REST API 现已开始进行版本化。有关详细信息，请参阅“关于 API 版本控制”。

用于管理 GitHub Enterprise Server 的 REST API 终结点

使用 REST API 管理 GitHub Enterprise Server 实例。

关于管理 GitHub Enterprise Server API

可以使用管理你的 GitHub Enterprise Server 实例 API 来管理GitHub Enterprise Server。例如，可以检索有关实例上运行的软件版本 GitHub Enterprise Server 的信息，或者在具有多个节点的实例上，查看复制的状态。

提示

可以使用此 API 替换在版本 3.15 中删除**** 的GitHub Enterprise Server 的功能。

在对管理 GitHub Enterprise Server API 的终结点进行 API 调用时指定端口号。如果实例使用 TLS，则端口号为 8443。否则，端口号为 8080。如果无法提供端口号，则需要将客户端配置为自动遵循重定向。有关详细信息，请参阅“配置 TLS”。

还可以使用 GitHub Enterprise Server 的 GitHub CLI 扩展在“管理 GitHub Enterprise Server API”中调用终结点。有关详细信息，请参阅 github/gh-es 存储库。

身份验证

若要对管理 GitHub Enterprise Server API 的终结点的请求进行身份验证，请将实例的根站点管理员帐户的密码指定为身份验证令牌。使用标准 HTTP 身份验证发送密码。 api_key 用户标识根站点管理员。以下示例演示此 API 的身份验证。将 ROOT-SITE-ADMINISTRATOR-PASSWORD 替换为密码，将 ADMINISTRATION-PORT 替换为 8443 或 8080。

curl -L -u "api_key:ROOT-SITE-ADMINISTRATOR-PASSWORD" 'http(s)://HOSTNAME:ADMINISTRATION-PORT/manage'

以管理控制台用户身份进行身份验证

管理控制台用户帐户还可以进行身份验证以访问这些终结点。有关详细信息，请参阅“管理对管理控制台的访问”。

若要使用管理控制台用户帐户的密码进行身份验证，请使用标准 HTTP 身份验证。在以下示例中，将 YOUR_USER_NAME 和 YOUR_PASSWORD 替换为帐户的用户名和密码。

curl -L -u "YOUR_USER_NAME:YOUR_PASSWORD" 'http(s)://HOSTNAME:ADMINISTRATION-PORT/manage'

查询参数

默认情况下，响应包括有关实例的所有已配置节点的信息。在具有多个节点的实例上，详细信息源自 /data/user/common/cluster.conf。可以使用以下查询参数来筛选响应，以获取有关特定节点的信息。

查询参数	说明
`uuid`	节点的唯一标识符。
`cluster_role`	群集中的节点所具备的角色。有关详细信息，请参阅“关于集群节点”。

可以通过用逗号分隔值来为查询参数指定多个值。例如，以下请求使用 curl 返回具有 web-server 或 storage-server 角色的任何节点。

curl -L -u "api_key:ROOT-SITE-ADMINISTRATOR-PASSWORD" 'http(s)://HOSTNAME:ADMINISTRATION-PORT/manage/v1/config/nodes?cluster_role=WebServer,StorageServer'

Get the configured SSH keys

Gets the configured SSH keys on all available nodes. For more information, see "Accessing the administrative shell (SSH)."

“Get the configured SSH keys”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized

“Get the configured SSH keys”的代码示例

请求示例

get/manage/v1/access/ssh

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/access/ssh

Response

Status: 200

[
  {
    "key": "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAAAgQCwcd9tu20xsJVBMWbIs+JDVQTvkLtrJ8A7eblSwC+zECver22QExC9d6zHy10MAyk1Ck7Bu6/0Z+rr/31vMcAOmNOLjExzWxBCXbtDf3758Qfw7FuvkaTE1sHztwSFi/yNhZSw7uPEWeGQiRY4UldSKG9zZvrZmxpnP8mNZzPriQ==",
    "fingeprint": "bd:26:d3:a2:ad:97:c0:7c:2d:57:a4:64:f3:fb:44:45"
  },
  {
    "key": "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAAAgQCg73sgCdLL9rWcmVTmvcOgAWZaokZTeNnMFEoaZWZcVtsP4Ed+ICuSbT1wDuyL/XXzr0O2u2bKwx8Np/3UGHb4R9Re+qiz//l5OCBd3KonO2GFICQwKGmVeZ+ki89aN2JDKgfemHbvclHYBD/r56lnbt/kinw7JqGPp+ndzH8nBw==",
    "fingeprint": "e1:1e:a0:f1:3f:a9:70:2d:99:dd:02:9d:39:1c:8b:a4"
  }
]

