LangGraph Control Plane LangGraph 控制平面
The term “control plane” is used broadly to refer to the control plane UI where users create and update LangGraph Servers (deployments) and the control plane APIs that support the UI experience.
"控制平面"这一术语被广泛用于指代用户创建和更新 LangGraph 服务器（部署）的控制平面 UI，以及支持 UI 体验的控制平面 API。 When a user makes an update through the control plane UI, the update is stored in the control plane state. The LangGraph Data Plane “listener” application polls for these updates by calling the control plane APIs.
当用户通过控制平面 UI 进行更新时，该更新会存储在控制平面状态中。LangGraph 数据平面“监听”应用程序通过调用控制平面 API 来轮询这些更新。
从控制平面 UI 中，您可以：
控制平面 UI 嵌入在 LangSmith 中。
本节描述控制平面 API 的数据模型。该 API 用于创建、更新和删除部署。有关更多详细信息，请参阅控制平面 API 参考。
部署是 LangGraph 服务器的实例。单个部署可以有多个版本。
修订是一个部署的迭代。当创建新的部署时，会自动创建一个初始修订。要部署代码更改或更新部署的密钥，必须创建新的修订。
本节描述控制平面的各种功能。
为简化操作，控制平面提供两种具有不同资源分配的部署类型：
CPU and memory resources are per replica.
CPU 和内存资源是按副本分配的。
Database disk size for
控制平面与 LangGraph 数据平面“监听”应用程序协同工作，自动为每个部署创建 Postgres 数据库。该数据库作为部署的持久化层。 When implementing a LangGraph application, a checkpointer does not need to be configured by the developer. Instead, a checkpointer is automatically configured for the graph. Any checkpointer configured for a graph will be replaced by the one that is automatically configured.
在实现 LangGraph 应用程序时，开发者无需配置检查点。相反，图会自动配置检查点。为图配置的任何检查点都会被自动配置的检查点替换。 There is no direct access to the database. All access to the database occurs through the LangGraph Server.
无法直接访问数据库。所有对数据库的访问都通过 LangGraph 服务器进行。 The database is never deleted until the deployment itself is deleted.
数据库永远不会被删除，直到部署本身被删除。
部署和修订的基础设施是异步创建和部署的。它们不会在提交后立即部署。目前，部署可能需要几分钟时间。
控制平面与 LangGraph 数据平面“监听”应用程序协同工作，以实现异步部署。
部署完成后，控制平面会监控部署并记录各种指标，例如：
这些指标在控制平面 UI 中以图表形式显示。
每次部署都会自动创建一个 LangSmith 跟踪项目。跟踪项目的名称与部署相同。在创建部署时，无需指定
当部署被删除时，跟踪记录和跟踪项目不会被删除。
"控制平面"这一术语被广泛用于指代用户创建和更新 LangGraph 服务器（部署）的控制平面 UI，以及支持 UI 体验的控制平面 API。 When a user makes an update through the control plane UI, the update is stored in the control plane state. The LangGraph Data Plane “listener” application polls for these updates by calling the control plane APIs.
当用户通过控制平面 UI 进行更新时，该更新会存储在控制平面状态中。LangGraph 数据平面“监听”应用程序通过调用控制平面 API 来轮询这些更新。
Control Plane UI 控制平面 UI
From the control plane UI, you can:从控制平面 UI 中，您可以：
- View a list of outstanding deployments.
查看待处理的部署列表。 - View details of an individual deployment.
查看单个部署的详细信息。 - Create a new deployment.
创建新的部署。 - Update a deployment. 更新部署。
- Update environment variables for a deployment.
更新部署的环境变量。 - View build and server logs of a deployment.
查看部署的构建和服务器日志。 - View deployment metrics such as CPU and memory usage.
