To ensure smooth operation of the CXC infrastructure we impose a set of restrictions on what kind of traffic is allowed on the peering fabric. This page gives a summary of those restrictions. CXC Team reserves the right to disable ports that violate the rules below.
Frames forwarded to CXC ports must have one of the following EtherTypes:
| EtherType | Protocol |
|---|---|
| 0x0800 | IPv4 |
| 0x0806 | ARP |
| 0x86dd | IPv6 |
Participants may utilize a single layer-2 MAC address to connect a single layer-3 router to each port or LAG group allocated to them by CXC. Changes to the MAC address of a connected port must be requested in advance to ops@cloudxchange.id. CXC applies MAC locking on participant ports; frames from other MAC addresses are dropped.
Use of proxy ARP on the router interface is not allowed.
Frames forwarded to CXC ports shall not be addressed to a multicast or broadcast MAC destination address except as follows:
- Broadcast ARP packets
- Multicast ICMPv6 Neighbor Discovery packets
Traffic related to link-local protocols will be filtered out on ingress, and will not be forwarded to other CXC ports. Link-local protocols include, but are not limited to, the following list:
- OSPF, ISIS, RIP, EIGRP
- Spanning Tree BPDUs
- ICMP redirects
- ICMPv6 Redirects
- ICMPv6 MLD
- CDP, LLDP, EDP
- L2 Keepalives
- PIM-SM/PIM-DM
- BOOTP/DHCP
- VTP/DTP
- UDLD
Using application layer protocols to unleash malicious actions against other CXC participants over CXC infrastructure, is forbidden. The following list includes, but is not limited to:
- BGP hijacking
- DNS amplification/flood
- HTTP flood
- NTP amplification
- UDP flood
- ICMP flood