Set a new SSH key

Adds a SSH key to the authorized_keys file for your GitHub Enterprise Server instance. This will grant access via SSH to your instance. For more information, see "Accessing the administrative shell (SSH)."

“”Set a new SSH key 的参数

主体参数
名称, 类型, 说明
`key` string 必须 The public SSH key to add to the `authorized_keys` file.

“Set a new SSH key”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Set a new SSH key”的代码示例

请求示例

post/manage/v1/access/ssh

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/access/ssh \
  -d '{"key":"ssh-rsa AAAAB3NzaC1yc2EAAAADCIABAAAAgQCY/ZiDDOFWcZnYXPwMbvwQDofXPdHxLfxPK+HWGVPd1DLcDncYBUSB0bmCU2g9Sc+oHKLoHhXp0ivau9h+EpmQJ7V8vqsRdD9pc4aL/WAnUyF4o3Y7xL94rlRpVbVo/tNjzcvqxxyzBiYyy3GciCMpYQh/uKt56B94/5PNyIGEEw=="}'

Response

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "message": "SSH key added successfully",
    "modified": true
  }
]

Delete a SSH key

Deletes a SSH key from the authorized_keys file for your GitHub Enterprise Server instance. This will remove access via SSH to your instance. For more information, see "Accessing the administrative shell (SSH)."

“”Delete a SSH key 的参数

主体参数
名称, 类型, 说明
`key` string 必须 The public SSH key to remove from the `authorized_keys` file.

“Delete a SSH key”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Delete a SSH key”的代码示例

请求示例

delete/manage/v1/access/ssh

curl -L \
  -X DELETE \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/access/ssh \
  -d '{"key":"ssh-rsa AAAAB3NzaC1yc2EAAAADCIABAAAAgQCY/ZiDDOFWcZnYXPwMbvwQDofXPdHxLfxPK+HWGVPd1DLcDncYBUSB0bmCU2g9Sc+oHKLoHhXp0ivau9h+EpmQJ7V8vqsRdD9pc4aL/WAnUyF4o3Y7xL94rlRpVbVo/tNjzcvqxxyzBiYyy3GciCMpYQh/uKt56B94/5PNyIGEEw=="}'

Response

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "message": "SSH key removed successfully"
  }
]

Get the system requirement check results for configured cluster nodes

Checks if the minimum requirements for system hardware resources are met on each configured cluster node. This endpoint may take several seconds to reply.

“Get the system requirement check results for configured cluster nodes”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`500`	Internal error

“Get the system requirement check results for configured cluster nodes”的代码示例

请求示例

get/manage/v1/checks/system-requirements

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/checks/system-requirements

Response

Status: 200

{
  "status": "OK",
  "nodes": [
    {
      "hostname": "ghe-local-app",
      "status": "OK",
      "roles_status": [
        {
          "status": "OK",
          "role": "ConsulServer"
        },
        {
          "status": "OK",
          "role": "JobServer"
        },
        {
          "status": "OK",
          "role": "WebServer"
        }
      ]
    },
    {
      "hostname": "ghe-local-app2",
      "status": "OK",
      "roles_status": [
        {
          "status": "OK",
          "role": "ConsulServer"
        },
        {
          "status": "OK",
          "role": "GitServer"
        }
      ]
    }
  ]
}

Get the status of services running on all cluster nodes

Gets the status of all services running on each cluster node. This endpoint may take several seconds to reply.