查看部署指标，如 CPU 和内存使用情况。 - Delete a deployment. 删除部署。
控制平面 UI 嵌入在 LangSmith 中。
Control Plane API 控制平面 API
This section describes the data model of the control plane API. The API is used to create, update, and delete deployments. See the control plane API reference for more details.本节描述控制平面 API 的数据模型。该 API 用于创建、更新和删除部署。有关更多详细信息，请参阅控制平面 API 参考。
Deployment 部署
A deployment is an instance of a LangGraph Server. A single deployment can have many revisions.部署是 LangGraph 服务器的实例。单个部署可以有多个版本。
Revision 修订
A revision is an iteration of a deployment. When a new deployment is created, an initial revision is automatically created. To deploy code changes or update secrets for a deployment, a new revision must be created.修订是一个部署的迭代。当创建新的部署时，会自动创建一个初始修订。要部署代码更改或更新部署的密钥，必须创建新的修订。
Control Plane Features 控制平面功能
This section describes various features of the control plane.本节描述控制平面的各种功能。
Deployment Types 部署类型
For simplicity, the control plane offers two deployment types with different resource allocations:Development and Production.为简化操作，控制平面提供两种具有不同资源分配的部署类型：
Development 和 Production 。
| Deployment Type 部署类型 | CPU/Memory CPU/内存 | Scaling 扩展 | Database 数据库 |
|---|---|---|---|
| Development 开发 | 1 CPU, 1 GB RAM | Up to 1 replica 最多 1 个副本 | 10 GB disk, no backups 10 GB 磁盘，无备份 |
| Production 生产环境 | 2 CPU, 2 GB RAM | Up to 10 replicas 最多 10 个副本 | Autoscaling disk, automatic backups, highly available (multi-zone configuration) 自动扩展磁盘、自动备份、高可用性（多区域配置） |
CPU 和内存资源是按副本分配的。
Immutable Deployment Type
Once a deployment is created, the deployment type cannot be changed.
不可变的部署类型 一旦创建部署，部署类型不能更改。
不可变的部署类型 一旦创建部署，部署类型不能更改。
Self-Hosted Deployment
Resources for Hybrid and Self-Hosted deployments can be fully customized. Deployment types are only applicable for Cloud deployments.
自托管部署 混合部署和自托管部署的资源可以完全定制。部署类型仅适用于云部署。
自托管部署 混合部署和自托管部署的资源可以完全定制。部署类型仅适用于云部署。
Production 生产
Production type deployments are suitable for “production” workloads. For example, select Production for customer-facing applications in the critical path.Production 类部署适用于“生产”工作负载。例如，对于关键路径中的面向客户的应用程序，选择 Production 。
Resources for Production type deployments can be manually increased on a case-by-case basis depending on use case and capacity constraints. Contact support@langchain.dev to request an increase in resources.Production 类型部署的资源可以根据具体使用场景和容量限制手动增加。如需增加资源，请联系 support@langchain.dev。
Development 开发
Development type deployments are suitable development and testing. For example, select Development for internal testing environments. Development type deployments are not suitable for “production” workloads.Development 类型部署适合开发和测试。例如，可选择 Development 用于内部测试环境。 Development 类型部署不适合“生产”工作负载。
Preemptible Compute Infrastructure
可抢占式计算基础设施上的
这种行为是预期的。可中断计算基础设施显著降低了部署
Development type deployments (API server, queue server, and database) are provisioned on preemptible compute infrastructure. This means the compute infrastructure may be terminated at any time without notice. This may result in intermittent…可抢占式计算基础设施上的
Development 类型部署（API 服务器、队列服务器和数据库）是在可抢占式计算基础设施上配置的。这意味着计算基础设施可能在任何时候无通知被终止。这可能导致间歇性问题……- Redis connection timeouts/errors
Redis 连接超时/错误 - Postgres connection timeouts/errors
Postgres 连接超时/错误 - Failed or retrying background runs
后台运行失败或重试
Development type deployment. By design, LangGraph Server is fault-tolerant. The implementation will automatically attempt to recover from Redis/Postgres connection errors and retry failed background runs.这种行为是预期的。可中断计算基础设施显著降低了部署
Development 类型应用的成本。根据设计，LangGraph Server 具有容错性。实现将自动尝试从 Redis/Postgres 连接错误中恢复，并重试失败的后台运行。Production type deployments are provisioned on durable compute infrastructure, not preemptible compute infrastructure.Production 永久型部署是在耐久型计算基础设施上配置的，而不是在可抢占型计算基础设施上配置的。Development type deployments can be manually increased on a case-by-case basis depending on use case and capacity constraints. For most use cases, TTLs should be configured to manage disk usage. Contact support@langchain.dev to request an increase in resources.Development 类型部署的数据库磁盘大小可以根据具体使用情况和容量限制手动调整。对于大多数使用场景，应配置 TTL 来管理磁盘使用。如需增加资源，请联系 support@langchain.dev。
Database Provisioning 数据库配置
The control plane and LangGraph Data Plane “listener” application coordinate to automatically create a Postgres database for each deployment. The database serves as the persistence layer for the deployment.控制平面与 LangGraph 数据平面“监听”应用程序协同工作，自动为每个部署创建 Postgres 数据库。该数据库作为部署的持久化层。 When implementing a LangGraph application, a checkpointer does not need to be configured by the developer. Instead, a checkpointer is automatically configured for the graph. Any checkpointer configured for a graph will be replaced by the one that is automatically configured.
