This document describes how to troubleshoot User Datagram Protocol (UDP) flooding on the Cisco Digital Content Manager (DCM).
There are no specific requirements for this document.
The information in this document is based on the Cisco DCM D9902.
The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, make sure that you understand the potential impact of any command.
Scenarios have been observed where the D9902 is connected to a Layer 2 (L2) switch and configured to receive unicast video streams. However, after the unicast video is streamed for approximately five minutes, the same LAN observes flooding on the switch, which causes an outage in the customer network. In this scenario, it was determined that the switch port connected to the DCM aged out the Media Access Control (MAC) address table, which caused the flooding because the destination MAC address was unknown to the switch.
The UDP flooding is a common issue in unidirectional scenarios. The Address Resolution Protocol (ARP) cache timer (with a default of four hours) in router/Layer 3 (L3) switches is always higher than the MAC address age timeout (with a default of five minutes). This means that there is always a chance that the MAC address information can be removed from switch if there is no response from the destination device.
Here are three different methods that you can implement in order to resolve this issue:
- The most reliable and simple solution for this issue is to create a dummy multicast join in the DCM. In this case, the DCM sends an Internet Group Management Protocol (IGMP) membership report to the switch, and the switch begins to poll the DCM periodically. In order to poll the DCM, the switch sends an IGMP membership query, which refreshes the MAC address table in the switch.
- Another solution for this issue is to decrease the value of the ARP cache timer on the switch so that it is less than or close to the MAC table aging timer. This causes the ARP packets to become broadcast, and the MAC address must be relearned before the L2 entry ages out.
- As a third option, you can configure a static MAC address entry on the switch, which remains even after a reboot and eliminates the timeout issue.