I'm looking for official documentation about this, and possibilities to fine-tune it.
- client is using a ***Vendor X*** firewall cluster/pair
- when switching the active-role both cluster-members seem to be active for a short time in parallel
- during this (short) time the switches try to "fix the network"
I remember the good old BCMSN class, somewhere in the small printed text it contained some info, this helped me to google this (the only documentation about MACFLAP on the web?):
https://www.cisco.com/c/en/us/support/docs/ip/hot-standby-router-protocol-hsrp/10583-62.html
Case Study #4: HSRP State Changes and Switch Reports SYS-4-P2_WARN: 1/Host <mac_address> Is Flapping Between Port <port_1> and Port <port_2> in Syslog
These error messages can appear:
2001 Jan 03 14:18:43 %SYS-4-P2_WARN: 1/Host 00:00:0c:14:9d:08 is flapping between port 2/4 and port 2/3
Feb 4 07:17:44 AST: %SW_MATM-4-MACFLAP_NOTIF: Host 0050.56a9.1f28 in vlan 1027 is flapping between port Te1/0/7 and port Te2/0/2
In Catalyst Switches, the switch reports a host MAC address that moves if the host MAC address moves twice within 15 seconds. A possible cause is an STP loop. The switch discards packets from this host for about 15 seconds in an effort to minimize the impact of an STP loop. If the MAC address move between two ports that is reported is the HSRP virtual MAC address, the problem is most likely an issue in which both HSRP routers go into the active state.
If the MAC address that is reported is not the HSRP virtual MAC address, the issue can indicate the loop, duplication, or reflection of packets in the network. These types of conditions can contribute to HSRP problems. The most common causes for the move of MAC addresses are spanning tree problems or physical layer problems.
Question
I wonder, if there is some official documentation ("Configuration Guide", "Command Reference"), i've been searching for a while.
Idea
... would be great, if this could be changed, maybe 4 moves within 15s...
MAC-Move-Notification can get configured very nicely, but the notifications are not the issue, the 15s blocking interval brings down the network...
...if nobody has an idea how to fix this, this posting might anyway be helpful to push the linked cisco-article back to front
Forum Tips: 1. Paste images inline - don't attach. 2. Always mark helpful and correct answers, it helps others find what they need.
