STP timers ?

[philipbarker]

Hi,

I have changed my STP timers on the root bridge, either manually or using 'spanning-tree vlan xxx root primary diameter 2' commands.

From my non-root switch I can see that the timers have been adjusted under the 'ROOT ID' section of 'sh spanning-tree'. However, under the 'BRIDGE ID' section of 'sh spanning-tree' the timers are still set to 2, 15, 15 respectfully.

Can I conclude that in the case of STP convergence that the network will converge with the new timers that I configured for the whole layer 2 network ? and that the timers under 'BRIDGE ID' will be ignored ?

Regards,

Phil.

[Kevin Dorrell]

While your current root bridge is still the root, the timers will be as you configured them. If the root bridge disappears, then a new Spanning tree will be formed using the timers according to the new root bridge. It is always the current root bridge that sets the pace.

BTW, diameter 2 seems a bit small - is your network really that simple? If the worst case of link failure occured, (notwithstanding a change of root bridge), would the diameter still be only 2? If not, then a link failure might make the Spanning tree fail.

OTOH, if it is fast convergence you are after, you would be better off converting to rapid-pvst.

Kevin Dorrell

Luxembourg

[Jon Marshall]

Hi Phil

As far as i know only the timers from the root bridge are relevant so the timers reported under the Bridge ID do not effect the convergence time. The switch will use the timers from the root switch.

HTH

Jon

[philipbarker]

My set up is purely lab. I have 2 switches, one Gig and 1 fast eth connection.

I have done a bit more digging and 'debug spann events' shows learning to forwarding taking 7 seconds from both the root bridge and the non-root bridge.

Thanks for your responses.

Phil.