starred/netbox
1
0
Fork 0
mirror of https://github.com/netbox-community/netbox.git synced 2026-01-11 21:10:29 +01:00
Code Issues 273 Packages Projects Releases 10 Wiki Activity

Support the assignment of virtual machines as cluster hosts #1228

New Issue
Closed
opened 2025-12-29 16:30:24 +01:00 by adam · 4 comments
adam commented 2025-12-29 16:30:24 +01:00
Owner
Copy Link

Originally created by @candlerb on GitHub (Sep 15, 2017).

Issue type

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

Environment

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

Description

A VM runs on pool of resources called a "cluster". At the moment, a cluster can only consist of physical devices. I would like a cluster to have VMs as hosts too [^1]

Use cases:

  • You run a number of lxd containers inside an lxd server, which itself is a VM.
    • "container1.example.com" is an lxd container
    • It runs insides a VM called "lxd1.example.com"
    • The host VM "lxd1.example.com" in turn runs on a virtualization cluster made up of physical devices
    • So this could be modelled as:
      • "container1.example.com" is a VM running on cluster "lxd1"
      • "lxd1" is a cluster of type "lxd", with member devices (VMs) "lxd1.example.com"
      • "lxd1.example.com" is a VM running on cluster "main"
      • "main" is cluster of type "vSphere" with member devices (physical) X, Y, Z
      • note that cluster "lxd1" could comprise multiple VMs; that is, lxd containers can be migrated between lxd hosts
  • You run docker containers inside a third-party cloud
    • you fire up a VM inside a cloud
    • you run docker on that VM
    • inside that VM you run multiple docker containers
      • So our cluster "mydocker" has member "docker1.example.com" which is a cloud VM, not a physical device
      • The cloud VM is a member of a cluster called "EC2 (eu-west)" or suchlike.

[^1] It would probably be OK if a cluster were limited to take either physical devices or virtual machines; that is, you have to select "physical cluster" or "virtual cluster".

That might make the UI less confusing, although it blocks the (relatively unusual) case of a mixed cluster.

Originally created by @candlerb on GitHub (Sep 15, 2017). ### Issue type [X] Feature request <!-- Requesting the implementation of a new feature --> [ ] Bug report <!-- Reporting unexpected or erroneous behavior --> [ ] Documentation <!-- Proposing a modification to the documentation --> ### Environment * Python version: 3.5.2 * NetBox version: v2.2-beta1 ### Description A VM runs on pool of resources called a "cluster". At the moment, a cluster can only consist of physical devices. I would like a cluster to have VMs as hosts too [^1] Use cases: * You run a number of lxd containers inside an lxd server, which itself is a VM. * "container1.example.com" is an lxd container * It runs insides a VM called "lxd1.example.com" * The host VM "lxd1.example.com" in turn runs on a virtualization cluster made up of physical devices * So this could be modelled as: * "container1.example.com" is a VM running on cluster "lxd1" * "lxd1" is a cluster of type "lxd", with member devices (VMs) "lxd1.example.com" * "lxd1.example.com" is a VM running on cluster "main" * "main" is cluster of type "vSphere" with member devices (physical) X, Y, Z * note that cluster "lxd1" could comprise multiple VMs; that is, lxd containers can be migrated between lxd hosts * You run docker containers inside a third-party cloud * you fire up a VM inside a cloud * you run docker on that VM * inside that VM you run multiple docker containers * So our cluster "mydocker" has member "docker1.example.com" which is a cloud VM, not a physical device * The cloud VM is a member of a cluster called "EC2 (eu-west)" or suchlike. ---- [^1] It would probably be OK if a cluster were limited to take *either* physical devices *or* virtual machines; that is, you have to select "physical cluster" or "virtual cluster". That might make the UI less confusing, although it blocks the (relatively unusual) case of a mixed cluster.
adam closed this issue 2025-12-29 16:30:24 +01:00
adam commented 2025-12-29 16:30:25 +01:00
Author
Owner
Copy Link

@jeremystretch commented on GitHub (Sep 15, 2017):

I looked into this possibility while fleshing out the initial data model and it's just a nightmare to implement. The VirtualMachine model already has a foreign key to the Cluster model to indicate assignment as a VM. Adding a second relationship between the two introduces a whole slew of validation issues: preventing a VM from hosting its own cluster, preventing infinite recursion, preventing physical devices from hosting Docker-based clusters, etc.

NetBox's role is as an IPAM/DCIM application and I believe the model introduced in the beta has satisfied that role. You can still create VM-based clusters and track IP assignments to Docker containers, you just can't model the assignment of VMs as cluster hosts. (The assignment of hosts to a cluster is optional.) IMO this is a reasonable compromise and keeps us from creeping into VM orchestration territory, for which most organizations will use a separate tool anyway.

