GATEWAY ERROR

[crisp geode]

since i shifted from .15 to .26 version my gateway loads slowly and hit internet errors initially it get connected to channels after that it get continue disconnected

20:18:11 [whatsapp] Inbound message +-> +(direct, image/jpeg, 73 chars)
20:20:50 [whatsapp] Web connection closed (status 408). Retry 1/12 in 2.38s… (status=408 Request Time-out Connection was lost)
20:20:50 [telegram] Polling stall detected (active getUpdates stuck for 175.81s); forcing restart. [diag inFlight=1 outcome=started startedAt=1777387674488 finishedAt=1777387674488 durationMs=30784 offset=0]
20:20:50 [telegram] [diag] polling cycle finished reason=polling stall detected inFlight=0 outcome=error startedAt=1777387674488 finishedAt=1777387850303 durationMs=175815 offset=0 error=Network request for 'getUpdates' failed!
20:20:50 [telegram] polling runner stopped (polling stall detected); restarting in 2.14s.
20:20:50 [discord] gateway: Gateway websocket closed: 1006
20:20:50 [discord] gateway: Gateway reconnect scheduled in 938ms (close, resume=true)
20:23:35 [telegram] [diag] closing stale transport before rebuild
20:23:35 [telegram] [diag] rebuilding transport for next polling cycle
20:23:35 [telegram] menu text exceeded the conservative 5700-character payload budget; shortening descriptions to keep 64 commands visible.
20:23:35 [health-monitor] [discord:default] health-monitor: restarting (reason: disconnected)

[raven current]

This looks less like one bad channel config and more like the host’s outbound network/TLS/DNS stalling: WhatsApp, Telegram, and Discord all drop around the same timestamp. The menu text exceeded... line is harmless.

First please paste the safe output of:

openclaw --version
node --version
which openclaw
openclaw gateway status --deep
openclaw status --all
openclaw channels status --probe

Docs triage flow: https://docs.openclaw.ai/help/faq and https://docs.openclaw.ai/gateway/troubleshooting

Also redact phone numbers/tokens from logs before pasting.

A couple likely angles:

• If this is a VPS/WSL/proxy setup, check DNS/IPv6/TLS egress, especially to api.telegram.org.
• If openclaw status --all shows memory/event-loop/liveness warnings, the gateway may be getting saturated after startup.
• If only Telegram keeps stalling, there are Telegram network knobs documented here: https://docs.openclaw.ai/channels/telegram#polling-or-network-instability

But don’t change config yet — the multi-channel simultaneous disconnect is the important clue. Paste those diagnostics and we can narrow it down. 🦐

[crisp geode]

Z:\openclaw-main>openclaw --version
OpenClaw 2026.4.26 (be8c246)
Z:\openclaw-main>which openclaw
'which' is not recognized as an internal or external command,
operable program or batch file.
Z:\openclaw-main>node --version
v22.14.0
Z:\openclaw-main>openclaw channels status --probe
🦞 OpenClaw 2026.4.26 (be8c246)
Your personal assistant, minus the passive-aggressive calendar reminders.
│
◇
Gateway reachable.

• Discord default: enabled, configured, running, connected, bot:@AYU, token:config, intents:content=limited, works
• Telegram default: enabled, configured, running, connected, mode:polling, bot:@crisp geodebot, token:config, works
• WhatsApp default: enabled, configured, linked, running, connected, in:9m ago, out:9m ago, dm:allowlist, allow:[phone],[phone]

Tip: status --deep adds gateway health probes to status output (requires a reachable gateway).

PS Z:\openclaw-main> openclaw gateway status --deep

🦞 OpenClaw 2026.4.26 (be8c246)
Say "stop" and I'll stop—say "ship" and we'll both learn a lesson.
│
◇
Service: Scheduled Task (registered)
File logs: ~\AppData\Local\Temp\openclaw\openclaw-2026-04-28.log
Command: C:\Program Files\nodejs\node.exe C:\Users\theay\AppData\Roaming\npm\node_modules\openclaw\dist\index.js gateway --port 18789
Service file: ~.openclaw\gateway.cmd
Service env: OPENCLAW_GATEWAY_PORT=18789
Config (cli): ~.openclaw\openclaw.json
Config (service): ~.openclaw\openclaw.json
Gateway: bind=lan (0.0.0.0), port=18789 (service args)
Probe target: ws://127.0.0.1:18789
Dashboard: http://192.168.1.13:18789/
Probe note: bind=lan listens on 0.0.0.0 (all interfaces); probing via 127.0.0.1.
Runtime: running (pid 22068, Startup-folder login item installed; listener detected on port 18789.)
Connectivity probe: ok
Capability: admin-capable
Listening: 0.0.0.0:18789
Troubles: run openclaw status
Troubleshooting: https://docs.openclaw.ai/troubleshooting

[raven current]

Good, this confirms the gateway is healthy right now and all three channel probes work. So we need to catch it during/after the disconnect.

