fix(subagent): scope drains and soften final flush

This commit is contained in:
imccyu
2026-08-01 21:49:27 +08:00
committed by Tianyi Cui
parent d7153768f5
commit a1b3bebb61
18 changed files with 203 additions and 111 deletions

View File

@@ -2,5 +2,5 @@
# side as of the last confirmed-consistent state. Both languages carry equal authority;
# after editing either side, bring the other along and re-record with:
# pnpm run verify-translation-pairing --write docs/core-data-structures/subagent.md
subagent.md: 4d389c106133114781ef0a3128032433fd0ed7ec
subagent.zh.md: cb374d03bbd9716ee53a1cecf938ff1dc1f8ce66
subagent.md: ce81eaf748ce4ce99b18cd218a99317998eeb45b
subagent.zh.md: cac1aa68614314fbad6a459a7eb877e6537a783d

View File

@@ -141,9 +141,9 @@ Follow-up authority comes from an exact live Agent tool context. The authenticat
For both operations the caller signal owns lookup, materialization, and admission only until inbox acceptance. Afterwards the manager owns the Activation independently: later caller cancellation neither cancels the accepted turn nor disposes the child, and the seam exposes no public subagent cancellation or steering operation.
Every Activation owns its `AgentHandle` and an `ownedChildren: Set<SessionId>`; because one Session has at most one live Activation, the child Session id identifies the live child without another runtime-incarnation reference. Starting a child or submitting parent-originated work registers the child in a continuation-managed parent's set before the child can run, and that parent cannot settle while the set is non-empty. A top-level or other non-continuation Agent has no Activation and stays outside the waiting graph. Child release happens only after the child Agent is quiescent, every child of that child is disposed, the final durability checkpoint settles, and the child's `AgentHandle` completes disposal.
Every Activation owns its `AgentHandle` and an `ownedChildren: Set<SessionId>`; because one Session has at most one live Activation, the child Session id identifies the live child without another runtime-incarnation reference. Starting a child or submitting parent-originated work registers the child in a continuation-managed parent's set before the child can run, and that parent cannot settle while the set is non-empty. A top-level or other non-continuation Agent has no Activation and stays outside the waiting graph. Child release happens only after the child Agent is quiescent, every child of that child is disposed, the best-effort final session flush settles, and the child's `AgentHandle` completes disposal.
Only `ctx.sessions.flush(session) === true` confirms durability; `false` or rejection reports `DURABILITY_FAILED`. Either way the manager still disposes the handle and releases ownership, because retaining a failed child would permanently pin its ancestors in `waiting` — the persisted child state may then be missing or stale on a later resume. Manager unload invokes an internal manager-wide drain that closes admission and disposes every live forest; `drainContinuableDescendants(parents)` closes admission only below exact live host-owned Agents and disposes their continuable descendants while unrelated forests remain live. Both await already-admitted materializations in their scope, propagate cancellation top-down, release handles child-first, and await every selected branch despite individual failures. Durable child Sessions survive that process-local teardown.
Final settlement awaits `ctx.sessions.flush(session)` but ignores its participation boolean because an arbitrary listener cannot prove that a persistence backend stored the state. Rejection is logged without failing the Activation, and the manager still disposes the handle and releases ownership; the persisted child state may then be missing or stale on a later resume. Manager unload invokes an internal manager-wide drain that closes admission and disposes every live forest; `drainContinuableDescendants(parents)` closes admission only below exact live host-owned Agents and disposes their continuable descendants while unrelated forests remain live. Both await already-admitted materializations in their scope, propagate cancellation top-down, release handles child-first, and await every selected branch despite individual failures. Durable child Sessions survive that process-local teardown.
```ts type-equiv
/** Attribution for a model coordinator's follow-up to one of its children. */

View File

@@ -141,9 +141,9 @@ Agent 收件箱是唯一的队列。每条继续执行消息都会成为一个 `
对于这两种操作,调用方 signal 仅在收件箱接受之前掌管查找、物化与准入。此后管理器独立掌管该 Activation之后的调用方取消既不会取消已接受的轮次也不会 dispose 子 agent并且该 seam 不对外暴露任何 subagent 取消或 steering中途引导操作。
每个 Activation 都拥有自己的 `AgentHandle` 和一个 `ownedChildren: Set<SessionId>`;由于一份会话至多有一个存活 Activation子会话 id 无需另一个运行时化身引用即可标识存活的子 agent。启动子 agent 或提交源自 parent 的工作,会在子 agent 能够运行之前将其注册到受继续执行管理的父级集合中;只要该集合非空,该父级就无法 settle。顶层或其他非继续执行的 Agent 没有 Activation处于 waiting 图之外。只有当子 Agent 已停稳、该子 agent 的每个子级都已 dispose、最终的持久性检查点结算完毕,且子 agent 的 `AgentHandle` 完成 dispose 之后,才会释放子 agent。
每个 Activation 都拥有自己的 `AgentHandle` 和一个 `ownedChildren: Set<SessionId>`;由于一份会话至多有一个存活 Activation子会话 id 无需另一个运行时化身引用即可标识存活的子 agent。启动子 agent 或提交源自 parent 的工作,会在子 agent 能够运行之前将其注册到受继续执行管理的父级集合中;只要该集合非空,该父级就无法 settle。顶层或其他非继续执行的 Agent 没有 Activation处于 waiting 图之外。只有当子 Agent 已停稳、该子 agent 的每个子级都已 dispose、best-effort 的最终会话 flush 结算完毕,且子 agent 的 `AgentHandle` 完成 dispose 之后,才会释放子 agent。
只有 `ctx.sessions.flush(session) === true` 才确认持久性;`false` 或 rejection 会报告 `DURABILITY_FAILED`。无论哪种情况,管理器仍会 dispose 该 handle 并释放所有权,因为保留一个失败的子 agent 会将其祖先永久钉在 `waiting`——此后持久化的子 agent 状态在后续恢复时可能缺失或陈旧。管理器卸载会调用内部的管理器全局 drain关闭准入并 dispose 每片在线森林;`drainContinuableDescendants(parents)` 只关闭由 host 确切拥有的在线 Agent 之下的准入,并 dispose 其可继续后代,而无关森林保持在线。两者都会等待各自作用域内已获准的物化过程,自顶向下传播取消,按 child-first 顺序释放 handle并且即使个别分支失败也会等待所有选中分支。持久化子会话不受该进程内拆卸的影响。
最终结算会等待 `ctx.sessions.flush(session)`,但会忽略其参与布尔值,因为任意 listener 都无法证明某个持久化后端已存储该状态。rejection 会被记录,但不会使 Activation 失败;管理器仍会 dispose 该 handle 并释放所有权,此后持久化的子 agent 状态在后续恢复时可能缺失或陈旧。管理器卸载会调用内部的管理器全局 drain关闭准入并 dispose 每片在线森林;`drainContinuableDescendants(parents)` 只关闭由 host 确切拥有的在线 Agent 之下的准入,并 dispose 其可继续后代,而无关森林保持在线。两者都会等待各自作用域内已获准的物化过程,自顶向下传播取消,按 child-first 顺序释放 handle并且即使个别分支失败也会等待所有选中分支。持久化子会话不受该进程内拆卸的影响。
```ts type-equiv
/** Attribution for a model coordinator's follow-up to one of its children. */