News Stay informed about the latest enterprise technology news and product updates.

The art and science of troubleshooting network problems

Troubleshooting network problems is part art and part science and requires creative use of tools and talent.

For the last few months, my notebook has been getting slower, and slower and slower. I know my way around Windows, and though I always appreciate Desktop Support's offer to re-image it, there's never time to rebuild it. So, I've just endured it.

Finally, this week I'd had enough and took the reluctant walk of shame to IT. New Guy at the helpdesk listened to my story, then took my laptop to his workbench and blew clouds of dust out of it with an air compressor. I was stunned; it's like a new machine (almost). I'd done everything I could think of, and when his troubleshooting process using air actually solved an "I/O" issue, I didn't want to believe it.

It reminded me again that troubleshooting performance problems in any technical field -- especially troubleshooting network problems -- is both art and science. Time and time again, the troika of management tools, administrator experience and methodical troubleshooting consistently yields timely resolution and happy users.

Check the cable (repeat three times)

A non-techie friend recently told a story about working with his company's helpdesk to resolve his home-office connectivity issue with a company-provided remote-branch VPN router. He proudly told me that he'd fixed it himself, following their instructions on the phone, by correcting his "reversed Ethernet polarity" problem. Suppressing a big laugh, I asked him to explain.

The most effective engineers are those who combine human and network engineering.

On the phone with the helpdesk, he'd been asked to double-check the cable between his home router and the VPN box. The helpdesk admin had paused and reluctantly told him it sounded like he had a problem with Ethernet polarity. The only solution: switch the ends of the cable. He had swapped the ends and voilà!, his problem was solved. I truly believe he had been working with one of the world's most talented support engineers.

The admin hadn't been confident that the customer had correctly seated both RJ45s, and rather than argue over assurances that they had been, he engineered a human solution that resulted in reseating both. Best of all, he had done it without making the end user feel foolish. It was a pragmatic combination of human engineering and years of experience with a single goal: getting off the phone quickly and closing the ticket.

Whether it's VLANs, ACLs, QoS or 802.1X

Whether supporting the core, becoming the master of data center switching or practicing the alchemy of turning business requirements into firewall rules, network engineers face similar challenges, only with greatly multiplied complexity. The customers often are more knowledgeable, but at the same time over-confident; and worse, some require political considerations. Because of services concentration, increased traffic and 24/7 business criticality, our systems require monitoring proactively and troubleshooting network problems almost immediately to restore access. The most effective engineers are those who combine human and network engineering.

IT managers can spot these admins easily from the systems they touch. Beyond helpdesk ticket close, reopen or reassignment rates; network performance monitor login and activity history; and low stress config management and firewall hygiene, do they occasionally stop by your office with a great improvement idea? In short, do they make it look easy?

Protect slack time to incubate creativity

Slack is perhaps the most important asset for technologists. We never have enough. We tend to overcommit, or at least underappreciate interrupt-request work cost. But more than just maintaining a buffer to keep admins out of trouble, slack is the most effective incubator of truly creative solutions that transform, not just maintain IT. We create slack with skill and automation, carefully selecting management tools and network hardware technology, and make time to learn and configure them well. Shelfware that is unneeded or encrusted with baroque features or too difficult to implement is just as detrimental to slack as overreliance on legacy or one-off manual solutions that never allow free time to invent.

Patrick Hubbard on network troubleshooting

The desktop tech who gave me a confident dude-nod while he blasted the crud out of my laptop, and the inventor of reversed Ethernet polarity share a common bond: They operate from a base of creative freedom they've brought about by using their tools and talent efficiently. They create a spare clock cycle here and there to provide outstanding service that their managers notice. Plus, they almost always bring the funny at lunch.

About the author:
Patrick Hubbard is a head geek and senior technical product marketing manager at SolarWinds. With 20 years of technical expertise and IT customer perspective, his networking management experience includes work with campus, data center, storage networks, VoIP and virtualization with a focus on application and service delivery in both Fortune 500 companies and startups in high tech, transportation, financial services and telecom industries. He can be reached at

This was last published in October 2013

Dig Deeper on Network management and monitoring

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

The key is the binary search. Keep dividing the problem in half to isolate the fault. This is difficult if it is not your network, but not impossible.
Thank you. My students will enjoy this.