“Get the status of services running on all cluster nodes”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Get the status of services running on all cluster nodes”的代码示例

请求示例

get/manage/v1/cluster/status

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/cluster/status

Response

Status: 200

{
  "status": "OK",
  "nodes": [
    {
      "hostname": "ghe-local-app",
      "status": "OK",
      "services": [
        {
          "status": "OK",
          "name": "es",
          "details": "Elasticsearch cluster is in sync (0 shards initializing, 0 shards unassigned)"
        },
        {
          "status": "OK",
          "name": "git-replication",
          "details": "Git replication is in sync"
        },
        {
          "status": "OK",
          "name": "kafka-lite-admin-healthcheck",
          "details": ""
        },
        {
          "status": "OK",
          "name": "kafka-lite-broker-is-reachable",
          "details": ""
        },
        {
          "status": "OK",
          "name": "memcache",
          "details": ""
        },
        {
          "status": "OK",
          "name": "metrics",
          "details": ""
        },
        {
          "status": "OK",
          "name": "mysql-replication",
          "details": "Replication is running"
        },
        {
          "status": "OK",
          "name": "mysql-failover",
          "details": ""
        },
        {
          "status": "OK",
          "name": "pages",
          "details": "Pages replication is in sync"
        },
        {
          "status": "OK",
          "name": "redis",
          "details": "Redis is OK"
        },
        {
          "status": "OK",
          "name": "storage",
          "details": "Storage replication is in sync"
        }
      ]
    },
    {
      "hostname": "ghe-local-app2",
      "status": "OK",
      "services": [
        {
          "status": "OK",
          "name": "kafka-lite-admin-healthcheck",
          "details": ""
        },
        {
          "status": "OK",
          "name": "kafka-lite-broker-is-reachable",
          "details": ""
        }
      ]
    }
  ]
}

Get the status of a ghe-config-apply run

Displays the current status of ghe-config-apply in the environment or the status of a historical run by ID.

“”Get the status of a ghe-config-apply run 的参数

查询参数
名称, 类型, 说明
`run_id` string The unique run ID of the `ghe-config-apply` run.

“Get the status of a ghe-config-apply run”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized

“Get the status of a ghe-config-apply run”的代码示例

请求示例

get/manage/v1/config/apply

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/apply

Response

Status: 200

{
  "running": true,
  "successful": false,
  "nodes": [
    {
      "run_id": "d34db33f",
      "hostname": "ghes-01.lan",
      "running": true,
      "successful": false
    }
  ]
}

Trigger a ghe-config-apply run

Triggers a run of ghe-config-apply from the ghes-manage agent on your Nomad Delegate instance. You can provide a run ID or allow one to be generated randomly.

“”Trigger a ghe-config-apply run 的参数

主体参数
名称, 类型, 说明
`run_id` string The run ID to execute `ghe-config-apply` with. If not provided, a run ID will be generated randomly.

“Trigger a ghe-config-apply run”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized

“Trigger a ghe-config-apply run”的代码示例

请求示例

post/manage/v1/config/apply

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/apply \
  -d '{"run_id":"d34db33f"}'

Response

Status: 200

{
  "run_id": "d34db33f"
}

List events from ghe-config-apply

Lists events from an in-process ghe-config-apply run on your Github Enterprise Server instance.

“”List events from ghe-config-apply 的参数

查询参数
名称, 类型, 说明
`last_request_id` string The unique ID of the last response from a host, used for pagination.

“List events from ghe-config-apply”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized

“List events from ghe-config-apply”的代码示例

请求示例

get/manage/v1/config/apply/events

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/apply/events

Response

Status: 200

{
  "nodes": [
    {
      "node": "ghes-01.lan",
      "last_request_id": "387cd628c06d606700e79be368e5e574:0cde553750689c76:0000000000000000",
      "events": [
        {
          "timestamp": "2023-01-01T13:00:00+00:00",
          "severity_text": "INFO",
          "body": "Validating services",
          "event_name": "Enterprise::ConfigApply::PhaseValidation#config_phase_validation",
          "topology": "multinode",
          "hostname": "ghes-01.lan",
          "config_run_id": "d34db33f",
          "trace_id": "387cd628c06d606700e79be368e5e574",
          "span_id": "0cde553750689c76",
          "span_parent_id": 0,
          "span_depth": 0
        }
      ]
    }
  ]
}

Initialize instance configuration with license and password

When you boot and set up a GitHub instance for the first time, you can use this endpoint to upload a license and set the initial root site administrator password.

