Compute API 参考

Compute API。

共 55 个端点,其中 18 个只读。每一个都同时是 orbit 命令行中的命令、SDK 中的方法和 MCP 服务器上的工具。

01

通用

端点权限参数CLI · MCP 工具
GET /compute/overview
Regions with their system images, the packages you can order, and your account's server limits
只读—orbit compute overview
compute_overview
02

instances

端点权限参数CLI · MCP 工具
GET /compute/instances
Your servers
只读—orbit compute instances list
compute_instances_list
POST /compute/instances
Create a server: region, package, system image, host name and SSH keys. Answers 202 with the queued job
读写请求体: JSONorbit compute instances create
compute_instances_create
GET /compute/instances/{id}
One server: state, power, the job in progress, addresses, resources and this period's traffic
只读id stringorbit compute instances get <id>
compute_instances_get
GET /compute/instances/{id}/options
What a server can switch to: packages, system images, ISO images and private networks
只读id stringorbit compute instances options <id>
compute_instances_options
POST /compute/instances/{id}/action
Power: start, shutdown, stop or reboot
读写id string
请求体: JSON
orbit compute instances action <id>
compute_instances_action
POST /compute/instances/{id}/destroy
Destroy a server and its disk; confirm with its host name
读写id string
请求体: JSON
orbit compute instances destroy <id>
compute_instances_destroy
POST /compute/instances/{id}/console
Open the graphical (vnc) or serial console: a one-time link valid for five minutes
读写id string
请求体: JSON
orbit compute instances console <id>
compute_instances_console
GET /compute/instances/{id}/usage
Hourly network, disk and CPU usage (at most 92 days)
只读id string
?from string
?to string
orbit compute instances usage <id>
compute_instances_usage
GET /compute/instances/{id}/metrics
Graph series: CPU, memory, network and disk over 1h, 1d, 1w or 1m
只读id string
?range string
orbit compute instances metrics <id>
compute_instances_metrics
GET /compute/instances/{id}/jobs
Recent operations on a server
只读id stringorbit compute instances jobs <id>
compute_instances_jobs
POST /compute/instances/{id}/reinstall
Reinstall from a system image (erases the disk); confirm with the host name
读写id string
请求体: JSON
orbit compute instances reinstall <id>
compute_instances_reinstall
POST /compute/instances/{id}/resize
Switch to another package (the server restarts; the disk only grows)
读写id string
请求体: JSON
orbit compute instances resize <id>
compute_instances_resize
POST /compute/instances/{id}/password
Set a new root password; the answer shows it once
读写id string
请求体: JSON
orbit compute instances password <id>
compute_instances_password
POST /compute/instances/{id}/hostname
Change the host name
读写id string
请求体: JSON
orbit compute instances hostname <id>
compute_instances_hostname
POST /compute/instances/{id}/rescue
Start in rescue mode, or leave it ({on: true|false})
读写id string
请求体: JSON
orbit compute instances rescue <id>
compute_instances_rescue
POST /compute/instances/{id}/iso
Insert an ISO image ({iso_id}) or eject it ({iso_id: null})
读写id string
请求体: JSON
orbit compute instances iso <id>
compute_instances_iso
GET /compute/instances/{id}/snapshots
Snapshots of a server
只读id stringorbit compute instances snapshots get <id>
compute_instances_snapshots_get
POST /compute/instances/{id}/snapshots
Take a snapshot
读写id string
请求体: JSON
orbit compute instances snapshots create <id>
compute_instances_snapshots_create
POST /compute/instances/{id}/snapshots/{snap}/rollback
Roll the server back to a snapshot
读写id string
snap string
请求体: JSON
orbit compute instances snapshots rollback <id> <snap>
compute_instances_snapshots_rollback
DELETE /compute/instances/{id}/snapshots/{snap}
Delete a snapshot
读写id string
snap string
orbit compute instances snapshots delete <id> <snap>
compute_instances_snapshots_delete
GET /compute/instances/{id}/backups
Backups of a server
只读id stringorbit compute instances backups get <id>
compute_instances_backups_get
POST /compute/instances/{id}/backups
Back up a server now
读写id string
请求体: JSON
orbit compute instances backups create <id>
compute_instances_backups_create
POST /compute/instances/{id}/backups/{backup}/restore
Restore a backup over the server's disk
读写id string
backup string
请求体: JSON
orbit compute instances backups restore <id> <backup>
compute_instances_backups_restore
DELETE /compute/instances/{id}/backups/{backup}
Delete a backup
读写id string
backup string
orbit compute instances backups delete <id> <backup>
compute_instances_backups_delete
GET /compute/instances/{id}/firewall
The server's firewall: on or off, default policies and rules
只读id stringorbit compute instances firewall get <id>
compute_instances_firewall_get
PUT /compute/instances/{id}/firewall
Replace the firewall settings and rules
读写id string
请求体: JSON
orbit compute instances firewall put <id>
compute_instances_firewall_put
GET /compute/instances/{id}/rdns
Reverse DNS of the server's addresses
只读id stringorbit compute instances rdns get <id>
compute_instances_rdns_get
PUT /compute/instances/{id}/rdns
Set or remove the reverse DNS name of an address ({address, ptr})
读写id string
请求体: JSON
orbit compute instances rdns put <id>
compute_instances_rdns_put
POST /compute/instances/{id}/ipv4
Add an IPv4 address
读写id string
请求体: JSON
orbit compute instances ipv4 add <id>
compute_instances_ipv4_add
DELETE /compute/instances/{id}/ipv4/{assignment}
Remove an added IPv4 address
读写id string
assignment string
orbit compute instances ipv4 remove <id> <assignment>
compute_instances_ipv4_remove
PUT /compute/instances/{id}/security-groups
Set a VPC server's security groups ({ids}, 1 to 5)
读写id string
请求体: JSON
orbit compute instances security_groups put <id>
compute_instances_security_groups_put
POST /compute/instances/{id}/networks
Put the server on a private network ({network_id})
读写id string
请求体: JSON
orbit compute instances networks attach <id>
compute_instances_networks_attach
DELETE /compute/instances/{id}/networks/{network}
Take the server off a private network
读写id string
network string
orbit compute instances networks detach <id> <network>
compute_instances_networks_detach
03

