Link to home
Start Free TrialLog in
Avatar of mikey250
mikey250

asked on

SWITCH STP QUERY

There are 2 parts to my questions just for my own clarity:

Election criteria

When switches decide who is the 'root bridge' or not it goes through the following checks & ive added 2 questions (?) along the process to gain clarity in Election criteria:

- Lowest root bridge id
  - 1st - which is a combination of the bridge priority & mac
  -  2nd - default priority is 32768

- Lowest root path cost
  - ie - 100 = 10mbps, 19 = 100mbps, 2 = 10 Gbps or 4 = 1 Gbps - so if the initially chosen 'root bridge' had the lowest bridge priority  & mac or even default priority of 32768 & a cost of say 19.

Does this mean that if a 2nd switch(s) for example has a 'cost 2' factor that this would be chosen as the 'root bridge'? unless obviously manually config'd!

continuing on below:

- Lowest Sender bridge id
  -  combination of bridge priority & mac - so if another switch as per above was detected to have a 'cost 2' factor also, it would then go to:

-  Lowest Sender port id
  -  which consists of a port priority and port id (based on port number) ie port 2 would be the choice and port 10 for eg would be lower and not chosen?
ASKER CERTIFIED SOLUTION
Avatar of Don Johnston
Don Johnston
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of mikey250
mikey250

ASKER

so what is the point of the 'Election criteria' if that is that? this is the part im trying to understand, as something must be different for this election criteria to take place?

Im assuming that if the required switch that needs to be the 'root bridge' happens to have a higher bridge id than the one currently in place, then presumably the 'Election criteria' would then take place as presumably there would be a manually configured 'bridge priority' added to start the Election process!!
so what is the point of the 'Election criteria' if that is that?

Root is responsible for creating the BPDU's that are used by all other switches to discover loops and block ports to prevent the loops from causing all of the problems that result from the loops.

SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
I don't know what you're reading, but there is only one factor used to elect the root bridge... The Bridge ID.

Think about it: Since it's impossible to have two bridges with the same Bridge ID, one will have to be lower. So why would we ever need to look beyond that one factor?

Are you sure that you're not talking about selecting the "root port"?
that must be where im getting my 'wires' crossed then!!  can you explain this election process clearer than i can so i can grasp? appreciated!!
>can you explain this election process clearer than i can so i can grasp?

I really can't think of any other way to put it. The election of the root bridge is pretty basic. Lowest bridge ID wins.
i get it now as ive spent all day yesturday reading other stuff!!!good stuff yes straight forward now!!
I got what the 'root bridge' was about but it was the election criteria i was getting confused with, thinking once i'd found the lowest bridge id and as you said is the chosen one.  but for some reason i thought as the switches all go through the election process, if it turned out that someones cost was higher then this would be chosen, INSTEAD of the initial root bridge.

Like when routing protocols such as rip, igrp, eigrp, ospf are all being used to maybe get to a destn, but due to maybe rip having a longer subnet mask, it maybe chosen over all other better admin distances of the others!!
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Cost, Bridge ID and Port ID are used (in that order) to select the root port on all non-root bridges.

i might have been hasty with my last comment, or is is as you say above, cost first for non-root bridges?
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
hi, just for clarification i know i entitled this thread iniatially Switch STP Query, but that was just to get interest in my question as sometimes i get no replies from experts at all, but yes 'root port' selection, in my book it titles first:

stp port roles: just as an eg.

- root
- designated
- non-designated
- disabled

then it goes onto stp reviews below: called the election criteria, so although you say cost 1st, in my mind your right although the book starts with Lowest root bridge id first ie before you actually know of who is the 'root bridge'

1.  Lowest root bridge id
    - 1st - which is a combination of the bridge priority & mac
    -  2nd - default priority is 32768

2.  Lowest root path cost

3.  Lowest Sender bridge id
     -  combination of bridge priority & mac

4.  Lowest Sender port id
     -  which consists of a port priority and port id (based on port number)
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
well ive just come from my ccnp course and this is how they teach as the instructor updates his ccnp every 6 months also!  but i do realise that the industry may disagree with cisco or vice versa, but i suppose for the purposes of passing the exam, they said stick to what they teach.
when you say one 'root bridge' i agree, but i suppose if there are 2 ports coming from the 'root bridge' for i suppose , backup purposes also, but both ports with 2 different costs, or the same, with i suppose human intervention dependant on 'network admins' decisions, then the election criteria would play a part!
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
ok I can accept.  just one more comment because you would know better than me!!!  ive just looked through my book and i rememer only coming across the word "inferior" once and they've got that word categorised under:

'backbone fast' - which detect indirect link failure

- initiated when a root or blocked port on a switch receives inferior bpdus from the designated bridge
- indirect knowledge of link failure
- switch sends root link queries (RLQ) to veify root bridges existence
- bypasses 20 second max-age timer (takes 30 secs to forward)
- enable on all switches

although just as an extra mention, they mentioned about 'uplinkfast' first', not that it holds relevance to 'backbone fast'

although prior to this it mentioned about 'uplinkfast'  having direct link failure and can move from one unworkable port to a second port to keep connection open and  in 3-5 seconds, which would increase the bridge priority to 49152 and adds 3000 to the spanning tree port cost of all interfaces to reduce the chance it will become the root bridge.
so im just thinking if 'backbone' fast wasn't even configured does your comments still apply!  although i realise your saying their wrong!! fair enough i can make a not in my book!!
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
ok, not going to disagree about 'real world' knowing.  thanks!!
ive ticked some of my boxes as there was a question about the 'election criteria' being: cost 1st and 2nd bridge id, but during my ccnp course it was taught as: bridge id 1st and cost 2nd and so on.  but donjohnston has just raised the point that it is in actual fact in the real world: cost 1st!!  so just acknowledging the point!!