Managing multicast connections for containers in a computing network
Described herein are systems, methods, and software to manage multicast communications for containers in a computing network. In one example, a first node is configured to monitor for registration packets from pods on the first node to join a multicast group. The first node further identifies a registration packet from a first pod of the pods, wherein the registration packet indicates a multicast IP address, and configures one or more forwarding rules in a virtual switch to direct packets with the multicast IP address as a destination IP address to a virtual interface for the first pod. Once configured, the first node is further configured to receive a packet with the multicast IP address as the destination and direct the packet based on the one or more or more forwarding rules.
1 . A method comprising:
in a first node, monitoring for registration packets from pods on the first node to join a multicast group;
in the first node, identifying a registration packet from a first pod of the pods, wherein the registration packet indicates a multicast IP address;
in the first node and in response to the registration packet, configuring one or more forwarding rules in a virtual switch to direct packets with the multicast IP address as a destination IP address to a virtual interface for the first pod; and
in the first node, identifying a packet with the multicast IP address as a destination IP address;
in the first node, communicating a notification to an external switch indicating that packets with the multicast address should be forwarded to the first node; and
in the first node, directing the packet to the virtual interface for the first pod based on the one or more forwarding rules.
2 . The method of claim 1 , wherein the registration packets comprise Internet Group Management Protocol (IGMP) packets or Multicast Listener Discovery (MLD) packets.
3 . The method of claim 1 further comprising:
in the first node, identifying an expiration of a timeout period for the registration packet; and
in the first node, removing the one or more forwarding rules in response to identifying the expiration.
4 . The method of claim 1 further comprising:
in the first node, receiving a notification from the first pod indicating a request to remove the first pod from the multicast group; and
in the first node, removing the one or more forwarding rules in response to the notification.
5 . The method of claim 4 , wherein the notification comprises an Internet Group Management Protocol (IGMP) packet.
6 . The method of claim 1 further comprising:
in the first node, initiating a timeout timer in response to identifying the registration packet;
in the first node, identifying a second registration packet from the first pod, wherein the registration packet indicates the multicast IP address; and
in the first node, restarting the timeout timer in response to the second registration packet.
7 . The method of claim 1 further comprising:
in a second node, identifying the packet with the multicast IP address as a destination IP address from a second pod on the second node; and
in the second node, communicating the packet via a network interface to the second node.
8 . The method of claim 1 , wherein the node comprises a virtual machine or a physical computer.
9 . A computing apparatus comprising:
a storage system;
a processing system operatively coupled to the storage system; and
program instructions stored on the storage system that, when executed by the processing system, direct the processing system to:
in a first node, monitor for registration packets from pods on the first node to join a multicast group;
in the first node, identify a registration packet from a first pod of the pods, wherein the registration packet indicates a multicast IP address;
in the first node and in response to the registration packet, configure one or more forwarding rules in a virtual switch to direct packets with the multicast IP address as a destination IP address to a virtual interface for the first pod; and
in the first node, identify a packet with the multicast IP address as a destination IP address;
in the first node, communicate a notification to an external switch indicating that packets with the multicast address should be forwarded to the first node; and
in the first node, direct the packet to the virtual interface for the first pod based on the one or more forwarding rules.
10 . The computing apparatus of claim 9 , wherein the registration packets comprise Internet Group Management Protocol (IGMP) packets or Multicast Listener Discovery (MLD) packets.
11 . The computing apparatus of claim 9 , wherein the program instructions further direct the computing apparatus to:
in the first node, identify an expiration of a timeout period for the registration packet; and
in the first node, remove the one or more forwarding rules in response to identifying the expiration.
12 . The computing apparatus of claim 9 , wherein the program instructions further direct the computing apparatus to:
in the first node, receive a notification from the first pod indicating a request to remove the first pod from the multicast group; and
in the first node, removing the one or more forwarding rules in response to the notification.
13 . The computing apparatus of claim 12 , wherein the notification comprises an Internet Group Management Protocol (IGMP) packet.
14 . The computing apparatus of claim 9 , wherein the program instructions further direct the computing apparatus to:
in the first node, initiate a timeout timer in response to identifying the registration packet;
in the first node, identify a second registration packet from the first pod, wherein the registration packet indicates the multicast IP address; and
in the first node, restart the timeout timer in response to the second registration packet.
15 . A system comprising:
a first node computer; and
a second node computer communicatively coupled to the first node computer and configured to:
monitor for registration packets from pods on the first node to join a multicast group;
identify a registration packet from a first pod of the pods, wherein the registration packet indicates a multicast IP address;
in response to the registration packet, configure one or more forwarding rules in a virtual switch to direct packets with the multicast IP address as a destination IP address to a virtual interface for the first pod; and
receive a packet with the multicast IP address as a destination IP address;
communicate a notification to an external switch indicating that packets with the multicast address should be forwarded to the second node; and
direct the packet to the virtual interface for the first pod based on the one or more forwarding rules.
16 . The system of claim 15 , wherein the registration packets comprise Internet Group Management Protocol (IGMP) packets.
17 . The system of claim 15 , wherein the second node computer is further configured to:
identify an expiration of a timeout period for the registration packet; and
remove the one or more forwarding rules in response to identifying the expiration.
18 . The system of claim 15 , wherein the second node computer is further configured to:
receive a notification from the first pod indicating a request to remove the first pod from the multicast group; and
removing the one or more forwarding rules in response to the notification.
19 . The system of claim 15 , wherein the second node computer is further configured to:
initiate a timeout timer in response to identifying the registration packet;
identify a second registration packet from the first pod, wherein the registration packet indicates the multicast IP address; and
restart the timeout timer in response to the second registration packet.
20 . The system of claim 15 , wherein the first node computer is further configured to:
identify the packet with the multicast IP address as a destination address from a second pod; and
communicate the packet via a network interface of the first node computer.