Important

To start the configuration process and apply the license, you need to POST to /manage/v1/config/apply

The root site administrator password provided when calling this endpoint is used to authenticate for all other endpoints in the GHES Manage API and the Management Console UI.

Note

The request body for this operation must be submitted as multipart/form-data data. You can can reference the license file by prefixing the filename with the @ symbol using curl. For more information, see the curl documentation.

“”Initialize instance configuration with license and password 的参数

主体参数
名称, 类型, 说明
`license` string 必须 The content of your .ghl license file.
`password` string 必须 The root site administrator password.

“Initialize instance configuration with license and password”的 HTTP 响应状态代码

状态代码	说明
`202`	Accepted
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Initialize instance configuration with license and password”的代码示例

请求示例

post/manage/v1/config/init

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: multipart/form-data" \
  http(s)://HOSTNAME/manage/v1/config/init \
   --form 'license=@enterprise.ghl' --form 'password=provide-password-here!'

Response

Status: 202

Get the enterprise license information

Gets information about the license that is currently set for the enterprise.

“Get the enterprise license information”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Get the enterprise license information”的代码示例

请求示例

get/manage/v1/config/license

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/license

Response

Status: 200

[
  {
    "advancedSecurityEnabled": true,
    "advancedSecuritySeats": 0,
    "clusterSupport": false,
    "company": "GitHub",
    "croquetSupport": true,
    "customTerms": true,
    "evaluation": false,
    "expireAt": "2025-01-02T07:59:59Z",
    "insightsEnabled": true,
    "insightsExpireAt": "2025-01-02T07:59:59.999Z",
    "learningLabEvaluationExpires": "2023-01-02T07:59:59Z",
    "learningLabSeats": 100,
    "perpetual": false,
    "referenceNumber": "32a145",
    "seats": 0,
    "sshAllowed": true,
    "supportKey\"": "",
    "unlimitedSeating": true
  }
]

Upload an enterprise license

Uploads a new enterprise license. In order to apply it right away, use the apply query parameter.

Note

“”Upload an enterprise license 的参数

查询参数
名称, 类型, 说明
`apply` boolean Whether to instantly apply changes from the license. Otherwise the new license can be applied using the `/manage/v1/config/apply` endpoint.

主体参数
名称, 类型, 说明
`license` string 必须 The content of your .ghl license file.

“Upload an enterprise license”的 HTTP 响应状态代码

状态代码	说明
`201`	Created
`202`	Accepted
`401`	Unauthorized
`500`	Internal error

“Upload an enterprise license”的代码示例

请求示例

Select the example type

put/manage/v1/config/license

curl -L \
  -X PUT \
  -u "api_key:your-password" \
  -H "Content-Type: multipart/form-data" \
  http(s)://HOSTNAME/manage/v1/config/license \
   --form 'license=@enterprise.ghl'

Created

Status: 201

Check a license

Check the status of the license that is currently set for the enterprise.

“Check a license”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Check a license”的代码示例

请求示例

get/manage/v1/config/license/check

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/license/check

Response

Status: 200

[
  {
    "status": "valid"
  }
]

Get GHES node metadata for all nodes

Get node metadata for all configured nodes in the current cluster. For more information, see "About clustering."

“”Get GHES node metadata for all nodes 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.
`cluster_roles` string The cluster roles from the cluster configuration file.

