STP Problem

I have an idea of whats going on but I want a second opinion. Can anyone tell by looking at the picture if there is a loop? (Side Note: Only this switch produces about 150 packets of STP traffic per second)

c:%5CDocuments%20and%20Settings%5Cdtommasino%5CDesktop%5Cstp.jpg

c:%5CDocuments%20and%20Settings%5Cdtommasino%5CDesktop%5Cstp.jpg

Find more posts tagged with

Save $250 on 2025 certification boot camps from Infosec!

Book now with code EOY2025

Button

Comments

NightShade03

show spanning-tree summary (from that switch)

Xenz

NightShade03 wrote: »

show spanning-tree summary (from that switch)

Nice... Do you happen to know if the so called engineer even has pruning enabled?

NightShade03

Yes pruning is enabled, BUT only because the person before him set it up that way, he didn't know until I pointed it out to him....

NightShade03

kryolla wrote: »

Im pretty sure you know this but just in case VTP pruning has nothing to do with spanning tree

Yup! Just answering his question.

Cyanic

Xenz wrote: »

If using SPAN to monitor traffic, the source MAC would be changed to the port on which the PC is connected that you're monitoring from or does SPAN not change anything? I've never used this feature hence why I'm asking. So really the only thing out of the ordinary in the screenshot is the TCN notifications?

SPAN ports change nothing except they will send the packets untagged.

Catalyst Switched Port Analyzer (SPAN) Configuration Example - Cisco Systems

Edit: Also, SPAN traffic will not included error/dropped packets, like packets with bad CRCs.

Xenz

I was hoping you'd tell me no!

My question still stands from above to anyone able to answer, is there any recommended best practice or is it up to the engineer to pick an arbitrary number regarding whether to implement MST?

Xenz

Cyanic wrote: »

SPAN ports change nothing except they will send the packets untagged.

Catalyst Switched Port Analyzer (SPAN) Configuration Example - Cisco Systems

Edit: Also, SPAN traffic will not included error/dropped packets, like packets with bad CRCs.

Thanks, I was looking for a page like that, instead I got a bunch of pages on configuring it on certain switches instead of an overview of what it does. Cisco's maze wins again. I'll have to lab this up later to practice with it. Thanks Cyanic

Forsaken_GA

Xenz wrote: »

I was hoping you'd tell me no!

My question still stands from above to anyone able to answer, is there any recommended best practice or is it up to the engineer to pick an arbitrary number regarding whether to implement MST?

We used to use MST, but had to move to RPVST+ when we added Force10 E1200's to our network as distribution switches. We found out, after 16 hours of downtime, that Cisco's implementation of MST is not quite the same as the standard and was incompatible. We would have preferred to use it, as all of our VLAN's have the same spanning tree anyway.

Xenz

Forsaken_GA wrote: »

We used to use MST, but had to move to RPVST+ when we added Force10 E1200's to our network as distribution switches. We found out, after 16 hours of downtime, that Cisco's implementation of MST is not quite the same as the standard and was incompatible. We would have preferred to use it, as all of our VLAN's have the same spanning tree anyway.

I see, how many VLANs did you support if you can divulge that information? The only thing I found on Cisco docs about whether to use MST or not was in the hypothetical situation of 1000 VLANs. The BCMSN exam library book was equally worthless so just curious. The book made it seem so much better in performance, but the details regarding it was about 1 page long and 3/4's of that was IOS output.

kryolla

you might have better luck on CCDP books for the why on things i.e choosing OSPF instead of EIGRP. CCNP is basically a how to

Xenz

Good point. Gogo safari books online.

*note* Thanks for the help Kryolla/Forsaken

Forsaken_GA

Xenz wrote: »

I see, how many VLANs did you support if you can divulge that information? The only thing I found on Cisco docs about whether to use MST or not was in the hypothetical situation of 1000 VLANs. The BCMSN exam library book was equally worthless so just curious. The book made it seem so much better in performance, but the details regarding it was about 1 page long and 3/4's of that was IOS output.