@jeremystretch commented on GitHub (Sep 15, 2017): I looked into this possibility while fleshing out the initial data model and it's just a nightmare to implement. The VirtualMachine model already has a foreign key to the Cluster model to indicate assignment as a VM. Adding a second relationship between the two introduces a whole slew of validation issues: preventing a VM from hosting its own cluster, preventing infinite recursion, preventing physical devices from hosting Docker-based clusters, etc. NetBox's role is as an IPAM/DCIM application and I believe the model introduced in the beta has satisfied that role. You can still create VM-based clusters and track IP assignments to Docker containers, you just can't model the assignment of VMs as cluster hosts. (The assignment of hosts to a cluster is optional.) IMO this is a reasonable compromise and keeps us from creeping into VM orchestration territory, for which most organizations will use a separate tool anyway.
adam commented 2025-12-29 16:30:25 +01:00
Author
Owner
Copy Link

@candlerb commented on GitHub (Sep 15, 2017):

OK, understood.

Many objects in Netbox have either "description" or "comments" field. How about adding one to Cluster? Then such cases can be documented ad-hoc.

@candlerb commented on GitHub (Sep 15, 2017): OK, understood. Many objects in Netbox have either "description" or "comments" field. How about adding one to Cluster? Then such cases can be documented ad-hoc.
adam commented 2025-12-29 16:30:25 +01:00
Author
Owner
Copy Link

@jeremystretch commented on GitHub (Sep 20, 2017):

Fixed the comments field on the Cluster object in #1514

@jeremystretch commented on GitHub (Sep 20, 2017): Fixed the `comments` field on the Cluster object in #1514
adam commented 2025-12-29 16:30:25 +01:00
Author
Owner
Copy Link

@snowie-swe commented on GitHub (Nov 11, 2018):

I looked into this possibility while fleshing out the initial data model and it's just a nightmare to implement. The VirtualMachine model already has a foreign key to the Cluster model to indicate assignment as a VM. Adding a second relationship between the two introduces a whole slew of validation issues: preventing a VM from hosting its own cluster, preventing infinite recursion, preventing physical devices from hosting Docker-based clusters, etc.

NetBox's role is as an IPAM/DCIM application and I believe the model introduced in the beta has satisfied that role. You can still create VM-based clusters and track IP assignments to Docker containers, you just can't model the assignment of VMs as cluster hosts. (The assignment of hosts to a cluster is optional.) IMO this is a reasonable compromise and keeps us from creeping into VM orchestration territory, for which most organizations will use a separate tool anyway.

@jeremystretch
This feature would also be very handy for virtual appliances like Loadbalancers and Firewalls.
Worth checking in to again? as more and more infrastructure become virtual.

@snowie-swe commented on GitHub (Nov 11, 2018): > I looked into this possibility while fleshing out the initial data model and it's just a nightmare to implement. The VirtualMachine model already has a foreign key to the Cluster model to indicate assignment as a VM. Adding a second relationship between the two introduces a whole slew of validation issues: preventing a VM from hosting its own cluster, preventing infinite recursion, preventing physical devices from hosting Docker-based clusters, etc. > > NetBox's role is as an IPAM/DCIM application and I believe the model introduced in the beta has satisfied that role. You can still create VM-based clusters and track IP assignments to Docker containers, you just can't model the assignment of VMs as cluster hosts. (The assignment of hosts to a cluster is optional.) IMO this is a reasonable compromise and keeps us from creeping into VM orchestration territory, for which most organizations will use a separate tool anyway. @jeremystretch This feature would also be very handy for virtual appliances like Loadbalancers and Firewalls. Worth checking in to again? as more and more infrastructure become virtual.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
beta
breaking change
complexity: high
complexity: low
complexity: medium
needs milestone
netbox
pending closure
plugin candidate
pull-request
Mirrored from GitHub Pull Request
 severity: high
severity: low
severity: medium
status: accepted
status: backlog
status: blocked
status: duplicate
status: needs owner
status: needs triage
status: revisions needed
status: under review
topic: GraphQL
topic: Internationalization
topic: OpenAPI
topic: UI/UX
topic: cabling
topic: event rules
topic: htmx navigation
topic: industrialization
topic: migrations
topic: plugins
topic: scripts
topic: templating
topic: testing
type: bug
type: deprecation
type: documentation
type: feature
type: housekeeping
type: translation
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
adam
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: starred/netbox#1228
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?