“Get GHES node metadata for all nodes”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Get GHES node metadata for all nodes”的代码示例

请求示例

get/manage/v1/config/nodes

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/nodes

Response

Status: 200

{
  "topology": "Cluster",
  "nodes": [
    {
      "hostname": "data1",
      "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
      "cluster_roles": [
        "ConsulServer",
        "ElasticsearchServer",
        "GitServer",
        "StorageServer"
      ]
    },
    {
      "hostname": "data2",
      "uuid": "228406d4-f97c-11ed-ab01-062281bbcf03",
      "cluster_roles": [
        "ElasticsearchServer",
        "StorageServer",
        "PagesServer"
      ]
    }
  ]
}

Get the GHES settings

Gets a list of settings for a GitHub Enterprise Server instance.

“Get the GHES settings”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized

“Get the GHES settings”的代码示例

请求示例

get/manage/v1/config/settings

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/settings

Response

Status: 200

{
  "private_mode": false,
  "public_pages": false,
  "subdomain_isolation": true,
  "signup_enabled": false,
  "github_hostname": "ghe.local",
  "identicons_host": "dotcom",
  "auth_mode": "default",
  "expire_sessions": false,
  "configuration_id": 1401777404,
  "configuration_run_count": 4,
  "avatar": {
    "enabled": false,
    "uri": ""
  },
  "customer": {
    "name": "GitHub",
    "email": "stannis@themannis.biz",
    "uuid": "af6cac80-e4e1-012e-d822-1231380e52e9",
    "secret_key_data": "-----BEGIN PGP PRIVATE KEY BLOCK-----\nVersion: GnuPG v1.4.10 (GNU/Linux)\nlQcYBE5TCgsBEACk4yHpUcapplebaumBMXYMiLF+nCQ0lxpx...\n-----END PGP PRIVATE KEY BLOCK-----\n",
    "public_key_data": "-----BEGIN PGP PUBLIC KEY BLOCK-----\nVersion: GnuPG v1.4.10 (GNU/Linux)\nmI0ETqzZYgEEALSe6snowdenXyqvLfSQ34HWD6C7....\n-----END PGP PUBLIC KEY BLOCK-----\n"
  },
  "license": {
    "seats": 0,
    "evaluation": false,
    "perpetual": false,
    "unlimited_seating": true,
    "support_key": "ssh-rsa AAAAB3N....",
    "ssh_allowed": true,
    "cluster_support": false,
    "expire_at": "2016-04-27T00:00:00-07:00"
  },
  "github_ssl": {
    "enabled": false,
    "cert": null,
    "key": null
  },
  "ldap": {
    "host": null,
    "port": 0,
    "base": [],
    "uid": null,
    "bind_dn": null,
    "password": null,
    "method": "Plain",
    "search_strategy": "detect",
    "user_groups": [],
    "admin_group": null,
    "virtual_attribute_enabled": false,
    "recursive_group_search": false,
    "posix_support": true,
    "user_sync_emails": false,
    "user_sync_keys": false,
    "user_sync_interval": 4,
    "team_sync_interval": 4,
    "sync_enabled": false,
    "reconciliation": {
      "user": null,
      "org": null
    },
    "profile": {
      "uid": "uid",
      "name": null,
      "mail": null,
      "key": null
    }
  },
  "cas": {
    "url": null
  },
  "saml": {
    "sso_url": null,
    "certificate": null,
    "certificate_path": null,
    "issuer": null,
    "idp_initiated_sso": false,
    "disable_admin_demote": false
  },
  "github_oauth": {
    "client_id": "12313412",
    "client_secret": "kj123131132",
    "organization_name": "Homestar Runners",
    "organization_team": "homestarrunners/characters"
  },
  "smtp": {
    "enabled": true,
    "address": "smtp.example.com",
    "authentication": "plain",
    "port": "1234",
    "domain": "blah",
    "username": "foo",
    "user_name": "mr_foo",
    "enable_starttls_auto": true,
    "password": "bar",
    "discard-to-noreply-address": true,
    "support_address": "enterprise@github.com",
    "support_address_type": "email",
    "noreply_address": "noreply@github.com"
  },
  "ntp": {
    "primary_server": "0.pool.ntp.org",
    "secondary_server": "1.pool.ntp.org"
  },
  "snmp": {
    "enabled": false,
    "community": ""
  },
  "syslog": {
    "enabled": false,
    "server": null,
    "protocol_name": "udp"
  },
  "pages": {
    "enabled": true
  },
  "collectd": {
    "enabled": false,
    "server": null,
    "port": 0,
    "encryption": null,
    "username": null,
    "password": null
  },
  "mapping": {
    "enabled": true,
    "tileserver": null,
    "basemap": "company.map-qsz2zrvs",
    "token": null
  },
  "prometheus": {
    "enabled": false,
    "trusted_ips": "10.0.0.1, 192.168.1.0/8"
  }
}

