fix(telegram): expose spooled handler timeout config by khang-dogo · Pull Request #87891 · openclaw/openclaw

khang-dogo · 2026-05-29T04:51:02Z

Summary

expose channels.telegram.spooledUpdateHandlerTimeoutMs with top-level and per-account config support
pass the configured value into Telegram isolated polling sessions
document the timeout and regenerate bundled channel config metadata

Why

Telegram isolated ingress already fails stale claimed spool updates after a handler timeout, but operators could not tune that timeout from channels.telegram. A wedged .processing file can therefore keep a same-chat lane blocked until the default 25 minute timeout.

Runtime Proof

I ran a redacted local runtime harness against the PR head using the real TelegramPollingSession, real spool files, real reply-fence timeout recovery path, and a fake Telegram Bot runtime to avoid live network calls. The harness configured isolatedIngress.spooledUpdateHandlerTimeoutMs = 200, wrote two same-lane spooled updates, wedged update 42, and verified update 43 drained only after timeout recovery restarted isolated ingress.

{
  "configuredTimeoutMs": 200,
  "configuredAbortGraceMs": 200,
  "elapsedMs": 475,
  "workerCreateCount": 2,
  "botConstructCount": 2,
  "events": [
    "first:42",
    "second:43"
  ],
  "failedUpdateIds": [
    42
  ],
  "pendingUpdateIds": [],
  "timeoutLog": "[telegram] Telegram isolated polling spool handler timed out behind update 42 on lane telegram:-100:topic:10 after 205ms; marking the update failed, aborting active reply work, and restarting isolated ingress so later updates can drain."
}

This shows the configured timeout was honored by the isolated-ingress runtime path: update 42 was failed after the configured timeout budget and the later same-lane update 43 drained after the restart.

Validation

pnpm config:channels:check
pnpm exec vitest run extensions/telegram/src/config-schema.test.ts extensions/telegram/src/polling-session.test.ts passed locally while preparing the patch. A later rerun in this workspace timed out in Vitest/Rolldown pre-test bundling with repeated plugin timing warnings before any assertion output.

clawsweeper · 2026-05-29T04:52:39Z

Thanks for the contribution. ClawSweeper proposes closing this for now: the implementation may be reasonable, but passing review and proof does not establish that OpenClaw should add this product surface.

Close for unconfirmed product direction: the patch is coherent and proof-backed, but it adds a new public Telegram operator config contract without maintainer sponsorship for the key name, bounds, default, and support commitment.

This is a proposal only until the separate default-off apply policy is enabled and all live maintainer-signal checks pass. A maintainer can sponsor the direction, request a narrower version, or apply clawsweeper:human-review to keep it open.

Review details

Best possible solution:

Keep the internal recovery path and the linked lane-release fix separate; expose a public timeout knob only if a maintainer sponsors the API name, bounds, default, docs, and support contract.

Do we have a high-confidence way to reproduce the issue?

Yes for the claimed config gap: current main has the internal isolated-ingress timeout hook but no public schema/type/docs/monitor wiring for channels.telegram.spooledUpdateHandlerTimeoutMs. I did not run tests because this review is read-only.

Is this the best way to solve the issue?

No as a merge path: the implementation is narrow and plausible, but adding a public Telegram config knob is not the best default without maintainer sponsorship. The safer path is to land the runtime lane-release repair separately and add this tuning surface only after the API contract is accepted.

Security review:

Security review cleared: Security review cleared: the diff changes Telegram config schema/types/docs/generated metadata/tests and one option-plumbing line, with no dependency, lockfile, workflow, secret, permission, or code-execution surface concern found.

AGENTS.md: found and applied where relevant.

What I checked:

Repository policy makes this compatibility-sensitive: Root policy treats config/default additions as compatibility-sensitive and expects review metrics plus maintainer-visible risk for public operator surfaces. (AGENTS.md:31, 82dfd8910743)
Telegram maintainer note checked: The Telegram maintainer note requires stronger live proof for transport, streaming, topics, callbacks, authorization, and reply-context changes; this PR is an internal timeout/config surface, so the runtime harness is relevant but product direction remains separate. (.agents/maintainer-notes/telegram.md:37, 82dfd8910743)
Current main has an internal timeout hook: Current main resolves opts.isolatedIngress?.spooledUpdateHandlerTimeoutMs or the environment override and falls back to the 25-minute isolated-ingress backlog stall budget. (extensions/telegram/src/polling-session.ts:393, 82dfd8910743)
Current main does not expose the public config key: The monitor passes adjacent Telegram account config into TelegramPollingSession, but current main does not pass a public spooledUpdateHandlerTimeoutMs value. (extensions/telegram/src/monitor.ts:301, 82dfd8910743)
Current schema gap confirmed: Current main exposes pollingStallThresholdMs in the Telegram account schema but has no public spooledUpdateHandlerTimeoutMs schema field. (src/config/zod-schema.providers-core.ts:298, 82dfd8910743)
PR adds the public config contract: The PR adds spooledUpdateHandlerTimeoutMs to the Telegram schema with 30,000 to 3,600,000 ms bounds and documentation text. (src/config/zod-schema.providers-core.ts:293, 1bdf43652ce7)

