm?
And since Docker and lxd are there, may I suggest Kubernetes and Docker Swarm?

@jeremystretch
Copy link
Member

@RyanBreaker Thanks. Ping me when you want this merged.

@RyanBreaker
Copy link
Contributor

@cimnine I think it's already a bit extensive as it is, and it is only initial data that someone can optionally import, more or less as examples.

@jeremystretch If you're ready I'm ready, I think it's good.

@cimnine
Copy link
Contributor

cimnine commented Oct 30, 2017

@RyanBreaker Yeah, I thought so minutes after I commented.

@jeremystretch
Copy link
Member

Merged #1604

@lock lock bot locked as resolved and limited conversation to collaborators Jan 18, 2020
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
type: feature Introduction of new functionality to the application
Projects
None yet
Development

No branches or pull requests

8 participants