Files
PaperClipAI/server/src/services/plugin-agent-runtime-state.ts
T
DottaandPaperclip 006d5691a6 feat(server): support volume-free web server handoff (#18518)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Its web server also owns controller admission and user commands.
> - A rolling deployment starts a replacement before the old server
exits.
> - Shared database state alone does not preserve local agent files,
browser chunks, or in-process commands.
> - This pull request adds an opt-in ephemeral server contract and a
local overlap test.
> - The benefit is a web server that can relinquish traffic and local
files without losing acknowledged work.

## Linked Issues or Issue Description

Refs #18403, #18327, #18335, #18341.

**Current behavior**

The control-plane web server still has local agent-file authority and
local command targets. A new container can miss saved bytes or send a
command to the wrong process. An existing browser tab also needs chunks
from its original build.

**Proposed behavior**

Eligible servers share only PostgreSQL, object storage and stable
secrets. A replacement waits for readiness. The old server drains
admitted requests, keeps owner commands available through handoff, and
releases controller leases. Explicit maintenance adopts existing files
before the volume is removed.

**Reason and benefit**

This is the application prerequisite for rolling app-only deployments.
Schema migrations and permanent multi-replica operation are outside this
change.

## What Changed

- Store agent-file manifests in PostgreSQL and immutable bytes in S3.
Treat local roots as disposable projections. Preserve existing
permission checks and edit conflicts.
- Add an explicit migration, eligibility and object-retention command to
the production runtime bundle.
- Add `/api/ready`, admitted-request drain, browser restart signaling
and retained immutable UI chunks.
- Carry transaction-aware work hints across controllers during handoff
and active execution. Keep listeners closed when idle.
- Deliver durable Stop and runtime responses to the execution owner.
Recheck request lifecycle and access before response dispatch.
- Reject local-only configuration in ephemeral mode. Keep the feature
disabled for existing installations by default.
- Add an isolated PostgreSQL/S3/proxy/browser deployment harness with
independent server homes and retained failure evidence.

## Verification

- All CI checks pass on `ce17b95019d3cf1cabd357d19b6d37f96782cc5c`:
workspace typecheck, build, Runner checks, server/general tests and
browser shards. Greptile reviewed the runtime changes at 5/5 with all
findings resolved; the final commit updates maps and the plan.
- Local targeted checks passed: coordinator 27; Stop/recovery 33;
PostgreSQL command delivery 6; route/workspace regressions 8; bootstrap
Stop 1; restart recovery 75. PostgreSQL command tests simulate the
provider edge.
- Local browser proof passed 91 reads and 19 saves; reconnect took
1,013ms. The tab opened Audit from S3 after deletion of A's whole
home/build. This uses compiled workspace code and local-trusted auth,
not released containers or live agents.
- Local workspace typecheck/build passed; current server and runtime
bundle passed. The earlier broad local suite was stopped after three
hours and is not a passing receipt. Its retained failures and isolated
reruns are documented in the plan.
- Stacked #18526 adds stricter graceful-exit proof, an unready
replacement control, live Runner qualification and a Stop handoff
correction. Hosted enablement remains gated on those results and Cloud
#1096. No hosted stack has been enabled.

## Risks

- Apply migration 0422 before enabling the feature. Existing data must
pass each owner's migration and eligibility checks. Do not discard a
volume based on a single migration page.
- Durable manifests remain authoritative after the opt-in is disabled.
Rolling back to a binary that ignores them is unsupported.
- Cold directory restoration and complete shutdown timing still need
qualification before enabling a hosted stack.
- Retained UI assets must cover the supported browser and rollback
window. Cleanup is explicit; no new idle maintenance polling is
introduced.
- This change does not claim production zero downtime or authorize a
fleet-wide enablement.

## Model Used

OpenAI Codex, GPT-6 family. The exact serving model ID and
context-window limit are not exposed in this session. Used reasoning,
code editing, shell execution and browser automation.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-01 01:58:54 -06:00

