STP Lab

[lordnikec03]

 Hello everyone, this is my lab exercise where I have to configure the STP topologies to match the given diagram. Just wondering out of 12 topologies how many of them are possible to be done? I am only able to configure 3 of them.

[Joseph W. Doherty]

I just skimmed through them, but at a simple glance, it appears, to me, all are possible.

What's one of the topologies you're unable to accomplish?

[Joseph W. Doherty]

As an aside, assuming all switches are alike and all links are alike, root placement, doesn't appear optimal for any.

[lordnikec03]

I am only able to do 9, 10 and 12 using the cost changing method. I found out that other topologies have a similar pattern at a specific switch, making it impossible for me to configure. Assuming the red dots are the switches and green lines are the links, this is the pattern I am talking about: 

Anyway I just started learning this so I might be wrong, could you please enlighten me on how to configure them?

[Joseph W. Doherty]

I'm confused by your drawing.

Otherwise, keep in mind, when you set an interface cost, to change the path to the root, that device's interfaces' cost is compared to the sum of alternate paths to the root, not just interface cost to interface cost on that device.  I.e. the total other path cost has to be less then the revised original primary path to the root.

I've attached a PT file with switches in the topology of your diagram.

I've set the bottom left switch as the root.

All interfaces are using a default cost of 19.

I can "toggle" which of the top switch's is blocked, which by default is the right most link, by changing interface cost on the top switch.  But, what you need to understand is why/how setting different STP cost values, on those interfaces, do, or do not, flip the blocked port to the other port.

Once you understand that, you can make the STP topology be whatever you want.

Before changing anything, you need to understand why PT defaulted the STP topology as it did, again with the root switch being the bottom left.

If you're still lost/confused, we can use the attached PT to "see" what needs to be seen, and make adjustments.

[Martin L]

Yes, I also think all are possible!  What are your study learning sources?

since u just started learning STP (at least I think u said so), finish reading chapter and make sure u understand STP concepts (lots of videos of youtube)

you can use Packet tracer - free simulator for learning netowrking - to build STP topologies and keep changing Root switch Primary to match your topology (using CLI commands).  you could also use Root swith secondary command and watch changes;

Note: you should go to CLN - cisco learning network - to learn networking basics, fundamentals and solved various learning techs;

this site is more of real-world issues but not focusing on learning tech.   

Regards, ML
**Please Rate All Helpful Responses **

[lordnikec03]

Hi, thank you for those who answered me. But I am still quite confused with the pattern. Can anyone just show me an example configuration (let's say for topology 1) so that I will have a better understanding? Given the default costs are 4 for every interfaces.

[Joseph W. Doherty]

For a better explanation of what I touched upon in my prior reply see https://www.computernetworkingnotes.com/ccna-study-guide/stp-port-and-path-costs-explained.html.

At the moment, don't have access to PT (responding from my phone).  So, I'll just touch on the first couple of changes.

For topology 1, using gig links (cost of 4), we set sw57 as root.

Switches 55 and 56 will, by default, see a full path cost, to the root (via their direct connection), as 4.

Sw54, by default, will see full path cost, to root, via both switches 55 and 56 as 8, a tie!  It will, for a tie, use the tie breaking rule, but as want to path to 56, we need to make that FULL path, to root, more costly.  If we set interface STP cost to (minimally) 5, FULL path cost root, via 56, will then be 9.  (Don't pass go, don't collect $200, until you understand changing interface cost to adjust FULL path cost.)

What do you need to do next, and why?

[lordnikec03]

Thanks for the response and the notes, I sort of understand what you are trying to say. But wouldn't you have trouble configuring S2?

[Joseph W. Doherty]

"S2"?  Ah, on my phone I was seeing S# as 5#.  Laugh, well "S" does make more sense.

Anyway, "trouble", yes, if more effort/work is required.  Heck, having to manual adjust any STP costing is "trouble".  Error prone too, but certainly possible.

So, let's start with, what are the shortest paths from S2 to root, and their path costs?  (Don't forget what we just did to S4.)

[lordnikec03]

Shortest paths from S2 to root are via either S2 -> S4 -> S5 -> S7 (Cost = 12) , S2 -> S4 -> S6 -> S7 (Cost = 13) or S2 -> S3 -> S5 -> S7 (Cost = 12) ? 

[Joseph W. Doherty]

Perfect, I believe.

Two of those paths are better than the third, and are also a tie.

Which path of the two cost 12 do you want to insure will be used?  What change should be made on S2?

[Joseph W. Doherty]

Oh, forgot to ask - do you something like this setup in PT so that you can make each change and see the results?

[lordnikec03]

Topology 1 will be the one at the bottom left with a label "1" after scrolling down. Do not mind the mess haha.

Would you mind configuring topology 1 so that I can get a clear picture from the topology and complete the rest? That will be very much helpful to me. I have been sitting all day just to figure this out.

[Joseph W. Doherty]

Spent some time working on this with PT yesterday.  It appears PT visual display for what port is blocked isn't always correct (a known issue described in your lab notes although supposedly not seen in this particular lab).  Also, blocked port didn't always change as I believed it should.

As PT operations not always being correct isn't a surprise, I spun up a 7 node L2 switch config in CML, but 6 of the 7 switches' consoles remained black; possibly due to a lack of RAM on my PC.  I may try CML again after rebooting my PC to maximize available free RAM.