Diameter on STP

[PETROBRASEISA]

The limit of 7 hops in a switched network with STP enabled is a physical limit or may I have more than 7 hops simply by adjusting the max age and forward delay parameters?

[Giuseppe Larosa · Accepted solution]

Hello Saint-Clair,

>> may I have more than 7 hops simply by adjusting the max age and forward delay parameters?

yes you need to adjust because current settings use an hello time of 2 seconds and 7 hops

so the learning and forwaring times are set to

7*2 +1

the max-age is set to 20

and so on

the spanning-tree root macro command accepts a diameter option that allows to specify the new diameter

Router(config-config)# spanning-tree mst instance_id root primary [diameter net_diameter [hello-time seconds]]

see

http://www.cisco.com/en/US/docs/switches/lan/catalyst6500/ios/12.2SX/configuration/guide/spantree.html#wp1102122

Hope to help

Giuseppe

[PETROBRASEISA]

Thank you so much!

[Jon Marshall]

Giuseppe

Not sure how this will help because with the diameter keyword the valid range is from 2 - 7 so if you used more than 7 hops then this command would not reflect your topology.

Note the link you provided was for mst not normal PVST+ or RPVST+ but the principle is the same -

http://www.cisco.com/en/US/docs/ios/lanswitch/command/reference/lsw_s2.html#wp1017396

Edit - do you know what happens if you just change the STP timers independently, presumably that would work ?

Jon

[Giuseppe Larosa]

Hello Jon,

thanks for your notes.

I took the wrong reference I usually point on that chapter because there is MST and Rapid PVST.

I think that you can change the timers indipendently until you keep the right relationships among them.

If the diameter option accepts only 2-7 it is useful only to reduce the STP size

Hope to help

Giuseppe

[Jon Marshall]

Hi Giuseppe

It's one of those things i have never needed to do ie. change the STP diameter to above 7.

I assumed, like you, that you could change the timers independently but i was just wondering what would happen if you did this but had already used the diameter keyword ?

Might have to boot up a couple of switches tomorrow :-)

Jon

[PETROBRASEISA]

Hi guys

I made this question because we have a switched network (made by a third-party group) with lots of redundant links. Keeping all the redundant links the network has 12 hops! Even discarding some redundant links the hops drop to 9.

[Jon Marshall]

Attached is a link to STP timers and tuning them - http://www.cisco.com/en/US/tech/tk389/tk621/technologies_tech_note09186a0080094954.shtml#stp_timers.

7 is not a physical limit as such but as dicussed you will need to modify these independently of the diameter option.

Jon

[PETROBRASEISA]

I've already read this note. After reading it I counted the hops on the network. Will the adjust on max age and forward delay be sufficient with 9 or 12 hops?

[Jon Marshall]

Based on those notes

Forward delay range = 4 to 30 seconds

Maximum age = 6 to 40 seconds

forward_delay = ((4 x hello) + (3 x dia)) / 2

max_age = (4 x hello) + (2 x dia) - 2

so for forward delay

4 x 2 = 8

3 x 12 = 36

8 + 36 = 44 / 2 = 22

max age

4 x 2 = 8

2 x 12 = 24

8 + 24 - 2 = 30

So forward delay = 22

maxage = 30

which are both within the ranges you can use.

** Warning - maths has never been my strong suit but i have checked it a couple of times :-) **

Jon

[PETROBRASEISA]

Ok Jon.

It matched with the maths that I made here too.

[Jon Marshall]

Well either we are both bad at maths or you should be okay :-)

[sarahr202]

Hi jon!

Just got one question.

i was looking at your math where you assume hello time=2 seconds.

I find in my cisco book, that hello-time=2 is assumed for a switched diameter=7.

Here the diameter is 12, so hello -time should be more than 2 sec.

i would really appreciate any correction.

Thanks a lot!

[Jon Marshall]

Sarah

Hello = 2 is the same whether the diameter is 1, 7 or any number in between. The hello timer does not scale with the diameter (or at least not as far as i know)

Does your book suggest how the hello timer is worked out from the diameter because i don't believe it is. The maxage and the fwd delay are calculated from the diameter and a default hello time.

Jon

[sarahr202]

Thanks Jon!

Here is excerpt from the book, page 199 , at the bottom of the page, inside the box marked as tip.The book is CCNP BCMSN official exam certification guide ,fourth edition,isbn# 1587201712

" The hello time is based on the time it takes for a bpdu to travel from the root bridge to a point seven switches away. This computation uses a hello time of 2 seconds"

Another one from the same book, page no 231-232

" As an example , suppose that a small network consists of three switches connected in traingule fashion.The longest path that a packet can take through the sample network is three switches.This is considerably less than the reference diameter of seven that is used to calculate the default timer values.Therefore you can safely assume that this network diameter is three,provided that no additional switches will be added to lengthen the longest path.suppose that a heloo time of 1 sec is also desired,to shorten the time neeeded to detect a dead neighbor .The fllowing command attempts to make the local switch become root bridge and automatically adjusts the stp timers.

switch(config) spanning-tree vlan 100 root primary diameter 3 hello-time 1"

Obviously the author associates the hellotime of 2 seconds with diameter of seven switches.

I think Jon ,you should write a book, it would be a best seller and i would be the first buyer!

thanks a lot!