Over Concurrency Limit.
# support
Hello, this is my first time using it. I created a new assistant, I changed Transcriber language to el and then tried writing Greek as first message (el is Greek language). Then, when I pressed "talk with AVA" I got no voice and every time since then, I get the error: "Error Over Concurrency Limit."
Contact us in live chat! or Telegram (@rforrank)
Hey Tom, what Voice were you using?
@jayvapi.ai I have similar situation with calls done via Web they do not launch at all and will count as parallel session against concurency limit for some time. Using ElevenLabs custom voice. See call id: 56e89d05-d99b-4e16-957f-10d2c68b30f0 , 8d8ecd71-949f-485c-8be3-8a556edcda60, 49bede88-35ca-4daa-b44b-ea23c961c2a4 In the Calls tab I have them without any details ...
Thanks for callID's, following up on this
Sorry for dropping the all on this. We had a bug with concurrency counting. Let us know if you still experience it
I'm getting this error now trying to send post request from postman { "message": "Over Concurrency Limit.", "error": "Bad Request", "statusCode": 400 }
{ "assistant": { "backgroundSound": "office", "clientMessages": [ "transcript", "hang", "function-call", "speech-update", "metadata", "conversation-update" ], "dialKeypadFunctionEnabled": true, "endCallFunctionEnabled": true, "endCallMessage": "this Is the end of the call message", "endCallPhrases": [ "thanks, have a great day!" ], "firstMessage": "Yo, this Is me, calling me !", "hipaaEnabled": false, "llmRequestDelaySeconds": 0.1, "maxDurationSeconds": 1800, "metadata": {}, "model": { "fallbackModels": [ "gpt-4-0125-preview", "gpt-4-0613" ], "maxTokens": 525, "messages": [ { "content": "You are laura, a friendly assistant. You gather Information about callers to qualify them for an appointment. If they qualify, collect appointment booking details.", "role": "assistant" } ], "model": "gpt-3.5-turbo-1106", "provider": "openai", "semanticCachingEnabled": true, "temperature": 1 },
"name": "laura", "numWordsToInterruptAssistant": 2, "recordingEnabled": true, "responseDelaySeconds": 0.4, "serverMessages": [ "end-of-call-report", "status-update", "hang", "function-call" ], "silenceTimeoutSeconds": 30, "transcriber": { "keywords": [ "keyword" ], "language": "cs", "model": "nova-2", "provider": "deepgram" }, "voice": { "provider": "azure", "speed": 1.25, "voiceId": "andrew" }, "voicemailDetectionEnabled": true, "voicemailMessage": "voicemail_message" }, "customer": { "name": "Trejon", "number": "censored" }, "maxDurationSeconds": 1800, "metadata": {}, "phoneCallProviderBypassEnabled": true, "phoneNumber": { "serverUrl": "censored", "twilioAccountSid": "censored", "twilioAuthToken": "censored", "twilioPhoneNumber": "censored" }, "type": "outboundPhoneCall" }
Have this almost every day...(
Please look into this issue @User
Hey sorry about that. We definitely know there's a bug where calls don't end properly and stay in "active" state. For now, just email us at support@vapi.ai. It's easy to reset manually + bump up your limit. We'll have a permanent fix soon, just need some time to investigate why it's happening for some and not others.