I can say that it was well short of 1000 vlan's. For us, our layout does not require that different vlan's have a different root (we manually prune vlan's, so that the only vlan traffic allowed on a trunk is for vlan's that a switch has setup on it), so it was just a little bit easier to maintain operationally. For our colo customers, they're not part of our Spanning Tree anyway, we just give them an ethernet drop and route their IP space to them over a /30, and what they do inside their network is their business.

cisco_trooper

NightShade03 wrote: »

The attached is the setup that we have running. The fiber line leads out of the building to a core L3 switch, switch 2 - 7 are config as a stack. I don't see redundancy here as this is a pyramid design just laid out differently. If there is no redundancy.....whats the point of STP?

Yikes, do any of these switches actually have dual uplinks? I don't see a bridging loop here...Although that is one hell of a physical architecture.

cisco_trooper

Xenz wrote: »

I was hoping you'd tell me no!

My question still stands from above to anyone able to answer, is there any recommended best practice or is it up to the engineer to pick an arbitrary number regarding whether to implement MST?

When choosing whether or not to utilize MST you are basically looking at the physical trees that can be built within your environment. If you only have two possible Spanning Tree outcomes when a link goes down, and you have 1000 VLANs, it makes quite a bit of sense to use MST. This of course depends on having a competent engineer on hand because it will require some planning to avoid downtime if you need to add VLANs to the MST. This is due in large part to the way MST works at a low level, but all that is not relevant here and can be found in the BCMSN study guides from Cisco Press.

Xenz

Yeah, I know the reasons to use MST, I was more curious as whether to use it is just an arbitrary number or if there is a best practice. 3 trees with 30 VLANs? 2 trees with 20? 2 trees with 100? This is what I was curious about. It seems it's more just an arbitrary number an engineer feels comfortable using.

Forsaken_GA

Xenz wrote: »

Yeah, I know the reasons to use MST, I was more curious as whether to use it is just an arbitrary number or if there is a best practice. 3 trees with 30 VLANs? 2 trees with 20? 2 trees with 100? This is what I was curious about. It seems it's more just an arbitrary number an engineer feels comfortable using.

Well that's kind of the point we're all trying to make. It's not really about the number of VLAN's. It's more a matter of does it make sense for your layout. It's a judgment call on the part of the designer and/or engineer.

We're answering your question, it's just not the answer you're expecting to hear hehe

cisco_trooper

Xenz wrote: »

Yeah, I know the reasons to use MST, I was more curious as whether to use it is just an arbitrary number or if there is a best practice. 3 trees with 30 VLANs? 2 trees with 20? 2 trees with 100? This is what I was curious about. It seems it's more just an arbitrary number an engineer feels comfortable using.

Ahh, so if I understand you correctly it seems like you are wondering if there is a ratio of VLANs to Spanning Tree instances that is accepted as an MST best practice. For that I haven't seen anything and it doesn't seem like MST gets a whole lot of attention so there may not be one. Worth researching if you've got the time to burn.

Xenz

Forsaken_GA wrote: »

Well that's kind of the point we're all trying to make. It's not really about the number of VLAN's. It's more a matter of does it make sense for your layout. It's a judgment call on the part of the designer and/or engineer.

We're answering your question, it's just not the answer you're expecting to hear hehe

I understand. I need to find a good whitepaper on it then. The way I've read and understood it, it was more about the capability of the engineer to configure it being one criteria and the layout of the network being another. The layout being that if you have enough STP instances that use the same STP topology it's better to use MST for performance purposes. This is what I was inquiring about. I haven't found a lot of time to research MST again in the CCDA/CCDP or BCMSN authorized self study so ignore me if I get too repetitive! I thought the answer would be a simple one.

@cisco_trooper I understand the downtime with MST as it reinitializes the tree.

I'll look at MST again ASAP. Thanks for the info and patience! One thing I'm finding out about the BCMSN/BSCI book is that they are fairly worthless. I'm much better off reading whitepapers and the Doyle Routing TCP/IP books. Heavy workload this semester so studying this stuff is going to be a pain.

cisco_trooper

Xenz wrote: »

One thing I'm finding out about the BCMSN/BSCI book is that they are fairly worthless. I'm much better off reading whitepapers and the Doyle Routing TCP/IP books.

Absolutely true. The CCNP level Cisco Material gives you just enough info to hang yourself. You HAVE to use other resources if you are studying to learn. Best of luck. Good to see someone pursuing in depth understanding.