Hi all, Some of our production outbound calls are failing intermittently due to an ended reason of "Twilio Connection Failed". It is interesting because most of the calls with the same number + assistant succeed. Here is the behavior we observe: - The call log shows an ended reason of "twilio connection failed' - The call log in Vapi Call Log table shows no logs, overview, cost, call times, or duration - The numbers were not disconnected -- we tested calls to verify the numbers were online - Phone number is managed by Vapi - Server messages are not sent for the call - Most outbound calls are succeeding with the same number, assistant, and in the same prod environment Call Record Examples: - 6a3a6f5e-4c45-4b22-a599-760c9f099c5f - ee891a8e-9bdb-40dd-8660-bd0f326fd6cc - 50c8ae3b-a72f-459f-aa33-ac0cf739bff5 - 294b02c2-e024-443c-bec4-3bf9128163f9 Any guidance on how to manage + debug these errors would be much appreciated!! Please let me know if any additional dets would help out. Thanks 🙏

message has been deleted

Hi @Vapi , I just checked through that thread and it was not relavant to this issue

@Jeff Hello! I'm sorry to hear about the intermittent issues you're experiencing with outbound calls. It's good that you've already verified that the numbers are online and that most calls are succeeding. Since the problem is intermittent and seems to be related to the Twilio connection, here are a few steps we can take to manage and debug these errors: 1. **Check Twilio Status**: Verify if there were any reported outages or issues with Twilio services at the times the failed calls occurred. You can check Twilio's status page for any incident reports. 2. **Review Application Logs**: Look into your application logs around the time of the failed calls to see if there are any errors or warnings that could provide more insight. 3. **Examine Network Issues**: Since the issue is with the connection, it could be related to network problems. Check for any network-related issues that could be causing intermittent failures. 4. **Twilio Debugging Tools**: Utilize Twilio's debugging tools and logs to get more detailed information about the failed calls. This may help identify if the issue is on the Twilio side. 5. **Contact Twilio Support**: If the above steps do not yield any results, reaching out to Twilio support with the call record examples you provided could help uncover more details. 6. **Monitor Call Patterns**: Look for patterns in the failed calls. Are they happening at specific times of the day, with specific destinations, or after certain application events? 7. **Test with Different Numbers**: If possible, try using different phone numbers to see if the issue persists. This could help determine if it's an issue with specific numbers. 8. **Update SDKs and Libraries**: Ensure that you are using the latest versions of any relevant SDKs or libraries, as updates often include fixes for known issues. If you need further assistance or if there's anything specific you would like me to do to help you debug this issue, please let me know!

Hi @Vapi , Here are the results of your suggestions: 1. I checked the Twilio status at the times of the events and did not see an outage: https://status.twilio.com/history 2. There are no logs displayed in the Vapi call record. I don't have additional application-level logs as it appeared that no system messages reached our prod server like they usually do 3. I don't have visibility into this as it appears to be an issue between Vapi and Twilio (these Twilio numbers were purchased through Vapi and are managed by Vapi) 4. I don't have access to the Twilio logs as the number is managed by Vapi 5. I don't manage the relationship with Twilio as the number is managed by Vapi 6. I did look for patterns, such as whether a bunch of calls occurred in quick succesion (i.e., it could be related to rate limiting) and did not notice any patterns 7. I have only observed this issue with one of our numbers so far. That number uuid is 888dba02-2c4e-4623-9b5d-ad662820c74d 8. I am using the latest API spec to trigger the call to Vapi. I could definitely use some further assistance

@Jeff Thank you for providing a detailed update on the steps you've taken. It seems like you've done a thorough job investigating the issue from your end. Given that the problem appears to be between VAPI and Twilio, and you don't have access to certain logs or the ability to manage the Twilio relationship directly, the next steps would involve deeper investigation from the VAPI side. Here's what we can do to assist you further: 1. **Internal Logs**: We will need to check VAPI's internal logs for the call records you've provided. This can help us understand if there were any errors or anomalies during the call setup process. 2. **System Messages**: We should investigate why system messages did not reach your production server as usual. This could be indicative of a deeper issue. 3. **Network and Service Configuration**: We'll review the network configuration and service setup between VAPI and Twilio to ensure there are no misconfigurations or issues that could lead to intermittent failures. 4. **Dedicated Support**: Since this issue is affecting production and you've exhausted the troubleshooting steps available to you, it would be best to escalate this to a dedicated support team within VAPI who can access the necessary logs and configurations. 5. **Escalation**: I will initiate an escalation process to ensure that your issue is addressed by the appropriate team with the right level of access and expertise. Please provide any additional details that you think might be relevant for the investigation. Meanwhile, I will proceed with the necessary internal escalations to ensure that this issue is looked into with the urgency it requires. If you have any other concerns or need further assistance, feel free to let me know.

Thanks @Vapi I will monitor this thread in case the team needs additional info

@Jeff Sweet. Have a great day!