"The internet is down" is one of the most common things a support technician hears, and one of the vaguest. It could mean a loose cable, a bad Wi-Fi connection, a DHCP problem, a DNS failure, a firewall rule, or one website having a bad day.
These are the study notes I use to practice a consistent routine: start at the bottom of the stack and work up, using the built-in Windows tools. The goal at the desk usually isn't to fix the network. It's to figure out quickly whether the problem is this device, this connection, or something bigger, and then either fix it or escalate it with useful information.
Step 1: Ask a few questions first
- Is it one site, every site, or one internal app?
- Is it just this user, or are people nearby affected too?
- Is the device on Wi-Fi or wired? On VPN?
- Did anything change recently: a new location, dock, or update?
If the whole floor is down, the problem isn't on this PC, and it's time to escalate.
Step 2: Check the physical layer (and the obvious)
- Wired: Is the cable seated at both ends? Is the link light on at the port or dock? Try a known-good cable or port. With USB-C docks, try plugging the network cable directly into the laptop if it has a port, or use a different adapter.
- Wireless: Is Wi-Fi turned on? Is airplane mode off? Is the device connected to the right network (not a guest or neighboring SSID)?
netsh wlan show interfaces shows the connected SSID, signal strength, radio type, and channel, which helps when someone is on the edge of coverage.
Step 3: ipconfig, the first real data point
Run ipconfig /all and look at the active adapter [1]:
- IPv4 address. An address that starts with 169.254 is a link-local (APIPA) address. Windows assigns one when it can't get an address from a DHCP server [2], so a 169.254 address points to a DHCP or connectivity problem, not DNS.
- Default gateway. If it's empty, the device can't leave its local subnet.
- DNS servers. Are they the ones the organization expects? An unexpected DNS server can mean a misconfiguration, or the device has picked up settings from a different network.
- DHCP enabled / lease times. Useful for knowing whether the address was assigned automatically or set by hand.
Common follow-ups [1]:
ipconfig /releaseandipconfig /renewto request a new DHCP lease.ipconfig /flushdnsto clear the local DNS resolver cache.
Step 4: ping, working outward
ping sends ICMP echo requests [3]. I like to work outward:
ping 127.0.0.1checks that the local TCP/IP stack responds.ping <default gateway>checks the local network segment.ping <a known internal server>checks the path inside the organization.ping <an external host>checks the path out to the internet.
Note where it fails. One caveat: many devices and firewalls block ICMP on purpose, so a failed ping on its own doesn't prove a host is down. I treat it as one clue among several.
Step 5: nslookup, separating DNS from everything else
If pinging an IP address works but the name doesn't resolve, DNS is the prime suspect. nslookup queries DNS directly [4]:
nslookup intranet.example.orgshows which DNS server answered and what address came back.nslookup intranet.example.org <other DNS server>lets you compare answers from a different server.
If one server resolves the name and another doesn't, that's very useful to put in an escalation ticket. If nothing resolves at all, go back and check the DNS servers listed in ipconfig /all.
Step 6: tracert, finding where the path stops
tracert shows each hop between the device and a destination [5]. It can help show whether traffic stops inside the building, at the edge of the network, or farther out. Hops that show * * * aren't always a problem, because many routers don't answer these probes. What matters most is whether the trace eventually reaches the destination.
For a specific service, PowerShell's Test-NetConnection can check a TCP port, for example Test-NetConnection server.example.org -Port 443 [6]. This helps when ping is blocked but you need to know whether a web service is reachable.
Wi-Fi vs. wired: a quick comparison
One of the fastest ways to narrow things down is to test the other connection type:
- Wired works, Wi-Fi doesn't: The problem is likely the wireless adapter, its driver, signal, authentication (certificates, credentials, WPA3 support), or the access point.
- Wi-Fi works, wired doesn't: Look at the cable, port, dock or adapter, or the switch port configuration (an escalation item).
- Neither works: Look at the device itself (drivers, VPN client, security software, proxy settings) or a wider outage.
What to put in the ticket
If I need to escalate to the network team, I'd include:
- Device name, user, location, and connection type
ipconfig /alloutput (or the key lines)- Which ping targets worked and which failed
nslookupresults, including which DNS server answered- Whether others nearby are affected
- What I already tried (renew, flush DNS, different cable, different network)
A clear escalation saves the next person from repeating the same tests, and it usually gets the problem fixed faster.
Sources
- Microsoft Learn, "ipconfig": https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/ipconfig
- IETF RFC 3927, "Dynamic Configuration of IPv4 Link-Local Addresses": https://www.rfc-editor.org/rfc/rfc3927
- Microsoft Learn, "ping": https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/ping
- Microsoft Learn, "nslookup": https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/nslookup
- Microsoft Learn, "tracert": https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/tracert
- Microsoft Learn, "Test-NetConnection": https://learn.microsoft.com/en-us/powershell/module/nettcpip/test-netconnection
I'm Rıza Korkusuz, an IT support professional in San Diego. I write these notes to organize what I'm learning about desktop and service desk work. More at rizakorkusuz.com/it-support and github.com/rizakorkusuz.