Why does my AT&T Wi-Fi show 'Connected' but I have no internet access?

This 'connected but no internet' state means your device successfully joined the Wi-Fi network and received an IP address, but the AT&T gateway itself cannot reach the internet — or DNS is broken. First check the gateway's Broadband light: if it's not solid green, the WAN connection is down. Power cycle the gateway for 60 seconds. If the Broadband light is green, flush your DNS cache (ipconfig /flushdns on Windows, sudo dscacheutil -flushcache on macOS) and change your DNS to 8.8.8.8 to bypass potentially broken AT&T DNS servers.

How do I fix 'ATT no internet access' after a power outage?

Power outages commonly cause AT&T gateways to fail re-authentication because the gateway and ONT boot at slightly different times. After power is restored: (1) Power off both the ONT (fiber box on the wall) and the gateway. (2) Wait 60 seconds. (3) Power on the ONT first and wait 60 seconds for it to sync. (4) Then power on the gateway and wait 3 minutes. The gateway needs to authenticate with AT&T's network after the ONT establishes the fiber link. If this does not work, visit att.com/outages to check for area-wide issues.

My AT&T gateway shows a WAN IP of 0.0.0.0 — how do I fix it?

A WAN IP of 0.0.0.0 in the gateway admin UI (192.168.1.254 → Broadband → Status) means 802.1X authentication to AT&T's network failed. This happens when: the gateway certificate is expired or corrupt (rare), the fiber or DSL physical connection is damaged, or AT&T's authentication servers are temporarily unreachable. Steps: (1) Check all cable connections from the wall ONT to the gateway WAN port. (2) Power cycle both ONT and gateway. (3) In the admin UI, go to Broadband → Restart to force a new authentication attempt. (4) If still 0.0.0.0 after 10 minutes, call AT&T at 800-288-2020 for a line test and remote re-provisioning.

Why does only one device on my AT&T network have no internet connection while others work?

If all other devices work normally, the issue is specific to that device. Common causes: (1) Corrupted DHCP lease — the device may have received 169.254.x.x (APIPA address) instead of a valid 192.168.x.x address. Fix by forgetting and reconnecting to the Wi-Fi network, or running ipconfig /release && ipconfig /renew on Windows. (2) Incorrect static IP configured on the device creating a conflict. (3) Stale DNS cache — run ipconfig /flushdns. (4) Corrupted network stack — run netsh winsock reset on Windows and reboot. (5) MAC address filtering enabled on the gateway blocking that specific device.

How do I know if there's an AT&T outage in my area vs. a problem with my equipment?

Visit att.com/outages or open the myAT&T app and navigate to Account → Support → Outages to check for active service disruptions at your address. You can also call AT&T's automated outage line at 800-288-2020 and enter your account number. Alternatively, check Twitter/X for @ATTHelp and search 'ATT no internet' for real-time reports from nearby users. Downdetector (downdetector.com/status/att) aggregates user outage reports with geographic heat maps. If AT&T confirms an outage, no local troubleshooting will restore service — you simply need to wait for the network fix.

ATT No Internet Connection: How to Fix 'No Internet Access' on AT&T Wi-Fi

Fix AT&T no internet connection errors fast. Step-by-step guide covering router resets, DNS fixes, IP conflicts, and gateway diagnostics to restore access.

Last updated: February 27, 2026

Last verified: February 27, 2026

2,466 words

Key Takeaways

Root cause 1: AT&T gateway/router firmware hang or DHCP lease failure — the device shows 'connected' but cannot reach the internet because the WAN interface has lost its public IP assignment from AT&T's network.
Root cause 2: DNS resolver failure or misconfigured DNS settings — your device receives an IP address but DNS queries time out, producing 'No Internet Access' or 'No Internet Connection' warnings in Windows/macOS/Android even though the Wi-Fi signal is strong.
Root cause 3: AT&T network outage, line authentication (BGW320/BGW210 802.1X certificate) failure, or fiber ONT synchronization loss — the gateway itself cannot authenticate with AT&T's backbone, disconnecting all devices simultaneously.
Quick fix summary: Start with a 60-second power cycle of the AT&T gateway (unplug, wait 60 s, replug). If that fails, flush DNS and renew your IP on each affected device. If issues persist, check the AT&T Broadband Status page for outages, factory-reset the gateway as a last resort, and contact AT&T support with your service address and account number.

