On MTS mobile data the signature is its own and easy to recognise: the connection does come up, the badge lights, and two or three seconds later traffic stops flowing. The app says so plainly: “Connected, but no traffic”. On home Wi-Fi the same app with the same settings runs fine — and that is the main clue. Below is what we measured in Moscow, why this is not a broken phone, and why changing the port does not help here.
In short. On MTS the handshake goes through, the “Protected” badge lights up — and about 2.5 seconds later traffic stops. It is not a broken phone: on Wi-Fi the very same settings run fine. Changing the port is useless, we tested it. Exactly one thing helped — masking the first packet: on 28 July the same phone on the same carrier connected and stopped dropping.
We build Mayak. The app shapes the first packet to fit the network and walks through the paths by itself — exactly what decides the outcome on MTS. 7 days free once you confirm your email, no card required. The trial is free and needs no card; without an email it is 3 days.
🍎 On iPhone Mayak works through the third-party Happ app — with the subscription link from your account. Happ is not in the Russian App Store; there incy (publisher LLC ITDEV) works instead, checked on 22 September 2026.
What it looks like, and how it differs from MegaFon
The difference between the two operators is substantial, and it tells you immediately where to look.
| Operator | Handshake | What happens next |
|---|---|---|
| MTS | succeeds | traffic dies after 2–3 seconds |
| MegaFon | never happens | a minute of “Connecting…” |
So on MTS the filter lets the start of the conversation through and intervenes in the established flow, whereas MegaFon cuts earlier — it never reaches a connection. People call both of these “it does not work”, and they are two different phenomena.
Why “connected” and “no internet” happen at the same time
This is the most common source of confusion. The badge means “the tunnel is up”: keys were exchanged and a route was built. It does not mean “data is flowing through it”. The MTS case lives precisely between those two states: the handshake passed, the badge is lit, and the flow behind it is gone. The same symptom has other, much simpler causes — five reasons a connection comes up while traffic does not.
What the measurement showed
We first saw it on 27 July 2026: on MTS in Moscow the flow died roughly 2.5 seconds after a successful handshake, consistently, attempt after attempt. Neither the phone, nor the account, nor the chosen country changed it.
On 28 July the same phone on the same operator, running an updated app with first packet mimicry enabled, connected and kept working. Exactly one thing changed — the shape of the traffic.
⚠️ An honest caveat: we did not measure throughput in that run, so there are no MTS speed numbers here and we will not invent any. One thing was established — the connection stopped dying.
“Is MTS blocking or throttling VPNs?” — what our measurement shows
That is how the case is most often described. The connection was not refused: the handshake went through, the tunnel came up, the icon lit. Had it been about the server address or the account, even that would not have happened. What died was the flow — roughly 2.5 seconds after everything was already up. And the same tunnel with first-packet mimicry, on the same day, lived and worked. So what differed was neither the address nor the port, but the shape of the traffic. The practical conclusion: looking for “another server” is pointless.
Changing the port is pointless
This was verified on MegaFon rather than MTS, but verified directly: a profile differing from the broken one only by port 443 did not work, while mimicry on an ordinary high port did. The advice “switch to 443”, which every search returns first, is not supported by a single measurement.
Why the same thing works on Wi-Fi
This is the first thing people notice: at home everything is alive, on mobile data it is not. The difference is neither the phone nor the plan — the path to the server simply runs through different equipment, and the cellular path has something on it that a home line does not.
We counted how common this is from our own connection log over 30 days (measured 5 September 2026): from mobile data 88.8 % of attempts succeed (309 of 348), from Wi-Fi 94.8 % (257 of 271). The difference is real, but note the other side of it: mobile data works nine times out of ten. If yours fails every time, the cellular network itself is probably not the cause, and it is worth working through the list of reasons why no app connects — the first of them is a taken slot in the system, and almost nobody checks it.
⚠️ That number is counted over our own users and our own app across one month, and our own devices are in that log too. It is an order of magnitude, not national statistics. The general write-up across all networks — why it fails on mobile data but works on Wi-Fi.
What to do when MTS drops after a couple of seconds
- Check whether it repeats. If the connection dies at roughly the same second every time, it is not radio and not chance.
- Compare with Wi-Fi. On a home network the same connection normally lives indefinitely — why a mobile network behaves differently.
- Let the app try its paths. Mayak does not get stuck on one route — how the ladder works.
- Confirm the connection is alive, not merely “connected”. Four checks in a minute.
Mayak does this by itself. The app does not stick to one route: it notices the drop within a couple of seconds and moves on to the next path by itself. We never ask for payment details. A confirmed email opens the access — 7 free days. An email is optional: without one the trial is 3 days.
🍎 On iPhone Mayak works through the third-party Happ app — with the subscription link from your account. Happ is not in the Russian App Store; there incy (publisher LLC ITDEV) works instead, checked on 22 September 2026.
Short version: on MTS what breaks is not the connection but the established flow. That is why reconnecting helps for a couple of seconds and changing the port does not help at all.
Short answers
Is MTS blocking or throttling VPNs? In our measurement the connection came up and the flow died — about 2.5 seconds after the handshake. The same tunnel with first-packet mimicry worked that same day. So what decided it for us was the shape of the traffic, not the server address and not the port — see above.
MTS blocked my VPN — what do I change in the settings? From what measurement supports: changing the port is pointless, mimicry helps. In Mayak it is on by default, and “Settings” → “Check connection” shows in a minute which path is not getting through — instead of guessing.
VPN stopped working on MTS — why exactly today? What the operator changed, and when, we do not know and will not invent. What we do know is the check that separates the causes: try the same connection over Wi-Fi. Works on Wi-Fi — it is the cellular network; fails everywhere — the cause is on the phone and is covered separately.
Why does it fail on MTS mobile data while Wi-Fi is fine? The path to the server runs through different equipment. Over 30 days of our own log, mobile data succeeds 88.8 % of the time and Wi-Fi 94.8 % — numbers and their caveats. If yours fails every time, the cellular network is probably not the cause.
VPN fails on MTS and the internet cuts out — same thing? No, and telling them apart is easy. If the internet drops without the tunnel on, that is connectivity, not the VPN. If everything is alive without the tunnel and goes silent a couple of seconds after it comes up, that is exactly the case covered here: “connected” and “no internet” at the same time.
Do your numbers hold everywhere? No. The runs were made IN MOSCOW in late July 2026, and the success share comes from our own log over the 30 days to 5 September. In another city and another month the picture may differ — which is why we always say where and when we measured. We did not measure throughput on MTS, and there are no numbers for it here.