Fix "deleteWebhook failed: network request failed"
Telegram API Unreachable at Startup
When OpenClaw initializes a Telegram skill, it calls deleteWebhook to clear any stale webhook before registering a new one. A network failure at this step means the bot cannot start.
This error occurs when OpenClaw's Telegram skill cannot reach api.telegram.org. The most common causes are an invalid bot token, a blocked network, or a missing proxy configuration.
Next Step
Fix now, then reduce repeat incidents
If this issue keeps coming back, validate your setup in Doctor first, then harden your config.
Jump to Fix
What the Error Looks Like
Step 1 — Verify Your Bot Token
An invalid or revoked token causes Telegram to reject the request before any network issue can be observed. Test your token directly:
curl "https://api.telegram.org/bot<YOUR_BOT_TOKEN>/getMe"
Expected: {"ok":true,...}
If you get {"ok":false,"error_code":401} your token is invalid — regenerate it via @BotFather using /revoke then /token.
Update your skill config with the new token:
openclaw config set skills.telegram.botToken "YOUR_NEW_TOKEN" openclaw gateway restart
Step 2 — Test Telegram API Connectivity
Run this from the same machine and user account that runs OpenClaw:
# Basic reachability curl -I https://api.telegram.org # With timeout (should respond in < 5s) curl --connect-timeout 5 https://api.telegram.org/bot<TOKEN>/getMe
curl: (6) Could not resolve host: api.telegram.orgDNS resolution is failing. See the DNS section below.
curl: (28) Connection timed out after 5000msTelegram API is blocked by your firewall or ISP. Common in mainland China and some corporate networks.
curl: (35) SSL connect errorTLS interception by a corporate proxy. Configure proxy settings or use a VPN.
Step 3 — Configure a Proxy or VPN
If api.telegram.org is blocked in your region or network, configure an HTTP/HTTPS proxy for OpenClaw:
# Set outbound HTTP proxy openclaw config set network.proxy "http://127.0.0.1:7890" # Or set via environment variable export HTTPS_PROXY="http://127.0.0.1:7890" openclaw gateway restart
Common proxy ports by tool
:7890— Clash / ClashX default:1080— shadowsocks / v2ray default:8080— corporate proxy (check IT):10809— v2rayN Windows default
DNS Resolution Failures
If api.telegram.org fails to resolve, try a public DNS server:
# Test with explicit DNS server nslookup api.telegram.org 8.8.8.8 # Linux: temporarily use Google DNS echo "nameserver 8.8.8.8" | sudo tee /etc/resolv.conf # Restart OpenClaw after changing DNS openclaw gateway restart
Firewall and Outbound Rules
OpenClaw needs outbound HTTPS (port 443) to 149.154.160.0/20 and 91.108.4.0/22 (Telegram's IP ranges). On Linux servers:
# Allow Telegram API servers outbound sudo iptables -A OUTPUT -d 149.154.160.0/20 -p tcp --dport 443 -j ACCEPT sudo iptables -A OUTPUT -d 91.108.4.0/22 -p tcp --dport 443 -j ACCEPT # Verify connectivity after curl --connect-timeout 5 https://api.telegram.org
Running OpenClaw as a systemd Service?
When running as a systemd service, environment variables like HTTPS_PROXY are not inherited from your shell. Add them directly to the service file:
# Edit your openclaw service file sudo systemctl edit openclaw-gateway.service # Add to [Service] section: [Service] Environment="HTTPS_PROXY=http://127.0.0.1:7890" Environment="NO_PROXY=localhost,127.0.0.1" # Reload and restart sudo systemctl daemon-reload sudo systemctl restart openclaw-gateway
Still Not Working?
Run the Doctor
Or use the web-based Error Doctor — paste the full error to get a specific fix.
Related Telegram Issues
These errors often appear together:
Did this guide solve your problem?