Openclaw telegram so slow | Friends of the Crustacean 🦞🤝 | Page 1

sour stag Apr 27, 2026, 5:17 PM

#

Openclaw telegram so slow, how do I check the latency?

olive garnetBOT Apr 27, 2026, 5:17 PM

#

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

finite hingeBOT Apr 27, 2026, 5:18 PM

#

First: what version are you on?

openclaw --version

To measure where the slowness is:

openclaw status
openclaw channels status --probe
openclaw channels logs --channel telegram

Then send a Telegram message and watch when it appears in the logs.

Quick read:

#

Message takes seconds to appear in logs → Telegram polling/network/DNS/proxy issue.
Message appears instantly, but reply takes long → model/session/tool execution latency. Send /status in Telegram too.
Reply is generated but Telegram send is slow/fails → outbound Telegram API/network issue.

For deeper Telegram HTTP diagnostics, temporarily enable:

{
  diagnostics: {
    flags: ["telegram.http"]
  }
}

Restart the gateway, then:

#

openclaw logs --follow --local-time

Also useful:

openclaw status --all

That’s designed to be pasteable/safe with tokens redacted.

Docs:
https://docs.openclaw.ai/help/faq
https://docs.openclaw.ai/channels/troubleshooting
https://docs.openclaw.ai/diagnostics/flags

sour stag Apr 27, 2026, 5:20 PM

#

2026-04-27T12:58:10.175-04:00 info {"subsystem":"gateway/channels/telegram"} [default] starting provider (@Taidigimon_bot)
2026-04-27T12:58:10.576-04:00 info {"subsystem":"gateway/channels/telegram"} Telegram menu text exceeded the conservative 5700-character payload budget; shortening descriptions to keep 78 commands visible.
2026-04-27T13:04:27.106-04:00 info {"subsystem":"gateway/channels/telegram"} [default] starting provider
2026-04-27T13:04:27.547-04:00 info {"subsystem":"gateway/channels/telegram"} Telegram menu text exceeded the conservative 5700-character payload budget; shortening descriptions to keep 78 commands visible.
2026-04-27T13:05:10.272-04:00 info {"module":"telegram-auto-reply"} {"chatId":"5024580021","senderUserId":"5024580021","username":"EnriqueMurillo5","firstName":"Enrique","matchKey":"none","matchSource":"none"} telegram pairing request
2026-04-27T13:08:10.488-04:00 info {"subsystem":"gateway/channels/telegram"} [default] starting provider (@Taidigimon_bot)
2026-04-27T13:08:10.683-04:00 info {"subsystem":"gateway/channels/telegram"} Telegram menu text exceeded the conservative 5700-character payload budget; shortening descriptions to keep 78 commands visible.
enriquemurillo@Mac-mini ~ %

finite hingeBOT Apr 27, 2026, 5:21 PM

#

Those lines don’t show latency yet — they mostly show Telegram provider restarting. The repeated:

starting provider

at 12:58, 13:04, 13:08 is the suspicious part. The “menu text exceeded…” line is harmless.

Please run/paste:

openclaw --version
openclaw gateway status
openclaw pairing list telegram
openclaw status --all

#

For actual Telegram processing latency, enable ingress timing and reproduce one message:

OPENCLAW_DEBUG_TELEGRAM_INGRESS=1 openclaw gateway

Then send a Telegram message and watch for lines like:

telegram ingress: ... contextReadyMs=...
telegram ingress: ... dispatchCompleteMs=...

Meaning:

contextReadyMs high → Telegram ingress / message context problem
dispatchCompleteMs high → model/tool/session processing is slow

#

repeated starting provider → gateway/channel is restarting; check earlier errors in logs

If you run OpenClaw as a service, also tail logs while testing:

openclaw logs --follow --local-time

Docs: https://docs.openclaw.ai/help/faq and https://docs.openclaw.ai/channels/troubleshooting

sour stag Apr 27, 2026, 5:24 PM

#

these latencies are normal?

13:24:05-04:00 info Telegram: ok (@Taidigimon_bot) (231ms)
13:24:05-04:00 info Agents: main (default)
13:24:05-04:00 info Heartbeat interval: 30m (main)
13:24:05-04:00 info Session store (main): /Users/enriquemurillo/.openclaw/agents/main/sessions/sessions.json (2 entries)
13:24:05-04:00 info - agent:main:telegram:direct:5024580021 (1m ago)
13:24:05-04:00 info - agent:main:main (6m ago)
13:24:06-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ channels.status 402ms conn=902e5024…ecaf id=5988dac5…c934

finite hingeBOT Apr 27, 2026, 5:24 PM

#

Yep — those are normal/healthy.