Set settings

Updates the settings on your instance. For a list of the available settings, see the Get settings endpoint.

Notes:

The request body only requires the settings parameters that should be updated to be specified, all other parameters will be unmodified or populated from the default values.
You cannot set the Management Console root site administrator password with this API endpoint. Use the ghe-set-password utility to change the management console password. For more information, see "Command-line utilities."

“Set settings”的 HTTP 响应状态代码

状态代码	说明
`204`	No Content
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Set settings”的代码示例

请求示例

put/manage/v1/config/settings

curl -L \
  -X PUT \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/config/settings \
  -d '{"public_pages":true}'

Response

Status: 204

Get the status of maintenance mode

Gets the status and details of maintenance mode on all available nodes. For more information, see "Enabling and scheduling maintenance mode."

“”Get the status of maintenance mode 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.
`cluster_roles` string The cluster roles from the cluster configuration file.

“Get the status of maintenance mode”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Get the status of maintenance mode”的代码示例

请求示例

get/manage/v1/maintenance

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/maintenance

Response

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "status": "scheduled",
    "scheduled_time": "2006-01-02T15:04:05+00:00",
    "connection_services": [
      {
        "name": "git operations",
        "number": 15
      },
      {
        "name": "mysql queries",
        "number": 6
      },
      {
        "name": "resque jobs",
        "number": 10
      },
      {
        "name": "aqueduct jobs",
        "number": 0
      }
    ],
    "can_unset_maintenance": true,
    "ip_exception_list": [
      "1.1.1.1"
    ],
    "maintenance_mode_message": "Scheduled maintenance for upgrading."
  }
]

Set the status of maintenance mode

Sets or schedules the maintenance mode. For more information, see "Enabling and scheduling maintenance mode."

“”Set the status of maintenance mode 的参数

主体参数
名称, 类型, 说明
`enabled` boolean 必须 Whether to enable maintenance mode.
`uuid` string The UUID of the node to target. This parameter is incompatible with maintenance mode scheduling. Only use `uuid` if the value of `when` is empty or `now`.
`when` string The time to enable maintenance mode. If this parameter is empty or set to `now`, maintenance mode is enabled immediately. Otherwise, maintenance mode is enabled at the specified time. The format is ISO 8601.
`ip_exception_list` array of strings The list of IP addresses to exclude from maintenance mode. IPv4, IPv6, and CIDR addresses are supported.
`maintenance_mode_message` string The message to display to users when maintenance mode is enabled.

“Set the status of maintenance mode”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Set the status of maintenance mode”的代码示例

请求示例

post/manage/v1/maintenance

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/maintenance \
  -d '{"enabled":true,"when":"2006-01-02T15:04:05+00:00","ip_exception_list":["192.168.1.0/24","1.1.1.1"]}'

Response

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "message": "maintenance mode scheduled with exception list [1.1.1.1]"
  }
]

Get the status of services running on all replica nodes

Gets the status of all services running on each replica node. This endpoint may take several seconds to reply.

“”Get the status of services running on all replica nodes 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.
`cluster_roles` string The cluster roles from the cluster configuration file.

