The biggest problems during migration were corporate issues, including dependencies on other devices such as F5, HSMs, and firewalls from Palo Alto. Everything was connected over the core switches, which made this a quite difficult part to manage and to onboard the teams and to understand the impact of the migration. We encountered an issue where we have two data centers, and unknowingly we were configuring the devices with the same VPC number on both devices. By doing that, we were not able to terminate the lease lines over those core switches. According to Cisco, we had to change the VPC number of one of the data center's core switches since they are in a VPC. We are still waiting, and it has been more than five to six years just to change the VPC number and migrate back the lease line. Now we are having cabin switches for the lease line and then the core switch. That VPC number issue was quite surprising for us. However, we have a very frequent issue with Cisco Nexus, particularly with the SFPs. Every time a port or the SFP heats up, it gets faulty. We have almost sixteen Nexus switches per data center. In a year, we experience one or two port issues per annum. We have standard Equinix data centers with standard installations, but somehow the temperature is always maintained and there is no ducting, yet the ports and the SFPs heat up or become faulty. This is a very frequent issue with the SFP-based port.
