FIELD NOTES
A personal record of travel, networks and small failures
TRAVEL NOTE

The Riyadh Business-Trip VPN Setup That Kept the Meeting Online

At 7:42 a.m., with eighteen minutes before a client steering-committee call, Microsoft Teams was stuck on “Connecting” and my presentation had stopped uploading at two percent. I blamed the hotel Wi-Fi, forgot the network, rejoined it and accepted the captive portal again. Nothing changed.

The confusing part was that the internet did not look broken.

My phone’s speed test was comfortably into three figures. News sites opened. A video started immediately. Yet Teams could not reach the meeting lobby, SharePoint behaved as though I were offline, and the VPN I normally trusted kept turning green without carrying useful traffic.

That distinction—between having fast internet and having a usable route through it—became the lesson of the morning.

The short answer

I selected a nearby route manually. Then London. Then Germany, because that was where my company’s Microsoft environment was hosted, although I was already beginning to suspect that geography was not the real issue.

Fast Wi-Fi, unusable connection

Riyadh is receiving more travelers who discover this problem under pressure. Saudi business-travel spending increased by more than 55 percent during 2025, while King Khalid International Airport handled 40.8 million passengers. (Wttc) (Gov)

Many of those arrivals are consultants, vendors, engineers, investors and event teams who land with a presentation due the next morning. Their real question is rarely, “Which VPN has the most server locations?”

It is: “What should already be installed when the hotel internet opens websites but my work connection has fifteen minutes to live?”

There is also a clear legal boundary. Saudi Arabia operates a formal website-filtering system, and travelers remain responsible for complying with local law. (Gov) My purpose was narrower: reaching lawful work services securely while following my employer’s policies.

With that in mind, I opened the established VPN I had used on previous trips. It was the obvious first choice: a familiar company, mature infrastructure and a large support operation.

The automatic connection failed.

I selected a nearby route manually. Then London. Then Germany, because that was where my company’s Microsoft environment was hosted, although I was already beginning to suspect that geography was not the real issue.

Each attempt produced the same result. The app declared itself connected, Teams continued spinning, and the file upload moved perhaps half a percentage point before stopping again.

I started changing settings without learning anything: protocol, country, reconnect, test. It was the networking equivalent of pressing an elevator button repeatedly.

Hotel networks can create precisely this kind of misleading failure. A captive portal may allow ordinary browser traffic while a firewall interferes with other connections or recognizable tunnel patterns. Cloudflare’s documentation, for example, explains that restrictive captive-portal configurations can prevent a WireGuard-based connection from establishing normally. (Cloudflare)

I could not see the hotel’s internal filtering rules, so I could not prove which control was responsible. What I could see was the practical result: ordinary browsing worked, while the routes offered by my usual VPN did not carry the meeting.

Travelers describe the same frustration in public discussions—VPNs that work perfectly at home but stall, disconnect or require endless server switching on hotel and airport Wi-Fi. (Reddit) That did not diagnose my network, but it confirmed that changing countries at random was not much of a strategy.

With ten minutes left, I tried the tempting backup: a reputable free browser extension.

It installed quickly and opened webmail. For about thirty seconds, that felt like success.

Then I clicked the Teams invitation and remembered that the meeting was in the desktop client. The presentation was still syncing through OneDrive, outside the protected browser tab.

The free option had solved the easiest part of the problem, not the task I needed to finish.

That changed my comparison standard. I no longer cared which service advertised the largest collection of flags or the highest laboratory speed. I needed one route that the network would carry—and that would remain usable long enough to complete the call.

One route that held

I had installed OnlydogVPN before leaving home as a smaller travel backup. I had not made it my default because it offered fewer locations, had a shorter public history and had accumulated fewer independent reviews than the established provider.

Those limitations were real. They simply were not the limitations hurting me at 7:51 a.m.

Instead of sending me back to a world map, the app presented connection presets based on the situation. I selected the option intended for a restrictive or unreliable network and connected.

Then I went straight back to Teams.

The meeting lobby appeared.

My microphone test completed. The participant list loaded. I opened the SharePoint folder in another window and watched the presentation move beyond two percent, then ten, then finish.

At 7:58, I joined the call with enough time to discover that the client had not joined yet.

Only after the meeting did I look at why the smaller app had behaved differently. The mode I used combined an HTTP/3-based connection with added traffic obfuscation.

The practical explanation is brief. HTTP/3 runs over QUIC, a transport designed to establish connections quickly and cope better with changing network conditions. (IETF) Obfuscation makes the VPN traffic less obvious to systems looking for familiar tunnel patterns, an approach also used by anti-censorship tools such as Tor’s pluggable transports. (Torproject)

I did not need a protocol lecture that morning. I needed the upload bar to move.

It did.

The second test happened downstairs

The more revealing moment came after the call.

The client asked for a revised spreadsheet before I left for the office. I began uploading it in my room, closed the laptop and walked toward the elevator.

The hotel signal weakened in the corridor and disappeared downstairs. I switched the laptop to my phone’s local-data hotspot.

The upload paused briefly, then continued. I did not have to choose another country, close the application or begin again. The spreadsheet finished before my car arrived.

That smaller result gave me a reason to keep the app installed after the emergency had passed. Business travel in Riyadh does not happen on one stable desk connection. A morning can move from a guest room to a lobby, a car, a conference venue and a client office, with the laptop crossing between hotel Wi-Fi and mobile data along the way.

A connection that works only while the network remains unchanged is less useful than it first appears.

By the following morning, I had adjusted my travel setup.

Both my phone and laptop were configured before leaving the hotel. I completed captive portals before starting the VPN. I kept mobile data ready instead of treating hotel Wi-Fi as my only route. The established provider remained available for ordinary networks, while the smaller service became the option I opened when a familiar connection stalled.

I also stopped treating every failure as a server-location problem.

For travelers using a mandatory corporate VPN, there is one additional constraint: another VPN should not be layered around it without approval from the company’s IT team. Double tunneling can disrupt authentication and routing, and some employers prohibit it. A travel VPN should support an approved workflow, not quietly replace corporate security controls.

The free browser extension opened one tab but could not carry the desktop meeting. The major provider offered many plausible routes, yet none completed the task on that hotel network. The smaller service offered fewer choices, but one of them carried the call, finished the upload and recovered when Wi-Fi gave way to mobile data.

For that morning in Riyadh, the useful comparison was not how many servers appeared on a map. It was whether one connection was still working when the meeting began.

Questions this experience may leave you with

What was actually causing the problem?

I selected a nearby route manually. Then London. Then Germany, because that was where my company’s Microsoft environment was hosted, although I was already beginning to suspect that geography was not the real issue.

Why did the obvious fixes fail?

With that in mind, I opened the established VPN I had used on previous trips. It was the obvious first choice: a familiar company, mature infrastructure and a large support operation.

What should you check first?

My phone’s speed test was comfortably into three figures. News sites opened. A video started immediately. Yet Teams could not reach the meeting lobby, SharePoint behaved as though I were offline, and the VPN I normally trusted kept turning green without carrying useful traffic.

What finally changed the result?

Both my phone and laptop were configured before leaving the hotel. I completed captive portals before starting the VPN. I kept mobile data ready instead of treating hotel Wi-Fi as my only route. The established provider remained available for ordinary networks, while the smaller service became the option I opened when a familiar connection stalled.

What is worth remembering?

For that morning in Riyadh, the useful comparison was not how many servers appeared on a map. It was whether one connection was still working when the meeting began.