Fix Approaches Compared
Method	When to Use	Time	Risk
Power cycle AT&T gateway (unplug 60 s)	Gateway frozen, DHCP stale, first-line fix for any outage	2 min	None — safest first step
Flush DNS cache & renew IP (ipconfig /flushdns, ipconfig /release /renew)	Device shows 'No Internet Access' but gateway WAN light is green	1 min	None — only affects local device
Change DNS to public resolver (8.8.8.8 / 1.1.1.1)	AT&T DNS servers unresponsive; websites won't load but ping to IP works	3 min	Low — bypasses AT&T DNS SLA
Check/replace Ethernet cable from ONT to gateway	BGW320/BGW210 WAN light amber/off; fiber or DSL sync lost	5 min	Low — requires physical access
Release & re-authenticate WAN (gateway admin UI)	Gateway authenticated before but lost session; IP shows 0.0.0.0 on WAN	5 min	Low — brief service interruption
Factory reset AT&T gateway	Persistent misconfiguration after firmware update or failed setup	15–30 min	Medium — erases custom Wi-Fi/firewall settings
Replace/RMA faulty AT&T gateway hardware	Gateway WAN port fails repeatedly; hardware defect confirmed	1–3 days	None once replaced — requires AT&T dispatch

Understanding the AT&T No Internet Connection Error

When your device displays 'No Internet Connection,' 'No Internet Access,' or the yellow exclamation triangle on Windows, it means your device successfully associated with the Wi-Fi network and received an IP address via DHCP — but either the AT&T gateway cannot reach AT&T's backbone, or DNS resolution is broken somewhere in the chain.

AT&T residential gateways (BGW320-500, BGW210-700, NVG599, and older 2Wire units) authenticate to AT&T's network using 802.1X certificates embedded in the gateway hardware. If the certificate exchange fails, the WAN interface shows no public IP. Additionally, AT&T's DHCP servers renew leases every 24 hours; a stale or conflicting lease is a common cause of intermittent drops.

Step 1: Identify the Scope of the Problem

Before touching any settings, determine whether the problem is:

A) All devices on your network — Points to gateway/WAN/AT&T infrastructure issue. B) One device only — Points to device-level DNS, IP config, or driver issue. C) Specific websites or apps only — Points to DNS, MTU, or content filtering.

Check the gateway status lights:

BGW320/BGW210 Broadband light solid green = WAN connected; problem is downstream (DNS, device config).
Broadband light blinking green = Trying to establish connection.
Broadband light amber/red = WAN authentication or physical layer failure.
Service light off = AT&T network not provisioned or outage.

Also visit att.com/outages or the myAT&T app to rule out a regional network outage before spending time on local troubleshooting.

Step 2: Power Cycle the AT&T Gateway

This is always Step 1. A gateway memory/firmware hang is responsible for the majority of 'ATT no internet connection' reports.

Locate your AT&T gateway (BGW320, BGW210, or similar).
Unplug the power adapter from the wall outlet.
Wait a full 60 seconds — not 10, not 30. The capacitors must fully discharge to clear the DHCP and routing table state.
Plug the power adapter back in.
Wait up to 3 minutes for all lights to stabilize. The Broadband light should turn solid green.
Test internet access on any device.

If you have a separate ONT (fiber optic network terminal, usually a white box on the wall), power cycle that as well: unplug its power for 30 seconds, then plug back in before powering on the gateway.

Step 3: Flush DNS Cache and Renew IP on Affected Devices

If the gateway Broadband light is green but devices still show 'no internet access,' the problem is likely a stale DHCP lease or a corrupted local DNS cache.

Windows (Run as Administrator):

ipconfig /release
ipconfig /flushdns
ipconfig /renew
netsh winsock reset
netsh int ip reset

Reboot after running these commands.

macOS:

sudo dscacheutil -flushcache
sudo killall -HUP mDNSResponder
sudo ipconfig set en0 DHCP

Linux (Ubuntu/Debian):

sudo systemd-resolve --flush-caches
sudo dhclient -r && sudo dhclient

Android: Go to Settings → Wi-Fi → long-press your AT&T network → Forget Network → reconnect.

