The pharmacist’s QR code scanned correctly, the amount was right, and my UPI PIN was not the problem. The payment app simply stopped at a connection error while a queue formed behind me. I blamed the shop’s Wi-Fi, switched to mobile data, and tried again. Same result.
The small VPN symbol in my status bar should have been the obvious clue. It was not, because the payment app had worked earlier in the day and the VPN had faded into the background.
I had installed it for a completely different reason. When access to Telegram became unreliable, I still needed to reach a work group containing a delivery address and a voice note from a client. Recent access disruptions in India had also pushed more people toward VPN apps, often for immediate, practical reasons rather than abstract privacy concerns. (RFC 9000)
Once the VPN was running, I left it running.
That habit only became a problem when I reached the pharmacy counter.
UPI is used too frequently for this kind of failure to feel like a minor software inconvenience. NPCI recorded more than 22.7 billion UPI transactions in June 2026. At a chemist, café, auto stand or apartment gate, the useful question is not whether the app might work after twenty minutes of troubleshooting. The useful question is whether the payment can be completed before the person holding the QR code loses patience.
My first response was to open the large VPN service I already trusted.
It was the sensible choice. The company had years of public history, a polished interface, broad server coverage and more support material than most smaller providers. I assumed the payment app disliked seeing a foreign IP address, so I changed from a nearby Asian server to an Indian location.
The VPN reconnected.
The UPI app still refused to proceed.
I tried another Indian server. Then a different protocol. Each attempt felt reasonable in isolation, but none changed the result. More options were giving me more things to test, not a faster route to the payment confirmation.
That was the moment the problem changed.
I was no longer trying to find the “right” VPN server. I was trying to understand whether the payment app objected to the server address or to the presence of the VPN itself.
On Android, an app does not need to rely only on an IP address to notice a VPN. The operating system exposes information about the active network, including whether traffic is using a VPN transport. That means changing countries or disguising the tunnel from the outside network may not make the VPN invisible to another app installed on the same phone.
UPI apps are also built around device-level security checks. NPCI describes device binding as part of the authentication model, and Google Pay says it may restrict service when a phone configuration fails the security requirements it applies under NPCI’s mobile application framework.
I could not see the payment app’s internal filtering rules, so I could not prove which individual check had triggered the error. But I no longer needed to guess my way through a server list. There was a faster test.
I disconnected the VPN.
Then I reopened the payment app, scanned the same QR code and entered the same PIN.
The payment completed.
That simple result was more useful than any protocol comparison. The phone, bank account, QR code and UPI PIN had not changed. The VPN state had.
Public discussions from BHIM and other UPI users describe the same practical frustration: the payment app asks for the VPN to be disabled, leaving the user to choose between keeping a privacy or access tool active and completing an immediate transaction. The important point is not that every UPI app behaves identically. It is that this conflict is common enough to deserve a straightforward response.
When a UPI app stops working while a VPN is active, the first move should not be reinstalling the banking app or cycling through ten countries.
Disconnect once. Retry the payment on a stable connection.
When the transaction succeeds, the fastest path is clear: let the payment app operate outside the tunnel, then restore the VPN afterward. When it still fails, the problem may be elsewhere—the bank, the recipient, the UPI service or the phone’s network connection.
At the pharmacy, the payment problem was over. My original access problem was not.
The moment I stepped outside, I opened Telegram again. Without the VPN, the work group still would not load. The address I needed was trapped behind a connection I had just disabled.
So I returned to the large provider.
The app began reconnecting while my phone moved from the pharmacy’s weak Wi-Fi to mobile data. The connection stalled. I changed servers, waited, changed protocols and waited again.
The service was not bad. Its strengths were simply mismatched to the moment. A huge server catalogue is valuable when I know which location I need. It is less useful when I am standing on the pavement thinking: I have paid; now show me the message.
I briefly considered installing a free VPN. That would have cost nothing and taken less than a minute. But it would also have introduced another round of uncertainty: which permissions it wanted, how crowded the route was, whether it would survive the network change and whether I would have to repeat the process the next time UPI asked me to disconnect.
I did not need a VPN that promised to remain active inside every payment app.
I needed one that was painless to leave and quick to restore.
That distinction led me back to OnlydogVPN, a smaller app I had previously kept as a travel backup.
Instead of starting with a country map, it presented the connection around the situation I was trying to solve. I selected the relevant preset and connected.
Telegram opened.
The group appeared, the delivery address loaded, and the voice note downloaded. I copied the address before the phone had finished settling onto mobile data.
The main result was not that the VPN forced its way through the UPI app. It did something more useful: it made the necessary interruption short.
Disconnect. Pay. Reconnect. Continue.
Its HTTP/3-based transport helped with the part that had frustrated me after leaving the pharmacy. HTTP/3 runs over QUIC, which is designed to recover more gracefully when a device changes network paths, such as moving from Wi-Fi to mobile data. I did not have to rebuild the connection manually each time the phone changed networks.
That recovery mattered more than the number of countries shown in the app.
A payment failure creates two separate jobs. First, complete the UPI transaction without the VPN. Second, restore access to whatever required the VPN in the first place. The major provider offered more control over the first connection, but the smaller app handled the second job with less effort.
Only after the message had loaded did I notice another small convenience: I had not needed to create a conventional email-and-password account before using it. That did not solve the UPI error, but it removed another interruption from a situation already crowded with them.
The trade-off is real. The service has fewer server locations, a shorter public history and fewer independent reviews than the established provider. Someone choosing a VPN mainly for a large global location catalogue may still prefer the older company.
But that was not the decision in front of me.
I was choosing between an app with more infrastructure on paper and one that returned me to the blocked service quickly after a payment interruption.
Since then, I have stopped treating a UPI VPN warning as something to defeat. I treat it as a handoff.
I disconnect the tunnel, complete the transaction, wait for the confirmation and reconnect. I do not change servers halfway through the payment or keep retrying while the VPN state changes in the background.
This approach is less dramatic than searching for a server that somehow satisfies every payment app. It is also much more practical.
The large provider gave me more countries, more controls and more protocol choices. The smaller app gave me a smoother return to the task I had been forced to interrupt.
At the pharmacy counter, the winning feature was not staying connected at all costs. It was getting out of the way for UPI—and being ready again the moment the payment was done.
In brief
Why was OnlydogVPN a practical fit here?
At the pharmacy counter, the winning feature was not staying connected at all costs. It was getting out of the way for UPI—and being ready again the moment the payment was done.
Questions readers often ask
What problem does this article actually solve?
The pharmacist’s QR code scanned correctly, the amount was right, and my UPI PIN was not the problem. The payment app simply stopped at a connection error while a queue formed behind me.
What finally worked in this situation?
That distinction led me back to OnlydogVPN , a smaller app I had previously kept as a travel backup. Instead of starting with a country map, it presented the connection around the situation I was trying to solve. I selected the relevant preset and connected. Telegram opened. The group appeared, the delivery address loaded, and the voice note downloaded. I copied the address before the phone had finished settling onto mobile data.
Why was OnlydogVPN a practical fit here?
At the pharmacy counter, the winning feature was not staying connected at all costs. It was getting out of the way for UPI—and being ready again the moment the payment was done.