#uristri2_api
1 messages · Page 1 of 1 (latest)
👋 Welcome to your new thread!
⏲️ We'll be here soon! Typically we respond in a few minutes, but sometimes we might take a bit longer if the server is busy or if you have a particularly tricky question.
⏱️ We close idle threads, which makes them read-only. Once a thread is closed it won't be reopened, but you can always start a new thread if you have another question.
🔗 This thread will always be available, even after it's closed. You can find it again using Discord's search, or you can save this link: https://discord.com/channels/841573134531821608/1448737025481113640
📝 Have more to share? Add more details, code, screenshots, videos, etc. below.
HI 👋
The biggest symptom when this happens is a burst of: Timed out waiting for connection token.
This implies that the tablets are not able to communicate with the server generating the connection token.
Backend answers with the connection token
we can even see the spike in connection tokens to our backend
Do you have an example serial numbers for the M2?
And, to be clear, the error you are seeing means the client application did not receive the ConnectionToken, or even more specifically, the onFailure callback function was executed: https://docs.stripe.com/terminal/payments/connect-reader?terminal-sdk-platform=android&reader-type=tap-to-pay#connect-reader
We can see the connection token in the telemetry logs
But the error is thrown, right? Unless your logs are incorrect somehow. So that tells us the Failure callback function was triggered.
we can even see the spike in call from the frontend and with the tokens in the logs.
Okay but, agian, that is not what I am talking about.
You can be successfully requesting connection tokens, that doesn't matter. And your server can be generating them, that still isn't the problem. The issue is that, for whatever reason, the receipt of the connection token by the client application appears to be failing. So that's where we want to focus.
I'm saying all of this just so we don't spend time looking at the server or the request from the client to get the token.
Right, but how is it possible that it works fine for 29 days of the month, and then during a specific time window in the month, the SDK suddenly stops reading the token?
and then during a specific time window in the month
Is this consistent? Is it always during the same time period monthly?
is not consistent and is not the same time period.
It happened to us twice this week, and before that on October 30th.
It’s not something that happens constantly, it’s not something fixed. Rather, on completely random days and within a short time window, we get this timeout.
hello! fyi i am taking over this thread, give me just a bit to catch up
no hurry, thanks
Tap to Pay gets disconnected for several shops located in different physical locations.
are you saying that this occurs for all of the shops at the exact same time?
I’m not sure if it’s the exact same millisecond, but it does happen within the same half hour.
You can even see the spike in connection-token requests in the backend logs yesterday and on Monday.
barrage of questions:
- which SDK are you using?
- is the disconnection happening between both the M2 reader and the TTP connection at the same time?
- is the same device being used to handle TTP as well as to communicate with the M2 reader?
which SDK are you using? 4.6.0
is the same device being used to handle TTP as well as to communicate with the M2 reader? nope, two different tablets
is the disconnection happening between both the M2 reader and the TTP connection at the same time?
Let me double check to be 100% sure
Hi! I'm working with uristri2. If a reader is already connected and something happen with internet connection, what could happen to the terminal?
Another question: what happened in this case is that there was an immediate reconnection of the SDK, first by SECURITY_REBOOT and then USB_DISCONNECTED (the latter many times in succession beyond what is seen in the image)
is the disconnection happening between both the M2 reader and the TTP connection at the same time?
It seems so. Based on the logs and what our support team told us, both TTP and the M2 stop working simultaneously.
Something to keep in mind: when this issue occurs, the support team performs a few reboots on the devices and they manage to restore everything. But still, the real problem is the loss of operational continuity.
ok, thanks for all the info so far - i'm gonna keep digging and will keep you updated if i have any more questions or find anything
also i've got a few other threads i'm juggling so that might delay my responses a bit
Perfect, I use Discord frequently so I’ll keep an eye on the thread for an update.
Thanks a lot for the help solanum!
If you need more info, just ping us.
regarding the SDK version, is this android or IOS?
100% android
one other question, have y'all looked into whether or not there were networking issues store side when these errors occurred?
also you're connecting via USB right? in your UsbConnectionConfiguration are you passing autoReconnectOnUnexpectedDisconnect?
yes, autoReconnectOnUnexpectedDisconnect = true
are you using the onDisconnect callback for your MobileReaderListener? if so i'm curious if you're seeing a DisconnectReason and if that tells us anything interesting
We have the speculation that everything starts with a hiccup in the connection on the tablets’ side.
However, we do see the tablets reconnecting to the sockets.
And it’s worth noting that these shops are located in different parts of the US with different ISP providers
Here there an example
ahh gotcha, i missed that
Hi 👋
I'm stepping back in as my colleague needs to go.
At this point, since the issue appears to be shared across multiple locations, it seems like the issue would have to be with creating the Connection Tokens.
Can you provide me with
- The account ID for the account making the server-side calls to create the Connection Tokens
- The time window for the last outage
- Details on how your server is hosted
Approximately every three weeks, we experience spikes in TapToPay disconnections and, at the same time, problems with the M2 devices. We have one device with the M2 connected (barista) and another with TapToPay connected (customer facing).
Sorry but approximately is not useful. We need exact timestamps
December 10, between 10 a.m. and 2 p.m., was one of these windows.
yesterday
same time on monday
Timezone?
Hi there, taking over for @drifting salmon as they had to step away. Can you provide the account ID for the account making the server-side calls to create the Connection Tokens?
acct_1DJxHZGnytxJlNo9
December 10, between 10 a.m. and 2 p.m.
in this window we had TapToPay disconnections in different stores and then when they try to use M2 the have problems to connect it (or reconnect them)
Got it, looking through the request logs now to find some specific info on failing requests
Thank you!
Hi there. I'm taking over for nobs here.
This is a pretty complex issue and I think at this point, it would be best to move to email and one of my colleagues can work with you asynchronously. I'll have our bot send you a dm that will explain how to do that