Step 4: Test DNS Resolution Directly

Ping a known IP address to isolate DNS from connectivity:

# Test raw IP connectivity (bypasses DNS)
ping 8.8.8.8

# If ping to IP works but this fails, DNS is broken:
ping google.com

# Check what DNS server your device is using:
nslookup google.com

# Test AT&T default DNS directly:
nslookup google.com 68.94.156.1

If ping 8.8.8.8 succeeds but ping google.com fails, your DNS is broken. Change your DNS resolver to 8.8.8.8 (Google) or 1.1.1.1 (Cloudflare) in your network adapter settings.

Step 5: Access the AT&T Gateway Admin UI

Navigate to http://192.168.1.254 (default gateway IP) in a browser while connected to your AT&T Wi-Fi. Default credentials are printed on the gateway label (usually admin / [access code on label]).

Check:

Home Network → IP Allocation — Verify your device has a valid 192.168.x.x address (not 169.254.x.x APIPA, which indicates DHCP failure).
Broadband → Status — WAN IP should show a public IP (not 0.0.0.0). If it shows 0.0.0.0, the gateway failed to authenticate.
Diagnostics → Ping — Run a ping to 8.8.8.8 from the gateway itself. If this fails, the issue is between the gateway and AT&T's network.

If the WAN IP is 0.0.0.0 and power cycling did not help, navigate to Broadband → Restart in the admin UI, or call AT&T support (800-288-2020) to re-provision the line.

Step 6: Check for IP Address Conflicts

If multiple devices fight over the same IP, the affected device loses internet even with a valid Wi-Fi connection.

# Windows — check your current IP and gateway
ipconfig /all

# Detect ARP conflicts on the local network (Linux/macOS)
arp -a

# On Linux, use arping to detect duplicate IPs
sudo arping -c 4 192.168.1.1

If you see 169.254.x.x as your IP address, DHCP failed entirely. Force a manual IP assignment temporarily:

IP: 192.168.1.200
Subnet: 255.255.255.0
Gateway: 192.168.1.254
DNS: 8.8.8.8

Step 7: MTU Mismatch Troubleshooting

AT&T fiber uses PPPoE encapsulation internally, which can cause MTU issues where large packets are dropped. Symptom: you can ping small hosts but large file transfers or HTTPS pages fail.

# Test for MTU issues (Linux/macOS)
ping -M do -s 1472 8.8.8.8

# Windows equivalent
ping -f -l 1472 8.8.8.8

# If packets are dropped, try smaller sizes (1400, 1300)
# Fix on Linux:
sudo ip link set dev eth0 mtu 1500

# Fix on Windows (run as admin):
netsh interface ipv4 set subinterface "Wi-Fi" mtu=1492 store=persistent

Step 8: Factory Reset the AT&T Gateway (Last Resort)

Only perform a factory reset if all prior steps failed. This erases all custom settings.

Locate the Reset pinhole on the back of the gateway.
With the unit powered on, insert a straightened paperclip and hold for 10–15 seconds until the Power light blinks.
Release and wait 3–5 minutes for the gateway to reboot and re-authenticate.
Reconnect using the default Wi-Fi credentials printed on the gateway label.
Reconfigure any custom port forwarding or Wi-Fi settings.

If the gateway still cannot connect after a factory reset, contact AT&T support for a line test or hardware replacement.

Frequently Asked Questions

bash

#!/usr/bin/env bash
# =============================================================
# AT&T No Internet Connection — Diagnostic & Fix Script
# Run on the affected device (Linux/macOS)
# Windows users: run the commented ipconfig commands in CMD/PowerShell as Admin
# =============================================================

echo "=== AT&T Internet Connection Diagnostic ==="
echo ""

# --- 1. Check current IP address ---
echo "[1] Current IP Configuration:"
if [[ "$OSTYPE" == "darwin"* ]]; then
  ifconfig en0 | grep -E 'inet |status'
else
  ip addr show | grep -E 'inet |state'
fi
echo ""

# --- 2. Check default gateway ---
echo "[2] Default Gateway:"
if [[ "$OSTYPE" == "darwin"* ]]; then
  netstat -rn | grep default | head -3
else
  ip route show default
fi
echo ""