Telegram: ok ... (231ms) = Telegram API/channel probe latency. Good.
channels.status 402ms = local Gateway RPC status call. Also fine.

Those don’t measure end-to-end reply time though. If Telegram still feels slow, the delay is probably after ingress: model/tool/session processing, or the channel restarting.

To measure the real path:

openclaw logs --follow --local-time

Then send a Telegram message and compare timestamps. For deeper timing, run the gateway with:

#

OPENCLAW_DEBUG_TELEGRAM_INGRESS=1 openclaw gateway

Look for:

telegram ingress: ... contextReadyMs=...
telegram ingress: ... dispatchCompleteMs=...

Also still paste:

openclaw --version

#

The earlier repeated starting provider lines are the bit I’d watch — if those keep happening, that points to restarts/polling instability, not normal Telegram latency.

sour stag Apr 27, 2026, 5:24 PM

#

13:24:05-04:00 info Telegram: ok (@Taidigimon_bot) (231ms)
13:24:05-04:00 info Agents: main (default)
13:24:05-04:00 info Heartbeat interval: 30m (main)
13:24:05-04:00 info Session store (main): /Users/enriquemurillo/.openclaw/agents/main/sessions/sessions.json (2 entries)
13:24:05-04:00 info - agent:main:telegram:direct:5024580021 (1m ago)
13:24:05-04:00 info - agent:main:main (6m ago)
13:24:06-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ channels.status 402ms conn=902e5024…ecaf id=5988dac5…c934
gateway connect failed: Error: gateway closed (1000):
Gateway not reachable. Is it running and accessible?
Gateway target: ws://127.0.0.1:18789
Source: local loopback
Config: /Users/enriquemurillo/.openclaw/openclaw.json
Bind: loopback
Hint: run openclaw doctor.
enriquemurillo@Mac-mini ~ %

finite hingeBOT Apr 27, 2026, 5:25 PM

#

That last part is the real problem — not Telegram latency:

gateway connect failed: gateway closed (1000)
Gateway not reachable

Telegram: ok ... (231ms) is fine. But the Gateway is closing / unreachable right after, so replies can feel slow or flaky because the service is restarting or not staying connected.

Please run these and paste the output:

openclaw --version
openclaw gateway status
openclaw doctor

#

Then restart once:

openclaw gateway restart
openclaw gateway status

And watch logs while sending one Telegram message:

openclaw logs --follow --local-time

If gateway status says the service is running but RPC/connectivity fails, that’s a Gateway/service issue. If it says not running, the fix is startup/service config rather than Telegram.

sour stag Apr 27, 2026, 5:57 PM

#

Its really invonvinient that I cant send you more than X characters

finite hingeBOT Apr 27, 2026, 5:57 PM

#

Yeah, that’s annoying 😅 Easiest workaround: don’t paste the full log inline.

Use one of these:

openclaw status --all > openclaw-status.txt

Then upload openclaw-status.txt here as a file.

Or for logs:

openclaw logs --plain --local-time > openclaw-logs.txt

Let it run while you send one slow Telegram message, press Ctrl+C, then upload the file.

#

If Discord still complains, trim it to the useful bits:

openclaw logs --plain --local-time | rg -i "telegram|gateway|error|warn|closed|restart|doctor"

Also, please include this separately because it’s short:

openclaw --version
openclaw gateway status

sour stag Apr 27, 2026, 5:57 PM

#

cant upload txt

finite hingeBOT Apr 27, 2026, 5:58 PM

#

No worries — paste short filtered chunks instead.

