#bwurtz999-terminal-busy

1 messages ยท Page 1 of 1 (latest)

opaque scaffoldBOT
sweet turret
#

๐Ÿ‘‹ taking a look

trail trout
#

Thank you!

opaque scaffoldBOT
trail trout
#

Hey karbi, any update here?

#

@elfin island

elfin island
#

๐Ÿ‘‹ catching up, give me a minute

opaque scaffoldBOT
#

bwurtz999-terminal-busy

elfin island
#

I had seen some other recent examples where readers were getting this error because the device's network connection was spotty but that doesn't appear to be the case here. From what I can tell in our logs, it looks like the reader still believes it's busy with another payment despite the earlier one succeeding. Give me a bit longer to see what else I can find

trail trout
#

Ok thank you

#

If this helps, the next time a payment is attempted, the reader will just spin and spin endlessly

#

It has to be completely restarted to fix it

elfin island
#

Yeah I think that matches what I see in the logs - for the next couple minutes the reader believes it's still busy so the requests are all failing

trail trout
#

Ok... do you know which payment the reader believes it is still processing?

#

And why does this happen?

elfin island
#

I'm pretty sure it still believes req_0bdYMbtkYeFHN7 is still processing for some reason

#

I'm not sure about the why yet though

trail trout
#

For what it's worth, my manager seems to think it has something to do with 'tap to pay'. I'm not there but he says it only happens when someone taps their card - never on a physical chip insert

#

To be fair, I don't know if he only thinks that because most people tap instead of insert, so it's just more likely to happen after a tap

#

But for now he is telling everyone to insert their card instead of tap

#

Do you see anything in the lifecycle of pi_3OOkHEHw0TZh88Zm14EbxEa6 that indicates I'm doing something wrong in my integration?

elfin island
#

As far as I can tell it's not an integration issue - it's the reader getting into a bad state where everything is saying the payment is complete and the reader for some reason just wasn't updated to also reflect that

trail trout
#

Ok. Any ideas on how to stop that from happening? When it happens everything shuts down and the lines back up very quickly

elfin island
#

I'm checking with some other folks to see if they see anything that I'm missing

trail trout
#

Ok thank you

#

@elfin island I have to drop for now but I can return later. I know these chats usually get archived after a period of inactivity, but I really need to continue this discussion. I need to figure out why this is happening. Is there a way you can avoid archiving this thread?

#

I would like to join again later this afternoon to gather more information

elfin island
#

Better thing to do would be to follow up over email with long-running discussions like this

opaque scaffoldBOT
trail trout
#

Do I have to go through support? Or can I email you directly?

elfin island
#

There isn't a way to email us directly, but if you go through support I can find it any put all the context there - we really need to push on the Terminal team to take a closer look (we had tried in the past, but there were so many examples that cluttered everything that it was hard to narrow everything down). Now that you have a really clear, recent example that seemingly doesn't work we can push again

#

When you write in to support it would also help if you could give more context about why you're calling cancel_action after processing the Pamynet Intent

trail trout
#

I do that after a payment to clear the screen. I guess I only need to do that for a cash payment now that I think about it

#

But actually, I retrieve the reader before every cancel_action and I only call it if there is a pending action

elfin island
#

Oh interesting! That's a new development - is there any chance you log the pending action already?