Legacy Dev Forum Posts

 View Only

Sign Up

  • 1.  Random puzzle of the day

    Posted 06-05-2025 18:30

    LighthouseMike | 2024-03-04 21:13:08 UTC | #1

    Good afternoon/evening, :smiley:

    So here's my situation (which is admittedly a weirdo and probably highly unlikely to ever happen in production - but still something I'd like to guess out if possible):

    1. I call into my app, press the Answer button and I'm on the call. Great, everything normal.
    2. I press my "cold transfer" button and transfer to the testing queue (same queue I'm on).
    3. I wrap-up the call (again,s till using the software I built using the SDKs) and again, everything seems to function as expected.
    4. Here's where we get into silly territory - but it's still something I'd like to know how to handle, just because it could realistically happen. I'm the only person in my queue. :laughing: So I've essentially transferred... to myself.

    When I do this, steps 1-3 repeat, and everything seems to work normally. I get pendingSession, accept, I get sessionStarted, and I'm in my call controls. I can put the call on hold, mute, etc. But when I go to transfer again, I get the error, "You can only transfer conversations that are in the connected state.". Ohhhh........kay? Sooooooo..... I'm clearly still connected......... and...... uhhhhhhh........????????????? :laughing: So I can mute and hold a call I'm not connected to. Must be a full moon outside? Was it a full-moon leap-year? :laughing:

    To add another layer of "interesting", when I end the call, I get two "sessionEnded" events (I hear my program say "call disconnected" twice). Struggleshooting this one his one isn't crazy-important... it's just another level of crazy. :crazy_face:

    Any ideas on where that missing puzzle piece might be? :laughing: Maybe the dog ate it (my sister's puppy will eat anything :laughing:).


    tim.smith | 2024-03-04 21:18:46 UTC | #2

    LighthouseMike, post:1, topic:24994
    But when I go to transfer again, I get the error, "You can only transfer conversations that are in the connected state."

    This sounds like a bug, assuming you've verified that the conversation is indeed in a connected state. This should be reported to Care for further investigation since it appears to be a server-side issue. They'll need to know the conversation ID, details of the request (endpoint and body), and the correlation ID from the response.


    LighthouseMike | 2024-03-05 16:47:06 UTC | #3

    tim.smith, post:2, topic:24994
    verified that the conversation is indeed in a connected state. This should be reported to Care for further investigation since it appears to be a server-side issue. They'll need to know the conversation ID, details of the request (endpoint and body), and the correlation ID from the response.

    Interesting... so... where do I start? What I mean is:

    • How do I "verify that the conversation is indeed in a connected state"? I'm not even entirely sure what that means. The call is connected in the sense that I can hear my background music and keys typing, put the call on mute and hold and all that... what exactly do I need to do to accomplish this verification?
    • What else can I do to rule out bugs on my end? Like could there be something else affecting this, org config for example? This morning I noticed I get a different result when I log into our app using the production org. Instead of the error about the call not being connected, I get no call controls (presumably because the call is already wrapped up? still debugging) and no error. So... I'm guessing there might be some setting between our dev org and our prod org that isn't how it ought to be... or something... naturally I'm clueless as usual. :laughing: But before I go and file a ticket or whatever I want to make sure there's nothing on my end that might provide an explanation. The fact that I'm consistently getting different behaviors from different orgs tells me it might be settings-related (Genesys is an extremely config-heavy system after all) and not a bug. If I can tweak memory address 9957 to 0xFB120096 :laughing: and it starts working on me, then it's probably a config-rig problem.
    • If I make those adjustments and it seems like this is in fact something that should go to Genesys support... how do I get a "correlation ID"? Conversation IDs I know; I use those constantly and know how to get them. But correlation ID is a new one on me.

    Thank you btw :smiley:


    tim.smith | 2024-03-05 17:04:25 UTC | #4

    LighthouseMike, post:3, topic:24994
    How do I "verify that the conversation is indeed in a connected state"?

    Look in the normal Genesys UI to confirm you're connected.

    LighthouseMike, post:3, topic:24994
    What else can I do to rule out bugs on my end?

    Make sure your problem report is repeatable. If you can repeat the same set of instructions, confirm that the Genesys UI shows the call is connected, and you get that error every time, that demonstrates a pattern that Care can troubleshoot. If this has happened a few times and you can't reproduce it, there's not much Care can do and they'll tell you it's a bug in your code.

    LighthouseMike, post:3, topic:24994
    If I make those adjustments and it seems like this is in fact something that should go to Genesys support... how do I get a "correlation ID"?

    https://help.mypurecloud.com/articles/find-a-correlation-id/


    LighthouseMike | 2024-03-05 17:13:06 UTC | #5

    Okay, well it's definitely repeatable, and consistently. So I'll do what you're suggesting and keep you guys posted (whether that be here or with support). Thanks as always for your help, Tim. :smiley:


    system | 2024-04-04 17:13:34 UTC | #6

    This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.


    This post was migrated from the old Developer Forum.

    ref: 24994