“Get the status of services running on all replica nodes”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Get the status of services running on all replica nodes”的代码示例

请求示例

get/manage/v1/replication/status

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/replication/status

Response

Status: 200

{
  "status": "OK",
  "nodes": [
    {
      "hostname": "ghe-local-primary",
      "status": "OK",
      "services": []
    },
    {
      "hostname": "ghe-local-replica",
      "status": "OK",
      "services": [
        {
          "status": "OK",
          "name": "redis",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "elasticsearch",
          "details": "cluster is in sync (0 shards initializing, 0 shards unassigned)"
        },
        {
          "status": "OK",
          "name": "git",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "pages",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "alambic",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "git-hooks",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "consul",
          "details": "replication is in sync"
        },
        {
          "status": "OK",
          "name": "mysql",
          "details": "replication is in sync"
        }
      ]
    }
  ]
}

Apply an upgrade

Applies an upgrade package on a GitHub Enterprise Server node. You can run a full upgrade (pre-upgrade + upgrade), or a specific phase. Either upgrade_package_file or version is required. When phase is omitted, both pre-upgrade and upgrade phases run sequentially.

“”Apply an upgrade 的参数

主体参数
名称, 类型, 说明
`upgrade_package_file` string Absolute path to the upgrade package file. Must be under `/var/lib/ghe-updates/` and the path must be canonical (no `..` or duplicate slashes).
`version` string Target upgrade version in semantic version format (e.g. `3.19.3`). The version is normalized to `major.minor.patch` format.
`target_partition` string Target block device partition for the upgrade.
`phase` string Upgrade phase to run. Valid values are `pre-upgrade` or `upgrade`. Omit for a full upgrade. 可以是以下选项之一: `pre-upgrade`, `upgrade`
`skip_reboot` boolean Whether to skip the reboot after the upgrade completes. Default is `false`. 默认: `false`
`uuid` string The UUID of a specific node to upgrade. When omitted, the upgrade is applied to all nodes.

“Apply an upgrade”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`404`	Not found
`500`	Internal error

“Apply an upgrade”的代码示例

请求示例

post/manage/v1/upgrade/apply

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  https://HOSTNAME/manage/v1/upgrade/apply \
  -d '{"version":"3.19.3"}'

OK

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "message": "upgrade started",
    "error": ""
  }
]

Download an upgrade package

Starts an asynchronous GitHub Enterprise Server upgrade package download on the targeted node set.

On single-node and high availability topologies, the gateway fans out directly to the targeted nodes and returns a 200 OK response containing one result per node, unless an orchestration option is provided. On cluster topologies, or whenever max_parallel, node_timeout, or max_attempts is provided, the gateway starts an in-memory throttled orchestration and returns 202 Accepted.

If version is omitted, each agent resolves the latest eligible upgrade package version for that node. The download runs in the background; use the upgrade download status endpoint to monitor progress.

“”Download an upgrade package 的参数

主体参数
名称, 类型, 说明
`version` string Semantic version of the upgrade package to download. If omitted, the agent resolves the latest eligible version.
`force` boolean Whether to remove an existing package file before starting a new download. 默认: `false`
`uuids` array of strings Node UUIDs to target after gateway query filters are applied. Blank values are ignored; if all supplied values are blank, the request is rejected.
`max_parallel` integer Selects gateway-side orchestration and limits the number of nodes downloading at the same time. Non-positive values use the gateway default, and values above the target node count are capped to that count. 默认: `2`
`node_timeout` string Selects gateway-side orchestration and marks an in-flight node download as stuck after this Go duration. Defaults to 30m and must be at most 2h. 默认: `30m0s`
`max_attempts` integer Selects gateway-side orchestration and sets the maximum number of attempts per node. Non-positive values use the gateway default. The maximum accepted value is 5. 默认: `3`

“Download an upgrade package”的 HTTP 响应状态代码

状态代码	说明
`200`	Direct fan-out accepted the download request on the targeted nodes. Individual node failures are returned in each node result and do not change the HTTP status code.
`202`	Gateway-side throttled orchestration was accepted and started in the background.
`400`	Bad request
`401`	Unauthorized
`404`	Not found
`409`	Download orchestration could not start because another orchestration is running, or because a targeted node is already downloading a package.
`500`	Internal error

“Download an upgrade package”的代码示例

请求示例

post/manage/v1/upgrade/download

curl -L \
  -X POST \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  https://HOSTNAME/manage/v1/upgrade/download \
  -d '{"version":"3.19.3","force":false}'

Direct fan-out accepted the download request on the targeted nodes. Individual node failures are returned in each node result and do not change the HTTP status code.