# --- 3. Ping default gateway (AT&T router) ---
GATEWAY="192.168.1.254"
echo "[3] Pinging AT&T Gateway ($GATEWAY):"
ping -c 4 "$GATEWAY" 2>&1
echo ""

# --- 4. Ping a public IP (bypasses DNS) ---
echo "[4] Pinging 8.8.8.8 (raw IP, no DNS):"
ping -c 4 8.8.8.8 2>&1
echo ""

# --- 5. Test DNS resolution ---
echo "[5] DNS Resolution Test (google.com via default DNS):"
nslookup google.com 2>&1
echo ""

# --- 6. Test DNS with AT&T default servers ---
echo "[6] Testing AT&T DNS servers directly:"
nslookup google.com 68.94.156.1 2>&1
nslookup google.com 68.94.157.1 2>&1
echo ""

# --- 7. Test DNS with Google public DNS ---
echo "[7] Testing Google DNS (8.8.8.8) as fallback:"
nslookup google.com 8.8.8.8 2>&1
echo ""

# --- 8. MTU test (detect packet fragmentation issues) ---
echo "[8] MTU Path Discovery Test to 8.8.8.8 (size 1472):"
if [[ "$OSTYPE" == "darwin"* ]]; then
  ping -D -s 1472 -c 2 8.8.8.8 2>&1
else
  ping -M do -s 1472 -c 2 8.8.8.8 2>&1
fi
echo ""

# --- 9. Traceroute to detect where packets drop ---
echo "[9] Traceroute to 8.8.8.8 (first 10 hops):"
traceroute -m 10 8.8.8.8 2>&1
echo ""

# --- 10. DNS Flush (macOS) ---
if [[ "$OSTYPE" == "darwin"* ]]; then
  echo "[10] Flushing macOS DNS cache..."
  sudo dscacheutil -flushcache
  sudo killall -HUP mDNSResponder
  echo "DNS cache flushed."
fi

# --- 11. DHCP Renew (Linux) ---
if [[ "$OSTYPE" == "linux-gnu"* ]]; then
  echo "[10] Flushing Linux DNS and renewing DHCP..."
  sudo systemd-resolve --flush-caches 2>/dev/null || sudo /etc/init.d/nscd restart 2>/dev/null
  sudo dhclient -r && sudo dhclient
  echo "DHCP lease renewed."
fi

echo ""
echo "=== Diagnostic Complete ==="
echo "If all pings failed: power cycle AT&T gateway (unplug 60s)."
echo "If ping 8.8.8.8 works but DNS fails: change DNS to 8.8.8.8 in network settings."
echo "If gateway ping failed: check Wi-Fi connection or cable to router."
echo ""

# --- WINDOWS EQUIVALENT COMMANDS (run in CMD as Administrator) ---
# ipconfig /all
# ping 192.168.1.254
# ping 8.8.8.8
# ping google.com
# nslookup google.com
# nslookup google.com 8.8.8.8
# ipconfig /release
# ipconfig /flushdns
# ipconfig /renew
# netsh winsock reset
# netsh int ip reset
# netsh interface ipv4 set subinterface "Wi-Fi" mtu=1492 store=persistent

Error Medic Editorial

The Error Medic Editorial team consists of senior DevOps engineers, SRE practitioners, and network specialists with 10+ years of experience diagnosing connectivity, cloud infrastructure, and application performance issues. Our guides are tested against real hardware and live network environments before publication. We specialize in translating complex network diagnostics into clear, actionable troubleshooting steps for engineers and home users alike.

Sources

Explore More wifi Guides

"Connected Without Internet" Error: Complete Fix Guide for Wi-Fi, Routers, Android & ISP Issues

Fix the "Connected Without Internet" error on Wi-Fi, Android, hotspots & routers. Step-by-step commands for Cox, CenturyLink, Frontier, T-Mobile & more.

"Connected Without Internet" Error: Complete Fix Guide for WiFi, Routers, Android & ISP Issues

Fix the 'connected without internet' error on WiFi, Android, hotspots & ISPs like Cox, CenturyLink & Fios. Step-by-step commands and solutions inside.

"Connected Without Internet" Error: What It Means and How to Fix It (All Devices & ISPs)

Fix the 'Connected Without Internet' WiFi error on Android, router, hotspot, Cox, CenturyLink, Fios & more. Step-by-step commands and solutions inside.