I’ll be honest about this: the network troubleshooting market has a paid-tools problem. SolarWinds, Datadog, and the rest are useful if you’re at scale, but the pricing has gotten silly for small teams. A three-person IT shop doesn’t need a $30k annual monitoring contract to figure out why a VoIP call keeps stuttering.
Good news: free tools handle probably 80% of the daily stuff. Ping, traceroute, a few browser-based checkers, iPerf, Wireshark. That’s a real toolkit, and it costs zero.
Starting With the Fundamentals
Start with ping. It’s the fastest way to rule out the dumbest problems: ping 8.8.8.8, and if it doesn’t reply, the issue isn’t with the app the user is complaining about, it’s their local network or their ISP. Save yourself the detour.
Then traceroute. On Windows it’s called tracert (annoying but whatever), and you point it at the actual destination server to watch for hops with weird latency. Anything over 100ms that shouldn’t be is where your path is getting screwy.
If you’d rather run these from a browser, MXToolbox and DNSChecker do the same checks from a bunch of locations around the world. That’s the only way to settle the “it works for me” vs “it’s broken for me” argument that inevitably shows up in every incident channel.
Verifying Proxy and IP Configurations
Proxies are where I get stuck the most. Not because the proxy is actually broken (usually), but because the configuration is subtly off in a way that doesn’t throw errors: headers get stripped, the wrong country ends up on the outbound request, or WebRTC leaks the real IP even though everything else looks fine.
You catch this with header inspection. There are free web tools where you can check proxy free against a real endpoint and see exactly what your setup is sending, followed by WhatIsMyIPAddress or IPLocation to verify the country and a WebRTC leak test for the browser side. Fifteen minutes total, usually catches whatever’s wrong.
Measuring Speed and Bandwidth Honestly
Speedtest.net is fine, it’s not lying to you. But it optimizes for the nearest server and near-perfect test conditions, so it’s giving you your best possible number, not your actual working performance. Treat it as a ceiling, not an average.
Fast.com (Netflix built it) is more honest for anyone who cares about streaming, since it tests against Netflix’s own CDN. For anything past that, iPerf3 is what you want, though it needs a cooperating server on the far end. Most IT teams I’ve worked with just keep one running on a spare box for exactly this reason.
And latency matters more than bandwidth for anything real-time. A gigabit line with 200ms jitter delivers a worse video call than a 100Mbps line with a rock-steady 20ms ping. PingPlotter is great for graphing this over hours or days, so you can catch the pattern where “the network gets bad at 1:15pm” turns out to correlate with the sales team’s daily standup.
Digging Deeper With Free Analysis Tools
When ping, traceroute, and header checks all come back fine but users still say things are broken, it’s time for Wireshark. Free, open source, captures every packet crossing the wire. The learning curve is real (I won’t pretend otherwise), but there are enough community tutorials that anyone patient can pick up the common cases inside a week.
Nmap and its GUI cousin Zenmap handle port scanning. If a firewall is quietly dropping traffic (and firewalls love to do this without logging anything), Nmap tells you which ports are actually reachable. Fast, cheap, no false confidence about what’s really open.
For DNS stuff, Google’s Public DNS Dashboard and Cloudflare’s 1.1.1.1 diagnostic pages flag propagation delays and resolver issues. Knowing how DNS resolution works matters here, because DNS-caused slowness looks exactly like bandwidth-caused slowness from the user’s side. I’d estimate 60% of “the internet is slow” tickets turn out to be DNS.
Putting It All Together
The gap between free and paid troubleshooting has narrowed enough that most small IT teams could ditch half their subscriptions. What used to require a full monitoring platform now fits inside a browser and a terminal window.
Ping and traceroute cover the basics, header checks catch proxy weirdness, and speed tests keep the sales team happy when they ask. iPerf and Wireshark are for the stuff nothing else can explain. Get comfortable with those five, and most incidents stop being mysterious. Click here