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "message": "download started",
    "error": ""
  }
]

Get the upgrade package download status

Returns the aggregate and per-node status for GitHub Enterprise Server upgrade package downloads.

While a gateway-side throttled orchestration is running, the response includes an orchestration object with queue-level counts and the status is scoped to the gateway-filtered node set. When no orchestration is active, the gateway fans out to the targeted agents and aggregates their per-node download status.

Use uuid query filters to inspect a subset of nodes.

“”Get the upgrade package download status 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.

“Get the upgrade package download status”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Get the upgrade package download status”的代码示例

请求示例

Select the example type

get/manage/v1/upgrade/download/status

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  https://HOSTNAME/manage/v1/upgrade/download/status

OK

Status: 200

{
  "upgrade_id": "550e8400-e29b-41d4-a716-446655440000",
  "package_name": "github-enterprise-3.19.3-kvm.pkg",
  "status": "downloading",
  "error_message": "",
  "distribution_progress": {
    "completed_nodes": 2,
    "total_nodes": 5
  },
  "nodes": [
    {
      "hostname": "node-1",
      "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
      "upgrade_id": "550e8400-e29b-41d4-a716-446655440000",
      "package_name": "github-enterprise-3.19.3-kvm.pkg",
      "status": "completed",
      "progress": "100%",
      "error": ""
    },
    {
      "hostname": "node-2",
      "uuid": "228406d4-f97c-11ed-ab01-062281bbcf03",
      "upgrade_id": "550e8400-e29b-41d4-a716-446655440000",
      "package_name": "github-enterprise-3.19.3-kvm.pkg",
      "status": "downloading",
      "progress": "63%",
      "error": ""
    }
  ]
}

Get the upgrade status

Gets the upgrade status from all configured nodes of a GitHub Enterprise Server instance. Returns information about the current upgrade phase, completion status, and individual step details when is_verbose mode is enabled.

“”Get the upgrade status 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.
`is_verbose` boolean Whether to include detailed step information.

“Get the upgrade status”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`400`	Bad request
`401`	Unauthorized
`500`	Internal error

“Get the upgrade status”的代码示例

请求示例

get/manage/v1/upgrade/status

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  https://HOSTNAME/manage/v1/upgrade/status

OK

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "uuid": "1b6cf518-f97c-11ed-8544-061d81f7eedb",
    "release_version": "3.19.2",
    "release_sha": "abc123def456",
    "status": "in_progress",
    "pre_upgrade_status": "completed",
    "upgrade_last_completed_step": "extract_root_filesystem",
    "is_running": true,
    "failed": false,
    "steps": [
      {
        "phase": "pre-upgrade",
        "step": "signature_verification",
        "status": "completed",
        "started_at": "2024-06-02T10:30:00Z",
        "ended_at": "2024-06-02T10:31:00Z",
        "error": ""
      }
    ],
    "messages": [],
    "error": ""
  }
]

Get all GHES release versions for all nodes

Gets the GitHub Enterprise Server release versions that are currently installed on all available nodes. For more information, see "GitHub Enterprise Server releases."

“”Get all GHES release versions for all nodes 的参数

查询参数
名称, 类型, 说明
`uuid` string The UUID which identifies a node.
`cluster_roles` string The cluster roles from the cluster configuration file.

“Get all GHES release versions for all nodes”的 HTTP 响应状态代码

状态代码	说明
`200`	OK
`401`	Unauthorized
`500`	Internal error

“Get all GHES release versions for all nodes”的代码示例

请求示例

get/manage/v1/version

curl -L \
  -u "api_key:your-password" \
  -H "Content-Type: application/json" \
  http(s)://HOSTNAME/manage/v1/version

Response

Status: 200

[
  {
    "hostname": "ghe-local-primary",
    "version": {
      "version": "3.9.0",
      "platform": "azure",
      "build_id": "fc542058b5",
      "build_date": "2023-05-02"
    }
  }
]