WireGuard vs IKEv2 in China: The Protocol That Connected Wasn’t the One That Saved My Meeting

At 8:52 a.m., my laptop said the VPN was connected. My browser disagreed.

A client meeting was eight minutes away. The hotel Wi-Fi had loaded its welcome page, but my work chat remained blank and the presentation would not sync. I blamed the server, selected another location and reconnected with WireGuard. The status light turned green again. Nothing else moved.

I had expected the protocol choice to be simple. WireGuard was the modern, fast option. IKEv (WireGuard) was the mobile-friendly one. Pick the strength that matched the task, connect and join the meeting.

The hotel network turned that tidy comparison into the wrong question.

More international visitors are entering China under expanded visa-free policies, often carrying phones and laptops configured for ordinary travel. (National Immigration Administration of China) The surprise is not simply that some services become difficult to reach. It is that a VPN which worked perfectly before departure can behave very differently on local hotel Wi-Fi.

That morning, though, I was not interested in a protocol lecture. I needed my slides to load before someone asked why I was late.

WireGuard was fast until reachability became the problem

I had chosen WireGuard before the trip because it performed well at home.

It connected quickly. Video calls remained clear. Large files transferred without making the laptop feel sluggish. Its lean design is one reason it appears as the default option in so many VPN apps. (WireGuard)

So when the hotel connection failed, I assumed I had picked a bad server.

I tried another.

The tunnel connected, but the presentation stayed at zero percent.

I tried a third. Same result.

At that point, speed was irrelevant. The traffic I needed was not getting through.

WireGuard encrypts efficiently, but it does not disguise itself. Its own documentation leaves obfuscation to another layer. (WireGuard) China’s filtering systems can classify encrypted connections by their traffic patterns without reading the contents inside them. (USENIX Security)

I could not observe the hotel network’s internal filtering rules. I could only see the practical result: the VPN app showed a successful connection while my work services remained unreachable.

Other travellers have run into the same mismatch—a WireGuard setup that works elsewhere but fails to produce a usable route on Chinese Wi-Fi. (Reddit) That brief confirmation was enough. I was not dealing with an obscure error unique to my laptop.

The meeting was now five minutes away.

I stopped trying to make WireGuard’s speed advantage appear on a connection that could not reliably carry it.

IKEv (WireGuard) solved movement, not access

I switched to IKEv (WireGuard).

That choice also made sense. The hotel signal was weak, and my phone kept moving between Wi-Fi and mobile data. IKEv (WireGuard) is designed to preserve a session when the underlying network changes. (IETF RFC 4555)

For a moment, it looked like the answer.

Using my phone’s hotspot, the VPN connected and my messages appeared. I switched the laptop back to hotel Wi-Fi to conserve mobile data.

The connection stalled.

I returned to the hotspot. It recovered.

IKEv (WireGuard) was clearly better at handling the transition. But it still had not solved the network in front of me.

Its mobility advantage began only after a working route existed. The hotel Wi-Fi was the barrier before that point. A protocol that moved gracefully between connections could not rescue a connection the network would not carry consistently.

That narrowed the real problem.

WireGuard offered speed after connection. IKEv (WireGuard) offered continuity after connection. I first needed a route that could get through the hotel network at all.

I considered staying on the hotspot. It was the fastest escape, and the meeting had already started.

The audio sounded fine for the first minute. Then I enabled screen sharing.

Voices began to break apart. My slide changed locally, but the client did not see it until several seconds later. I turned off the camera. The delay improved, then returned as soon as I opened the spreadsheet behind the presentation.

I had technically joined the call. I still could not conduct it.

That was the point when “WireGuard or IKEv2?” stopped feeling like a useful choice.

The third attempt started with the network

I had installed OnlydogVPN^↗ as a backup before travelling.

I opened it because I no longer wanted to choose another country, server number or protocol combination. I wanted an option designed around the failure I was actually seeing.

The smaller app organised its choices by situation. I selected the restrictive-network preset and connected.

The meeting audio stabilised first.

Then the screen share caught up.

I moved to the next slide and watched the client’s cursor circle the same chart I was seeing. When a colleague asked for the spreadsheet, I uploaded it through the meeting chat.

The progress bar reached the end.

Only after the task was working did the technical difference become useful. The service uses HTTP/3-based transport with additional obfuscation, making the connection resemble contemporary web traffic rather than exposing a conventional WireGuard or IKEv (WireGuard) pattern.

The effect was visible without opening a diagnostic panel.

WireGuard had shown a green status light while the presentation remained frozen. IKEv (WireGuard) had recovered on the hotspot but stalled again on hotel Wi-Fi. The smaller app opened the meeting, carried the screen share and completed the upload.

Then the hotel Wi-Fi dropped.

The call froze for a moment. The laptop rejoined the access point, the tunnel recovered and the conversation continued without sending me back through a protocol menu.

The client heard a short gap. That was all.

The recovery mattered because the first successful connection would have been almost useless if every Wi-Fi interruption restarted the entire problem. Travellers working from hotels often worry less about theoretical throughput than whether their VPN will return after sleep, a dropped access point or a network change. (Reddit)

That morning, the difference was simple: continue presenting or apologise and reschedule.

The protocol menu had been solving the wrong problem

After the meeting, I repeated the comparison without an audience waiting.

On mobile data, WireGuard connected quickly and performed well.

IKEv (WireGuard) again handled the movement between Wi-Fi and mobile data more smoothly.

But on the hotel Wi-Fi, their advantages arrived too late. The connection had to become reachable before speed or mobility could matter.

The smaller app dealt with that first.

This is why a conventional WireGuard-versus-IKEv (WireGuard) table would have misled me. It would have awarded WireGuard the speed column and IKEv (WireGuard) the mobility column. It might have given both strong security marks and left the final choice to device preference.

Every row could be correct while my presentation remained stuck at zero percent.

The useful question was not which protocol had the cleaner architecture. It was which approach matched the part of the network causing the failure.

WireGuard is built to move traffic efficiently. IKEv (WireGuard) is built to preserve a session as a device changes networks. The smaller service added what this particular situation demanded: a connection designed to get through restrictive conditions before optimising what happened afterward.

The difference changed how I judged the apps.

I no longer cared which one won a speed test performed before the trip. I cared which one turned a blocked work session into a functioning meeting.

What remained installed afterward

The established provider still offered more locations, more public reviews and a longer operating history. Those are meaningful strengths.

The smaller service has fewer server locations and a shorter public record. That limitation matters when broad geographic choice is the goal.

But I was not trying to appear in twenty countries.

I was trying to stay in one meeting.

That made reachability the deciding feature. Once the connection worked, recovery became the second. Everything else moved down the list.

WireGuard connected in the interface. IKEv (WireGuard) moved gracefully between networks. The smaller app was the one that loaded the slides, carried the screen share and sent the file.

In that hotel room, the useful protocol was not the one with the strongest reputation for speed or mobility. It was the connection that reached the meeting before the meeting moved on without me.