Routing

Expand all | Collapse all

Receiving jnxl2aldInterfaceMacLimit SNMP traps from MX480

Jump to Best Answer
  • 1.  Receiving jnxl2aldInterfaceMacLimit SNMP traps from MX480

    Posted 04-11-2019 10:49

    After enabling SNMP traps on one of our production MX480 routers, I began receiving JUNIPER-L2ALD-MIB:jnxl2aldInterfaceMacLimit SNMP traps for a few AE interfaces. I'm having a hard time finding what this might be affecting. It looks like the MAC address limit for a specific interface is 1024, and this matches up with the results I get from show bridge mac-table brief | match interfacename, such as with the example trap below for ae2.452. Does anyone have any insight? Here's an example of one of the traps:

     

    sysUpTime=421 days 14 hours 26 minutes 23.19 seconds

    snmpTrapOID=JUNIPER-L2ALD-MIB:jnxl2aldInterfaceMacLimit

    jnxl2aldIntfRoutingInst.585=default-switch

    jnxl2aldIntfBridgeDomain.585=vlan-452+452

    jnxl2aldIntfLogicalRouter.585=

    ifDescr.99.585=ae2.452

    jnxl2aldIntfMacLimit.585=1024

    snmpTrapEnterprise=JUNIPER-CHASSIS-DEFINES-MIB:jnxProductNameMX480

     

    Thanks in advance for any guidance.



  • 2.  RE: Receiving jnxl2aldInterfaceMacLimit SNMP traps from MX480
    Best Answer

     
    Posted 04-11-2019 12:27

    Hello,

     

    This SNMP trap should've been generated when the no. of MAC addresses learned by interface reached the max no. of addresses that can be learned by that interface.  You can change the limit in bridge-domain configuration. Default limit is 1024 for access ports. Traffic for new MAC will still be forwarded if no packet drop action is specified.  More details available in the following link:

     

    https://www.juniper.net/documentation/en_US/junos/topics/task/configuration/layer-2-services-mac-bridge-domain-limiting-addresses-learned-from-a-logical-interface-in.html



  • 3.  RE: Receiving jnxl2aldInterfaceMacLimit SNMP traps from MX480

    Posted 04-11-2019 13:27

    Thanks for the explanation and the article. The interfaces I'm working with are logical interfaces on ae1 interfaces (for example, ae1.21, ae2.311, etc). According to the article, it looks like the "excess" addresses are deleted as the new MAC addresses are learned. I'm wondering if this means that the oldest addresses are deleted when a new MAC address is added. 

    The behavior is different for some configurations. For aggregated Ethernet interfaces and label-switched interfaces, the behavior is to learn all the new MAC addresses even when the limit has been reached. The excess addresses are later deleted. The learning limit does not apply to bridge domain trunk ports, because they have no counters for the individual domains, and those domains might have different MAC learning limits.

     

    If this is the case, could this result in unneeded broadcast traffic on the network as MAC addresses are added and removed from the MAC table? Like you explained, we're not seeing any problems with traffic being dropped on the network as there isn't a packet drop action specified. I know for sure that there are more than 1,024 devices on the interfaces that I'm working with and plan to increase the limit, but I'm very curious how this might be currently affecting the network and the mechanics behind it. 



  • 4.  RE: Receiving jnxl2aldInterfaceMacLimit SNMP traps from MX480

     
    Posted 04-11-2019 14:05

    Existing MAC addresses will not be removed. Here the excess MAC address means new MAC address. Broadcasts will be seen only for new mac addresses.