84 lines
6.6 KiB
TypeScript

import { and, desc, eq } from "drizzle-orm";
import { agents, environments, plugins, pluginCompanySettings, resourceLifecycleEvents, fileStorageDb, type Db } from "@paperclipai/db";
import type { PluginAgentRuntimeState, PluginAgentRuntimeIdentity } from "@paperclipai/plugin-sdk";
import { agentRecordQueries } from "../lib/agent-records.js";
import { conflict, forbidden, unprocessable } from "../errors.js";
import { agentInstructionsService } from "./agent-instructions.js";
import { pluginAgentConnectionAssignments } from "./plugin-agent-connections.js";
import { companySkillService } from "./company-skills.js";
import { withSkillActor } from "./skill-access.js";
import { readPaperclipSkillSyncPreference } from "@paperclipai/adapter-utils/server-utils";
import { skillVersionSelectionMap } from "./runtime-skill-selections.js";
import { instanceSettingsService } from "./instance-settings.js";
import { agentInstanceSecretService } from "./agent-instance-secrets.js";
import { environmentSkillConfig } from "./environment-skills.js";
async function readState(db: Db, pluginId: string, companyId: string, agentId: string, identityOnly: true): Promise<PluginAgentRuntimeIdentity | null>;
async function readState(db: Db, pluginId: string, companyId: string, agentId: string, identityOnly?: false): Promise<Omit<PluginAgentRuntimeState, "skills" | "connectionAssignments"> | null>;
async function readState(db: Db, pluginId: string, companyId: string, agentId: string, identityOnly = false): Promise<PluginAgentRuntimeIdentity | Omit<PluginAgentRuntimeState, "skills" | "connectionAssignments"> | null> {
return fileStorageDb(db).transaction(async tx => {
const [agent] = await tx.select().from(agents)
.where(and(eq(agents.companyId, companyId), eq(agents.id, agentId))).for("update");
if (!agent) return null;
const [plugin] = await tx.select().from(plugins).where(eq(plugins.id, pluginId));
const [settings] = await tx.select().from(pluginCompanySettings).where(and(eq(pluginCompanySettings.companyId, companyId), eq(pluginCompanySettings.pluginId, pluginId)));
if (plugin?.status !== "ready" || settings?.enabled === false || !plugin?.manifestJson.capabilities?.includes("agents.lifecycle.manage")
|| !plugin.manifestJson.resourceLifecycle?.includes("agent_runtime") || !agent.lifecycleRequiredPluginIds?.includes(pluginId)) {
throw forbidden("Agent runtime state requires the owning lifecycle plugin");
}
const [latest] = await tx.select({ id: resourceLifecycleEvents.id }).from(resourceLifecycleEvents)
.where(and(eq(resourceLifecycleEvents.companyId, companyId), eq(resourceLifecycleEvents.resourceType, "agent_runtime"), eq(resourceLifecycleEvents.resourceId, agentId)))
.orderBy(desc(resourceLifecycleEvents.id)).limit(1);
if (identityOnly) return {
revision: String(latest?.id ?? 0),
agent: { id: agent.id, companyId: agent.companyId, status: agent.status as PluginAgentRuntimeIdentity["agent"]["status"],
lifecycleState: agent.lifecycleState, lifecycleVersion: agent.lifecycleVersion },
instanceSecrets: await agentInstanceSecretService(tx as unknown as Db).read({ companyId, agentId, pluginId }),
};
const instructions = await agentInstructionsService(tx as unknown as Db).getBundleLocation(agent);
const records = agentRecordQueries(tx as unknown as Db);
const companyAgents = await records.listCompanyAgentEligibilityRows(companyId);
return {
revision: String(latest?.id ?? 0), agent: records.normalizeAgentRow(agent, companyAgents) as PluginAgentRuntimeState["agent"],
instructionsRootPath: instructions.rootPath, instructionsEntryFile: instructions.entryFile,
instanceSecrets: await agentInstanceSecretService(tx as unknown as Db).read({ companyId, agentId, pluginId }),
};
});
}
export async function pluginAgentRuntimeIdentity(db: Db, pluginId: string, companyId: string, agentId: string): Promise<PluginAgentRuntimeIdentity | null> {
return readState(db, pluginId, companyId, agentId, true);
}
export async function pluginAgentRuntimeState(db: Db, pluginId: string, companyId: string, agentId: string, environmentId?: string, installEnvironmentSkills = false): Promise<PluginAgentRuntimeState | null> {
for (let attempt = 0; attempt < 3; attempt++) {
const state = await readState(db, pluginId, companyId, agentId);
if (!state) return null;
const instance = await instanceSettingsService(db).get();
const effectiveEnvironmentId = environmentId ?? state.agent.defaultEnvironmentId ?? instance.defaultEnvironmentId;
const [environment] = effectiveEnvironmentId
? await db.select().from(environments).where(eq(environments.id, effectiveEnvironmentId)) : [];
if (environmentId && !environment) throw unprocessable("The selected environment is unavailable");
const { config: selectionConfig, skillKeys: environmentSkills, shadowedSkills } = await environmentSkillConfig(db, companyId, environment, state.agent.adapterConfig, installEnvironmentSkills);
const preference = readPaperclipSkillSyncPreference(selectionConfig);
const settings = instance.experimental;
const skills = await withSkillActor({ type: "agent", agentId, companyId, source: "agent_key" }, () =>
companySkillService(db).listRuntimeSkillEntries(companyId, {
selectionConfig, includeLegacy: state.agent.adapterType !== "paperclip_runner", snapshotLatest: true,
versionSelections: skillVersionSelectionMap(preference.desiredSkillEntries, { versionPinsEnabled: settings.enableBetaSkills }),
}));
if (environmentSkills.some(key => !skills.some(skill => skill.key === key && skill.sourceStatus !== "missing"))) {
throw unprocessable("An environment skill is unavailable to this agent");
}
const connectionAssignments = await pluginAgentConnectionAssignments(db, companyId, agentId, { publicMcp: settings.enablePublicMcp, chatConnectors: settings.enableChatConnectors });
const fallback = !state.instructionsRootPath ? await agentInstructionsService(db).exportFiles(state.agent, { rejectSymlinks: true }) : null;
const entryFile = fallback?.entryFile ?? state.instructionsEntryFile;
const entryContent = fallback?.files[entryFile];
const current = await readState(db, pluginId, companyId, agentId);
if (!current) return null;
if (current.revision === state.revision) return { ...current, instructionsEntryFile: entryFile, ...(entryContent === undefined ? {} : { instructionsEntryContent: entryContent }), skills, connectionAssignments, ...(shadowedSkills?.length ? { shadowedSkills } : {}) };
}
throw conflict("Agent runtime changed while resolving synchronization inputs; retry");
}