Run these one at a time and paste the output:

openclaw --version

openclaw gateway status

openclaw doctor 2>&1 | tail -80

Then for the useful log slice:

#

LOG="$(ls -t /tmp/openclaw/openclaw-*.log | head -1)"
rg -i "telegram|gateway|closed|restart|starting provider|polling|error|warn|crash|failed" "$LOG" | tail -80

If that’s still too long, use:

LOG="$(ls -t /tmp/openclaw/openclaw-*.log | head -1)"
rg -i "telegram|closed|restart|error|warn|failed" "$LOG" | tail -40

Paste them in multiple messages if needed. The main thing I’m looking for is why the gateway closes/restarts, because your Telegram 231ms probe looked fine.

sour stag Apr 27, 2026, 5:59 PM

#

3:46:39-04:00 error [tools] image failed: Local media path is not under an allowed directory: /tmp/chrome-oauth.png raw_params={"image":"/tmp/chrome-oauth.png","prompt":"Read the Chrome page error. What does it say and what likely caused it?"}
13:46:39-04:00 error [tools] image failed: Local media path is not under an allowed directory: /tmp/chrome-oauth.png raw_params={"image":"/tmp/chrome-oauth.png","prompt":"Read the Chrome page error. What does it say and what likely caused it?"}
13:46:46-04:00 error [tools] image failed: Failed to optimize image raw_params={"image":"/Users/enriquemurillo/.openclaw/workspace/chrome-oauth.png","prompt":"Read the Chrome page error. What does it say and what likely caused it?"}
13:46:46-04:00 error [tools] image failed: Failed to optimize image raw_params={"image":"/Users/enriquemurillo/.openclaw/workspace/chrome-oauth.png","prompt":"Read the Chrome page error. What does it say and what likely caused it?"}
13:48:12-04:00 warn agent/embedded {"subsystem":"agent/embedded"} {"event":"embedded_run_failover_decision","tags":["error_handling","failover","assistant","surface_error"],"runId":"02d0c6ce-f9da-4674-9505-cd96a31e207d","stage":"assistant","decision":"surface_error","failoverReason":null,"profileFailureReason":null,"provider":"openai-codex","model":"gpt-5.5","sourceProvider":"openai-codex","sourceModel":"gpt-5.5","profileId":"sha256:9206c49937ca","fallbackConfigured":false,"timedOut":false,"aborted":true} embedded run failover decision
13:48:14-04:00 info gateway/channels/telegram {"subsystem":"gateway/channels/telegram"} telegram sendMessage ok chat=5024580021 message=499
13:51:46-04:00 info hooks/session-memory {"subsystem":"hooks/session-memory"} Session context saved to ~/.openclaw/workspace/memory/2026-04-27-gog-auth.md
13:52:01-04:00 warn memory {"subsystem":"memory"} chunks_vec not updated — sqlite-vec unavailable. Vector recall degraded. Further duplicate warnings suppressed.
13:52:09-04:00 info gateway/channels/telegram {"subsystem":"gateway/channels/telegram"} telegram sendMessage ok chat=5024580021 message=506
13:53:05-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ channels.status 165ms conn=f5df210d…0432 id=27a2edea…c844
13:53:52-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ channels.status 166ms conn=69dcb217…91fc id=55e43fa1…45dc
13:54:16-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ health 231ms conn=c1de6d35…69eb id=55e2c219…a689
13:55:18-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=150s queueDepth=1
13:55:48-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=180s queueDepth=1
13:55:59-04:00 error Error: TUI exited from signal SIGABRT
13:56:18-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=210s queueDepth=1
13:56:22-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ channels.status 169ms conn=815de716…5bf7 id=e3097cd0…f18f
13:56:23-04:00 info gateway/ws {"subsystem":"gateway/ws"} ⇄ res ✓ health 234ms conn=c0c5f892…b2b2 id=c5e0dec4…af25
13:56:48-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=240s queueDepth=1
13:57:18-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=270s queueDepth=1
13:57:39-04:00 warn memory {"subsystem":"memory"} chunks_vec not updated — sqlite-vec unavailable. Vector recall degraded. Further duplicate warnings suppressed.
13:57:48-04:00 warn diagnostic {"subsystem":"diagnostic"} stuck session: sessionId=main sessionKey=*** state=processing age=300s queueDepth=1

finite hingeBOT Apr 27, 2026, 5:59 PM

#

This log explains it: Telegram is not the slow part. Telegram sends are OK, health/RPC is ~165–234ms.

The problem is here:

stuck session: state=processing age=150s ... 300s queueDepth=1
TUI exited from signal SIGABRT

Your main session is wedged processing a run, so Telegram messages queue behind it.

Try this in Telegram as a standalone message:

stop

#

or:

/stop

Then start fresh:

/new

If Telegram won’t respond, restart the gateway:

openclaw gateway restart

#Openclaw telegram so slow