jobs

端点权限参数CLI · MCP 工具
GET /compute/jobs/{id}
One operation: queued, running, done or failed
只读id stringorbit compute jobs get <id>
compute_jobs_get
04

vpcs

端点权限参数CLI · MCP 工具
GET /compute/vpcs
Your VPCs (private networks of a region) with their subnets
只读?region_id stringorbit compute vpcs list
compute_vpcs_list
POST /compute/vpcs
Create a VPC in a region: {region_id, name, size: 20–28 (default 24)}; its range is chosen from the region's private pool and it gets a first subnet
读写请求体: JSONorbit compute vpcs create
compute_vpcs_create
GET /compute/vpcs/{id}
One VPC with its subnets
只读id stringorbit compute vpcs get <id>
compute_vpcs_get
PATCH /compute/vpcs/{id}
Rename a VPC
读写id string
请求体: JSON
orbit compute vpcs patch <id>
compute_vpcs_patch
DELETE /compute/vpcs/{id}
Delete a VPC with no servers in it
读写id stringorbit compute vpcs delete <id>
compute_vpcs_delete
05

subnets

端点权限参数CLI · MCP 工具
POST /compute/vpcs/{id}/subnets
Add a subnet to a VPC: {name, size} (a block of the VPC's range)
读写id string
请求体: JSON
orbit compute subnets create <id>
compute_subnets_create
DELETE /compute/subnets/{id}
Delete a subnet with no servers in it (a VPC keeps at least one)
读写id stringorbit compute subnets delete <id>
compute_subnets_delete
06

eips

端点权限参数CLI · MCP 工具
GET /compute/eips
Your elastic public IPs and what they are bound to
只读?region_id stringorbit compute eips list
compute_eips_list
POST /compute/eips
Allocate an elastic public IP in a region: {region_id, name}
读写请求体: JSONorbit compute eips allocate
compute_eips_allocate
POST /compute/eips/{id}/associate
Bind the elastic IP to a VPC server ({server_id}); bound to another server, it moves in one step
读写id string
请求体: JSON
orbit compute eips associate <id>
compute_eips_associate
POST /compute/eips/{id}/disassociate
Unbind the elastic IP from its server (the server then has no internet access, in or out)
读写id string
请求体: JSON
orbit compute eips disassociate <id>
compute_eips_disassociate
DELETE /compute/eips/{id}
Release an unbound elastic IP
读写id stringorbit compute eips release <id>
compute_eips_release
07

security_groups

端点权限参数CLI · MCP 工具
GET /compute/security-groups
Your security groups with their rules
只读?region_id stringorbit compute security_groups list
compute_security_groups_list
POST /compute/security-groups
Create a security group: {region_id, name, description, policy_in, policy_out, rules: [{direction, action, proto, dport, source (address, network or sg:<id>), comment}]}
读写请求体: JSONorbit compute security_groups create
compute_security_groups_create
GET /compute/security-groups/{id}
One security group
只读id stringorbit compute security_groups get <id>
compute_security_groups_get
PATCH /compute/security-groups/{id}
Change a security group (rules are replaced as a whole); its servers follow
读写id string
请求体: JSON
orbit compute security_groups patch <id>
compute_security_groups_patch
DELETE /compute/security-groups/{id}
Delete a security group no server uses
读写id stringorbit compute security_groups delete <id>
compute_security_groups_delete
08

networks

端点权限参数CLI · MCP 工具
GET /compute/networks
Your private networks
只读—orbit compute networks get
compute_networks_get
POST /compute/networks
Create a private network in a region
读写请求体: JSONorbit compute networks create
compute_networks_create
DELETE /compute/networks/{id}
Delete a private network with no servers on it: 204 when it never reached a host, otherwise 202 while it is removed from the hosts (state deleting)
读写id stringorbit compute networks delete <id>
compute_networks_delete
继续

相关文档

账户 API 参考

Orbit API 中 账户 的全部 10 个端点:方法与路径、所需令牌权限、参数,以及对应的命令行命令与 MCP 工具。

组织 API 参考

Orbit API 中 组织 的全部 33 个端点:方法与路径、所需令牌权限、参数,以及对应的命令行命令与 MCP 工具。

服务 API 参考

Orbit API 中 服务 的全部 15 个端点:方法与路径、所需令牌权限、参数,以及对应的命令行命令与 MCP 工具。

← 全部文档