fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes #48177) by yassinebkr · Pull Request #48282 · openclaw/openclaw

yassinebkr · 2026-03-16T14:44:25Z

Summary

Hoists resolveTelegramTransport() out of createTelegramBot() so the transport and its sticky IPv4 fallback state persists across polling restarts.

Problem

When Node 22+ autoSelectFamily=true and the host has unstable IPv6 to api.telegram.org, the transport correctly enables sticky IPv4 fallback. But this state is lost on every polling restart because createTelegramBot() creates a fresh transport each time, resetting stickyIpv4FallbackEnabled = false.

This causes a repeating cycle of IPv6 timeouts — the reporter observed 47 polling stalls in 20 hours with 30s-2min message loss per stall.

Changes

bot.ts: Accept optional telegramTransport in TelegramBotOptions, use it if provided
monitor.ts: Resolve transport once before polling session creation
polling-session.ts: Pass transport through to bot creation

Testing

Minimal diff (+15 lines). All changes are additive — existing code paths without a pre-resolved transport still work identically.

AI-assisted (Claude Sonnet 4).

Fixes #48177

greptile-apps · 2026-03-16T14:46:15Z

Greptile Summary

This PR fixes a regression where the sticky IPv4 fallback state inside TelegramTransport was silently reset on every polling restart, causing repeated IPv6 connection timeouts. The fix is correct: resolveTelegramTransport() is hoisted to monitor.ts so the transport (and its internal stickyIpv4FallbackEnabled + stickyIpv4Dispatcher closure state) survives across the restart loop in TelegramPollingSession. All existing code paths without a pre-resolved transport are unaffected.

The change is minimal, additive, and accurately targets the root cause.
One minor issue: resolveTelegramTransport is imported in polling-session.ts but never called there — only TelegramTransport (the type) is used. This is dead code that should be trimmed.
No concerns about shared dispatcher state or race conditions; undici agents are designed for reuse, and JavaScript's single-threaded model means the stickyIpv4FallbackEnabled flag is safely mutated within the async fetch closure.

Confidence Score: 5/5

Safe to merge — the fix is minimal, additive, and correctly targets the root cause with no behaviour change on existing code paths.
The change is a small, well-scoped fix that only affects the transport wiring path. All existing callers that don't supply a telegramTransport continue to work exactly as before. The one unused import is a trivial cleanup item, not a blocker.
No files require special attention beyond the minor unused-import cleanup in polling-session.ts.

Prompt To Fix All With AI

This is a comment left during a code review.
Path: extensions/telegram/src/polling-session.ts
Line: 7

Comment:
**Unused import of `resolveTelegramTransport`**

`resolveTelegramTransport` is imported here but never called anywhere in `polling-session.ts`. Only the `TelegramTransport` type is actually used. The transport is now resolved once upstream in `monitor.ts` and passed in via `opts.telegramTransport` — this file no longer needs the function itself.

```suggestion
import { type TelegramTransport } from "./fetch.js";
```

How can I resolve this? If you propose a fix, please make it concise.

_{Last reviewed commit: 8e9c769}

extensions/telegram/src/polling-session.ts

…ixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean.

Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>

@yassinebkr

…penclaw#48282) (thanks @yassinebkr)

obviyus · 2026-03-17T16:26:35Z

Landed on main.

Landed source commit: a5a8711
Merge commit: 8139f83

Thanks @yassinebkr.

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]>

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]> (cherry picked from commit 8139f83)

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]>

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]>

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]> (cherry picked from commit 8139f83)

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]> (cherry picked from commit 8139f83)

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]> (cherry picked from commit 8139f83) # Conflicts: # CHANGELOG.md

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]> (cherry picked from commit 8139f83)

@yassinebkr

…ixes openclaw#48177) (openclaw#48282) * fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes openclaw#48177) Hoist resolveTelegramTransport() out of createTelegramBot() so the transport (and its sticky IPv4 fallback state) persists across polling restarts. Previously, each polling restart created a new transport with stickyIpv4FallbackEnabled=false, causing repeated IPv6 timeouts on hosts with unstable IPv6 connectivity. Changes: - bot.ts: accept optional telegramTransport in TelegramBotOptions - monitor.ts: resolve transport once before polling loop - polling-session.ts: pass transport through to bot creation AI-assisted (Claude Sonnet 4). Tested: tsc --noEmit clean. * Update extensions/telegram/src/polling-session.ts Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> * style: fix oxfmt formatting in bot.ts * test: cover telegram transport reuse across restarts * fix: preserve telegram sticky IPv4 fallback across polling restarts (openclaw#48282) (thanks @yassinebkr) --------- Co-authored-by: Yassine <[email protected]> Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Co-authored-by: Ayaan Zaidi <[email protected]>

openclaw-barnacle bot added channel: telegram Channel integration: telegram size: XS labels Mar 16, 2026

greptile-apps bot reviewed Mar 16, 2026

View reviewed changes

extensions/telegram/src/polling-session.ts Outdated Show resolved Hide resolved

openclaw-barnacle bot added size: S and removed size: XS labels Mar 17, 2026

obviyus self-assigned this Mar 17, 2026

yassinebkr and others added 5 commits March 17, 2026 21:52

Update extensions/telegram/src/polling-session.ts

62f8bf9

Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>

style: fix oxfmt formatting in bot.ts

611902b

test: cover telegram transport reuse across restarts

ee614fc

fix: preserve telegram sticky IPv4 fallback across polling restarts (o…

a5a8711

…penclaw#48282) (thanks @yassinebkr)

obviyus force-pushed the fix/issue-48177-sticky-ipv4-fallback branch from e6d86a4 to a5a8711 Compare March 17, 2026 16:25

obviyus merged commit 8139f83 into openclaw:main Mar 17, 2026
8 checks passed

yassinebkr deleted the fix/issue-48177-sticky-ipv4-fallback branch March 17, 2026 16:49

github-actions bot mentioned this pull request Mar 17, 2026

📡 Upstream Digest — 2026-03-17 16:58 UTC curtismercier/openclaw-mods#290

Open

alexey-pelykh mentioned this pull request Mar 23, 2026

Cherry-pick (blocked): Telegram adapter (1/2) (50 commits) remoteclaw/remoteclaw#1920

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes #48177)#48282

fix(telegram): persist sticky IPv4 fallback across polling restarts (fixes #48177)#48282
obviyus merged 5 commits intoopenclaw:mainfrom
yassinebkr:fix/issue-48177-sticky-ipv4-fallback

yassinebkr commented Mar 16, 2026

Uh oh!

greptile-apps bot commented Mar 16, 2026

Uh oh!

Uh oh!

Uh oh!

obviyus commented Mar 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

yassinebkr commented Mar 16, 2026

Summary

Problem

Changes

Testing

Uh oh!

greptile-apps bot commented Mar 16, 2026

Greptile Summary

Confidence Score: 5/5

Uh oh!

Uh oh!

Uh oh!

obviyus commented Mar 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants