How to troubleshoot network issues when sending API requests

Objective

How to troubleshoot network issues when sending API requests

Applies To

• Voice API
• SMS API
• API calls

Procedure

Customer Network → Vonage

Customers might experience a connection timeout when making an API call to our API endpoints. These timeouts could be caused by network issues between the customer network and Vonage. 

In order to identify the root cause of the issue -

• Try sending a request using command-line curl as below :
  

• curl -svvv -w "\ntime_namelookup: %{time_namelookup}\ntime_connect: %{time_connect}\ntime_appconnect: %{time_appconnect}\ntime_pretransfer: %{time_pretransfer}\ntime_redirect: %{time_redirect}\ntime_starttransfer: %{time_starttransfer}\ntime_total: %{time_total}\n\n" -o /dev/null $API_ENDPOINT

• Record the source and destination IPs and verify your proxy server and firewall configurations.
• If the above shows no issue, then run a packet capture on your WAN interface when you are experiencing timeouts. This will show if TCP connections from your Firewall/Server are arriving at Vonage.  

  tcpdump host rest.nexmo.com -ttt -vvv -s0 -i any -w $(hostname)-$(date +%Y%m%d%H%M%S).pcap

Vonage → Customer Network 

When sending callbacks to customer's webhook endpoints from our servers, there might be connection issues. Most of the cases are due to the firewall configuration on the customer's network. All traffic coming from the Vonage network might have been blocked, or some more advanced filtering could discard packets coming from specific servers, or they could have some throttling and block packages. 

In order to identify the root cause of the issue -

• Packet capture on your WAN interface: This will show if TCP connections from Vonage are arriving at your Firewall/Server.

  tcpdump host 174.36.197.202 -ttt -vvv -s0 -i any -w $(hostname)-$(date +%Y%m%d%H%M%S).pcap