cannot create one builder with multiple nodes on same host

Contributing guidelines

• I've read the contributing guidelines and wholeheartedly agree

I've found a bug and checked that ...

• ... the documentation does not mention anything about my problem
• ... there are no open or closed issues that are related to my problem

Description

$ /usr/libexec/docker/cli-plugins/buildx ls
NAME/NODE DRIVER/ENDPOINT STATUS  PLATFORMS
default * docker
  default default         running linux/amd64, linux/arm64, linux/riscv64, linux/ppc64le, linux/s390x, linux/386, linux/arm/v7, linux/arm/v6

$ /usr/libexec/docker/cli-plugins/buildx create --use    --driver docker-container --name mybuilder --platform linux/amd64 --node node1
mybuilder

$ /usr/libexec/docker/cli-plugins/buildx ls
NAME/NODE   DRIVER/ENDPOINT             STATUS   PLATFORMS
mybuilder * docker-container
  node1     unix:///var/run/docker.sock inactive linux/amd64*
default     docker
  default   default                     running  linux/amd64, linux/arm64, linux/riscv64, linux/ppc64le, linux/s390x, linux/386, linux/arm/v7, linux/arm/v6

$ /usr/libexec/docker/cli-plugins/buildx create --append --driver docker-container --name mybuilder --platform linux/arm64 --node node2
Error: invalid duplicate endpoint unix:///var/run/docker.sock
Usage:
  /usr/libexec/docker/cli-plugins/buildx create [OPTIONS] [CONTEXT|ENDPOINT] [flags]

Flags:
      --append                   Append a node to builder instead of changing it
      --buildkitd-flags string   Flags for buildkitd daemon
      --config string            BuildKit config file
      --driver string            Driver to use (available: [docker docker-container kubernetes])
      --driver-opt stringArray   Options for the driver
  -h, --help                     help for create
      --leave                    Remove a node from builder instead of changing it
      --name string              Builder instance name
      --node string              Create/modify node with given name
      --platform stringArray     Fixed platforms for current node
      --use                      Set the current builder instance

Global Flags:
      --builder string   Override the configured builder instance

Expected behaviour

I'd expect that I can have one builder that has multiple nodes for different architectures on the same (local) host.

Actual behaviour

/usr/libexec/docker/cli-plugins/buildx version
github.com/docker/buildx 0.0.0+unknown

That's weird, but that's what we get when we use amazonlinux:2 ?

docker run -it --rm amazonlinux:2

bash-4.2# amazon-linux-extras install docker -y
...
bash-4.2# /usr/libexec/docker/cli-plugins/buildx version
github.com/docker/buildx 0.0.0+unknown
bash-4.2# docker --version
Docker version 20.10.17, build 100c701

Buildx version

github.com/docker/buildx 0.0.0+unknown

Docker info

docker info
Client:
 Context:    default
 Debug Mode: false

Server:
 Containers: 3
  Running: 0
  Paused: 0
  Stopped: 3
 Images: 43
 Server Version: 20.10.17
 Storage Driver: overlay2
  Backing Filesystem: xfs
  Supports d_type: true
  Native Overlay Diff: true
  userxattr: false
 Logging Driver: json-file
 Cgroup Driver: cgroupfs
 Cgroup Version: 1
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
 Swarm: inactive
 Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: 9cd3357b7fd7218e4aec3eae239db1f68a5a6ec6
 runc version: 5fd4c4d144137e991c4acebb2146ab1483a97925
 init version: de40ad0
 Security Options:
  seccomp
   Profile: default
 Kernel Version: 5.10.167-147.601.amzn2.x86_64
 Operating System: Amazon Linux 2
 OSType: linux
 Architecture: x86_64
 CPUs: 2
 Total Memory: 7.663GiB
 Name: ip-10-20-11-55.us-west-2.compute.internal
 ID: KBGO:3A4V:KG6O:YASU:A74Z:2W4B:WEDV:GON6:EPLC:N3AF:N3AD:6OYJ
 Docker Root Dir: /var/lib/docker
 Debug Mode: false
 Registry: https://index.docker.io/v1/
 Labels:
 Experimental: true
 Insecure Registries:
  127.0.0.0/8
 Live Restore Enabled: false

also, this might be interesting:

docker run -it --rm amazonlinux:2

bash-4.2# amazon-linux-extras install docker -y
...
bash-4.2# /usr/libexec/docker/cli-plugins/buildx version
github.com/docker/buildx 0.0.0+unknown
bash-4.2# docker --version
Docker version 20.10.17, build 100c701

### Builders list

```text
/usr/libexec/docker/cli-plugins/buildx ls
NAME/NODE   DRIVER/ENDPOINT             STATUS   PLATFORMS
mybuilder * docker-container
  node1     unix:///var/run/docker.sock inactive linux/amd64*
default     docker
  default   default                     running  linux/amd64, linux/arm64, linux/riscv64, linux/ppc64le, linux/s390x, linux/386, linux/arm/v7, linux/arm/v6

Configuration

docker run -it --rm amazonlinux:2
amazon-linux-extras install docker -y

/usr/libexec/docker/cli-plugins/buildx ls
/usr/libexec/docker/cli-plugins/buildx create --use    --driver docker-container --name mybuilder --platform linux/amd64 --node node1
mybuilder
/usr/libexec/docker/cli-plugins/buildx create --append --driver docker-container --name mybuilder --platform linux/arm64 --node node2

Build logs

No response

Additional info

No response

Reproduce the issue with the buildx create --append --driver docker-container commands and two nodes using the same local Docker endpoint. Start by tracing the duplicate endpoint error in the builder creation path; done means one builder can contain multiple same-host nodes with different platforms, while existing duplicate-endpoint validation remains correct.