Likely related people:

joshavant: Authored the merged Telegram isolated-ingress and stalled spool-handler recovery work that shaped this timeout, reply-fence, and restart path. (role: feature-history owner; confidence: high; commits: fd244fd76de3, b7735f88fa27, 9921825e1795; files: extensions/telegram/src/polling-session.ts, extensions/telegram/src/polling-session.test.ts, extensions/telegram/src/telegram-ingress-spool.ts)
0xghost42: Authored the recent merged lone-active-spooled-handler timeout recovery PR on the same polling-session timeout path. (role: recent adjacent contributor; confidence: high; commits: 11b6c01198a0, 9f7577d9d08d; files: extensions/telegram/src/polling-session.ts, extensions/telegram/src/polling-session.test.ts)
vincentkoc: Recent history includes Telegram/config schema maintenance and adjacent spool claim preservation work near the affected surfaces. (role: recent config and Telegram area contributor; confidence: medium; commits: 2ddebf3897be, 49572863d3c1, 2da361a3457d; files: src/config/zod-schema.providers-core.ts, extensions/telegram/src/telegram-ingress-spool.ts, extensions/telegram/src/polling-session.ts)

Codex review notes: model internal, reasoning high; reviewed against 82dfd8910743.

khang-dogo · 2026-05-29T06:15:24Z

Added runtime proof to the PR body.

The proof uses the real TelegramPollingSession, real isolated-ingress spool files, and real reply-fence timeout recovery path with a fake Telegram Bot runtime to avoid live network calls. It shows spooledUpdateHandlerTimeoutMs = 200 causing update 42 to be failed after ~205ms and update 43 on the same lane draining after the restart.

@clawsweeper re-review

clawsweeper · 2026-05-29T06:15:28Z

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

Re-review progress:

State: Complete
Detail: The targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
Run: https://github.com/openclaw/clawsweeper/actions/runs/26621497182
Updated: 2026-05-29T06:21:57.447Z

fix(telegram): expose spooled handler timeout config

1bdf436

openclaw-barnacle Bot removed the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label May 29, 2026

openclaw-barnacle Bot added the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label May 29, 2026

openclaw-barnacle Bot removed the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label May 29, 2026

clawsweeper Bot removed the rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. label Jun 20, 2026

clawsweeper Bot added the rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. label Jun 20, 2026

openclaw-barnacle Bot added the triage: needs-pr-context Candidate: external PR body lacks required problem context or evidence. label Jun 20, 2026

This was referenced Jun 20, 2026

fix(telegram): release timed-out spooled lanes #93040

Closed

feat(telegram): add isolatedIngress config switch #95335

Closed

clawsweeper Bot added rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. and removed rating: 🦐 gold shrimp Decent PR readiness signal, but merge confidence is limited. labels Jul 5, 2026

steipete closed this Jul 9, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(telegram): expose spooled handler timeout config#87891

fix(telegram): expose spooled handler timeout config#87891
khang-dogo wants to merge 1 commit into
openclaw:mainfrom
khang-dogo:fix/telegram-spooled-handler-timeout-config

khang-dogo commented May 29, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 29, 2026 •

edited

Loading

Uh oh!

khang-dogo commented May 29, 2026

Uh oh!

clawsweeper Bot commented May 29, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Uh oh!

Conversation

khang-dogo commented May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Why

Runtime Proof

Validation

Uh oh!

clawsweeper Bot commented May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

khang-dogo commented May 29, 2026

Uh oh!

clawsweeper Bot commented May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

khang-dogo commented May 29, 2026 •

edited

Loading

clawsweeper Bot commented May 29, 2026 •

edited

Loading

clawsweeper Bot commented May 29, 2026 •

edited

Loading