02-11-2011 11:10 AM
I am in the process of deploying FCX 648 switches, and have recently encountered a spanning-tree reconvergence when I enable uplinks to MLX. I have 'spanning-tree root-protect' on MLX side, but when uplinks are enabled I still have a reconvergence and recalc.
Has anyone else experienced this, and if so what solution was implemented.
02-13-2011 11:28 AM
This is from the MLX configuration guide:
Root Guard may prevent network connectivity if improperly configured. It needs to be configured on
the perimeter of the network rather than the core. Also, Root Guard should be configured only on the
primary port of a LAG.
It sounds like you have Root Guard configured on your core switch (MLX).
02-14-2011 11:30 AM
What spanning tree are you using (STP,RSTP, MSTP etc)?
Note root guard will not do anything when MSTP is used.
Which switch do you want as the root? Have you set that switch with a lower prioity?
Can you please provide a network layout?
02-21-2011 10:59 AM
Thank you very much for your assistance, After further investigation, I noticed that links to existing Cisco 6500's are becoming root when I turn up new FCX's connected to our MLX's via root protected ports. For some reason turning up a new FCX causes a spanning tree recalc that leaves the Cisco as root bridge and puts network in chaotic state until FCX ports are disabled and MLX's reassume root. Could this be a result of not using VLAN 1 on Brocades, which is Cisco default VLAN?
02-21-2011 02:25 PM
What type of STP are you using on the FCX and MLX and 6500's?
Do you want the MLX to ALLWAYS be the prefered Root bridge? If so they lower its prioty to a lower value then the 6500's.
02-22-2011 09:23 AM
Unfortunately my name would dissapear from my paycheck if I post proprietary info of that nature, but I can say that the LAG's to Cisco's do not have root protect on primary lag member, could this be the reason that the Cisco's are becoming root. Or could it be the path costs perhaps?
02-22-2011 02:43 PM
Configure on your MLX a priority value of 0 (this is the highest priority). You need to do this for each VLAN that are running RSTP.
You can also use root guard on the LAG's to the CISCO however using the priority should stop that that too.
To designate a priority for a bridge, enter a command such as the following at the VLAN level.
NetIron(config)# vlan 20
NetIron(config-vlan-20)# rstp priority 0