在实现 LangGraph 应用程序时，开发者无需配置检查点。相反，图会自动配置检查点。为图配置的任何检查点都会被自动配置的检查点替换。 There is no direct access to the database. All access to the database occurs through the LangGraph Server.
无法直接访问数据库。所有对数据库的访问都通过 LangGraph 服务器进行。 The database is never deleted until the deployment itself is deleted.
数据库永远不会被删除，直到部署本身被删除。
A custom Postgres instance can be configured for Hybrid and Self-Hosted deployments.
一个自定义的 Postgres 实例可以配置用于混合部署和自托管部署。
一个自定义的 Postgres 实例可以配置用于混合部署和自托管部署。
Asynchronous Deployment 异步部署
Infrastructure for deployments and revisions are provisioned and deployed asynchronously. They are not deployed immediately after submission. Currently, deployment can take up to several minutes.部署和修订的基础设施是异步创建和部署的。它们不会在提交后立即部署。目前，部署可能需要几分钟时间。
- When a new deployment is created, a new database is created for the deployment. Database creation is a one-time step. This step contributes to a longer deployment time for the initial revision of the deployment.
当创建一个新的部署时，会为该部署创建一个新的数据库。数据库创建是一次性步骤。这个步骤导致部署的初始修订需要更长的部署时间。 - When a subsequent revision is created for a deployment, there is no database creation step. The deployment time for a subsequent revision is significantly faster compared to the deployment time of the initial revision.
当为部署创建后续修订时，没有数据库创建步骤。与初始修订的部署时间相比，后续修订的部署时间显著更快。 - The deployment process for each revision contains a build step, which can take up to a few minutes.
每个版本的部署过程都包含一个构建步骤，这可能需要几分钟的时间。
控制平面与 LangGraph 数据平面“监听”应用程序协同工作，以实现异步部署。
Monitoring 监控
After a deployment is ready, the control plane monitors the deployment and records various metrics, such as:部署完成后，控制平面会监控部署并记录各种指标，例如：
- CPU and memory usage of the deployment.
部署的 CPU 和内存使用情况。 - Number of container restarts.
容器重启次数。 - Number of replicas (this will increase with autoscaling).
副本数量（这将随着自动扩展而增加）。 - Postgres CPU, memory usage, and disk usage.
Postgres CPU、内存使用量和磁盘使用量。 - LangGraph Server queue pending/active run count.
LangGraph 服务器队列中待处理/活跃运行计数。 - LangGraph Server API success response count, error response count, and latency.
LangGraph 服务器 API 成功响应计数、错误响应计数和延迟。
这些指标在控制平面 UI 中以图表形式显示。
LangSmith Integration LangSmith 集成
A LangSmith tracing project is automatically created for each deployment. The tracing project has the same name as the deployment. When creating a deployment, theLANGCHAIN_TRACING and LANGSMITH_API_KEY/LANGCHAIN_API_KEY environment variables do not need to be specified; they are set automatically by the control plane.每次部署都会自动创建一个 LangSmith 跟踪项目。跟踪项目的名称与部署相同。在创建部署时，无需指定
LANGCHAIN_TRACING 和 LANGSMITH_API_KEY / LANGCHAIN_API_KEY 环境变量；它们会由控制平面自动设置。
When a deployment is deleted, the traces and the tracing project are not deleted.当部署被删除时，跟踪记录和跟踪项目不会被删除。