本页面为机器翻译。阅读英文原文。

Component

core/ai-pipeline/temporal

Temporal, a durable-execution engine. A multi-step job is written as ordinary program code against one of the project's client libraries; the engine records every step's outcome as an event and hands the next piece of work to a worker the application itself deploys. Because the run is a replayable event history rather than a position in a process, a worker that crashes mid-job is replaced and the job resumes where it stopped, and a step that fails is retried on the policy the code declared. That division is what separates this from a scheduler: the engine owns the run's history and its timers, while the application owns the code each step executes. It keeps none of its own durable state. A relational datastore holds the event histories and a second database on that same connection holds the visibility index the console and the list APIs read, so the store is wired as a topology edge to a datastore component the Aggregate already declares rather than bundled inside this one. The engine is reached over gRPC on the frontend service, which is the address the provides contract publishes; a browser console is deployed alongside it. Realized on Kubernetes through a Helm Recipe that sources the project's own chart. It has no standalone local-container form, so no Compose Recipe is provided. Each of the engine's services serves its own metrics, in pods of its own, so the Recipe publishes the pod labels selecting every one of them, the name of their metrics port and the path, for a scraper that discovers those pods directly.

来源

官方由本目录认定为官方发布者的签名身份签名。
内容摘要
sha256:60d848e16a196199e572a931bf940dc5c539909ce68a5fb35ed51a0c51fa1881
发布时间
2026-11-10T10:09:45Z
签名身份
https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

签名

  • 官方application/vnd.dev.sigstore.bundle.v0.3+json

    https://token.actions.githubusercontent.com/https://github.com/aggrega-ai/aggrega-hub/.github/workflows/catalog-publish-sign.yml@refs/heads/main

    sha256:b73cbac98a2f203410f5c676e05c155fa82c0f89872211a0655912026fe5d7fa

此版本的引用

oci://hub.aggrega.ai/component/core/ai-pipeline/temporal:1.0.0

开发者参数

开发者在 Aggregate 中设置的值。

名称类型默认值必需描述
console_enabledbooleantrue否Deploy the browser console beside the engine. The console reads the visibility index to list and inspect runs; the engine serves its gRPC API with or without it, so turning it off removes a browser surface rather than a capability.
admin_tools_enabledbooleantrue否Deploy the administrative command-line image as a long-running pod an operator can exec into to inspect namespaces, describe a run, or apply a schema change. Turning it off removes that entry point; it serves no application traffic either way.

运维规格

运维人员按环境调整的基础设施规格。

名称类型默认值必需描述
portinteger (min 1, max 65535)7233否Port the frontend service serves the gRPC API on. This is the address workers and clients dial, and the one the provides contract publishes. Most operators should omit this and let the Recipe assign the default (7233).
console_portinteger (min 1, max 65535)8080否Port the browser console listens on. It carries no application traffic; it exists so the console can be reached separately from the gRPC API.
history_shardsinteger (min 1)512否How many shards the event history is partitioned across, which bounds how far the engine can scale out. It is fixed for the life of an installation — the engine cannot repartition an existing history, so raising it later requires a new installation. Size it for the load the installation is expected to reach rather than for its first day.
source_versionstring—否Source artifact version (Helm chart). Overrides the Recipe's default source version. Changing from default does not guarantee value mapping compatibility.
computeobject—否—
overridesobject (open)—否Provider-specific overrides. Accepts any nested structure. Each Recipe documents which keys it reads.

输出契约

此组件通过 uses 拓扑向使用方发布的值。

名称类型默认值必需描述
hoststring—是In-cluster DNS name of the Service fronting the engine's frontend. A worker or a client that starts a run points its connection target at this name together with `port`.
portinteger—是TCP port that frontend accepts client connections on. It follows the port an operator set in the Release, so a consumer reads it rather than assuming a default.
protocolstring—否How the published address is spoken to — `grpc`, so a consumer configures one of the project's client libraries rather than an HTTP client.
metrics_pod_selectorstring—否Kubernetes label selector matching the pods of every engine service, for a scraper that discovers those pods directly rather than scraping one address. Published beside `metrics_pod_port` and `metrics_path`.
metrics_pod_portstring—否Name of the container port on those pods that carries the metrics, `metrics`.
metrics_pathstring—否Path the metrics are served at on that port, `/metrics`.

uses 套接字

此组件用于建立连接的依赖套接字。通过某个套接字接入的每个连接,其目标都必须满足该套接字的标签选择器,且每个连接都可以在该套接字声明的槽位中携带附加项。必需槽位须由 Aggregate 或 Release 通过该套接字接入的每个连接来填充。

metadata_store

选择器
component.aggrega.ai/category=datastore
注入
env

经由此套接字的每个连接可填充的槽位

槽位选择器必需注入敏感描述
credentialfeature.aggrega.ai/kind=secret否—是Password the engine authenticates to the relational datastore this connection reaches with. When attached, the Recipe points the engine at the produced Secret so the value never appears in the rendered manifest. Unattached, the engine is configured with no password, which suits a store reached without one and matches what the standard-library relational Recipes leave in place when nothing is attached to them.

附加槽位

Feature 附加到此组件时所用的命名绑定点。槽位的选择器限定哪个 Feature 可以填充该槽位,其 inject 列表则指明 Feature 的内容可以如何交付。必需槽位必须在每个实例上完成绑定,可以通过实例上的附加,也可以通过其上层作用域的附加;敏感槽位则声明为用于处理机密内容。Feature 也可以以任何槽位都未声明的名称进行附加。

未声明附加槽位。

自注入允许列表

env

关系

与其他已发布实体之间的类型化边。没有已发布目标的引用会呈现为断开的边,而不是链接。

使用

被使用

Component 概念参考 →

Aggrega 目录 — 收录组件、功能、配方以及渲染它们的插件的开放索引。

基于共享的 Aggrega 设计系统构建。

隐私