Two notes:

• which failing is normal on Windows. Use:

  where.exe openclaw
  

• The simultaneous WhatsApp 408 + Telegram network stall + Discord 1006 still points to intermittent network/host stall, not one channel config.

Please run these next, ideally right after it disconnects:

openclaw status --all
openclaw logs --limit 500 --max-bytes 1000000 --local-time
openclaw gateway stability

openclaw doctor
where.exe openclaw

Also quick network checks:

Test-NetConnection api.telegram.org -Port 443
Test-NetConnection discord.com -Port 443

One thing I noticed: gateway status says Scheduled Task registered, but runtime mentions Startup-folder login item installed too. That can be fine, but openclaw doctor should catch duplicate/stale service startup state if it exists.

If doctor shows a repair, paste it before running fixes unless it’s obvious. If this keeps reproducing on 2026.4.26, generate a sanitized bundle:

openclaw gateway diagnostics export

Docs: https://docs.openclaw.ai/gateway/diagnostics

[crisp geode]

PS Z:\openclaw-main> openclaw status --all
OpenClaw status --all

Overview

│ Item │ Value

│ Version │ 2026.4.26
│ OS │ windows 10.0.26200 (x64)
│ Node │ 22.14.0
│ Config │ ~.openclaw\openclaw.json
│ Dashboard │ http://192.168.1.13:18789/
│ Tailscale exposure │ off
│ Channel │ stable (default)
│ Update │ pnpm · up to date · npm latest 2026.4.26 │
│ Gateway │ local · ws://127.0.0.1:18789 (local loopback) · reachable 40ms · │
│ │ auth token · MSI (192.168.1.13) app 2026.4.26 windows 10.0.26200 │
│ Security │ Run: openclaw security audit --deep │
│ Gateway self │ unknown │
│ Gateway service │ Scheduled Task installed · registered · running (pid 22068, │
│ │ Startup-folder login item installed; listener detected on port │
│ │ 18789.) │
│ Node service │ Scheduled Task not installed │
│ Agents │ 1 total · 0 bootstrapping · 0 active · 29 sessions │
│ Secrets │ none │

Channels

│ Channel │ Enabled │ State │ Detail

Agents

│ Agent │ Bootstrap file │ Sessions │ Active │ Store

│ main (main) │ ABSENT │ 29 │ 25m ago │ ~. │
│ │ │ │ │ openclaw\agents\main\sessions\ses │
│ │ │ │ │ sions.json │

Z:\openclaw-main>where.exe openclaw
C:\Users\theay\AppData\Roaming\npm\openclaw
C:\Users\theay\AppData\Roaming\npm\openclaw.cmd

[raven current]

Thanks — that rules out a simple PATH/version split-brain. Your service is using the same global npm install:

C:\Users\theay\AppData\Roaming\npm\openclaw.cmd

and the gateway is reachable locally.

What we don’t have yet is the useful bit: logs/stability from the moment it drops. Please run these after the next disconnect:

openclaw logs --limit 500 --max-bytes 1000000 --local-time
openclaw gateway stability
openclaw status --deep
openclaw channels status --probe
openclaw doctor

Network checks too:

Test-NetConnection api.telegram.org -Port 443
Test-NetConnection discord.com -Port 443
Resolve-DnsName api.telegram.org
Resolve-DnsName discord.com

From what you pasted, all channels were working when probed, so the failure is intermittent. Because WhatsApp, Telegram, and Discord all disconnect together, I’d still suspect either:

1. Windows/native host network stall
2. DNS/IPv6/TLS issue
3. gateway event-loop/CPU stall
4. Windows sleep/power/network adapter saving state

Also: native Windows works, but WSL2 is still the more stable path for OpenClaw if this keeps happening: https://docs.openclaw.ai/platforms/windows