Error Codes

New disaster-server APIs should use the unified response envelope. Older endpoints or authentication middleware may still return msg; clients should handle both.

Response Envelope

{
  "code": 0,
  "message": "OK",
  "data": {},
  "meta": {},
  "trace_id": "..."
}

Error response:

{
  "code": 1000,
  "message": "invalid request",
  "data": null,
  "meta": {},
  "trace_id": "..."
}

Business Codes

code	HTTP Status	Meaning	Suggested Handling
0	200	Success	Read data and meta
1000	400	Bad request	Check body, path params, query params
2001	401	Unauthorized or invalid token	Login again or refresh token
2003	403	Forbidden	Check user permissions, Server RBAC, remote cluster permissions
3004	404	Resource not found	Check name, namespace, cluster, route
3009	409	State conflict	Check whether the current state allows the action
4000	502	Upstream error	Check Kubernetes API, Velero, object storage, or remote cluster
5000	500	Internal server error	Use trace_id to search server logs

Authentication Middleware Response

Authentication failures may return:

{
  "code": 401,
  "msg": "token is invalid"
}

Clients should handle both message and msg.

Troubleshooting

• 400: request schema or field names.
• 401: token expiry or missing Bearer prefix.
• 403: user permissions, ServiceAccount, remote cluster RBAC.
• 404: resource name and API group/version.
• 409: instance or operation state machine.
• 502: Kubernetes API, Velero, object storage, network.
• 500: use trace_id and server logs.