Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

Exceeding diamater 7 of STP

Hi,

I would like to built an Ethernet based access network, but the diameter would exceed the recommended value by 802.1D (7). My questions are as follows:

1, Is the usage of Rapid STP (802.1w) a solution for that diameter problem?

2, Have the STP timer any relevance in an RSTP environment (except "hello" of course)?

3, What problems can occur and should be considered if a part of the network runs STP and the other RSTP? (Considering convergence, diameter, timers, etc.)

thx for the help

VBala'zs

1 ACCEPTED SOLUTION

Accepted Solutions

Re: Exceeding diamater 7 of STP

Hi,

I think your STP understanding is deeper than mine.

But regarding your network topology and other vendors equipment - I'd be afraid of such a complex switched network.

Haven't you thought about involving L3 distribution layer?

Regards,

Milan

6 REPLIES

Re: Exceeding diamater 7 of STP

Hi,

my understanding is:

1) yes, there is no strict recommendation regarding network diameter in RSTP. See http://www.cisco.com/warp/public/473/122.pdf

for details about 802.1d STP timers.

On the contrary, IEEE 802.1w document itself (http://standards.ieee.org/getieee802/download/802.1w-2001.pdf) is mentioning the maximum diameter as 7:

"A worst-case reconfiguration (based on a maximum diameter of 7 for the Bridged LAN) would involve the

time taken for six such handshakes to take place in sequence before the reconfiguration had been propagated from the point of the original cut to the edge of the LAN (i.e., for the entire network to reconfigure)."

2)"Classic 802.1d timers such as forward delay and max_age are nearly only used as a backup and should not be necessary if point-to-point links and edge ports are properly identified set by the administrator, and if there is no interaction with legacy bridges."

See http://www.cisco.com/warp/customer/473/146.html.

3)I'd recommend to run RPVST+, not MST. It's much easier to configure and troubleshoot. It's also recommended the common root to be in the RSTP part. But I remember there was a TCN bug in RPVST+ implementation on Cat2950 causing TCN beeing sent even while edge (portfast) port going up. So I didn't dare to move to RSTP in my network yet and I'd recomment to check Release Notes, Filed Notices, Bug Tool carefully before doing that. I'd also try to start a testing network first.

Regards,

Milan

New Member

Re: Exceeding diamater 7 of STP

Hi,

thx for your valueable information. I have studied the docs. I have found the followings:

- 802.1w: chapter 9.3.4 Validation of received BPDUs

NOTE 2 - The operation of the RSTP Port Information state machine (see 17.21) checks that the value of the BPDU's Message Age parameter is less than that of its Max Age parameter, and if not, will immediately age out the received information.

It means for me that maximum number of hops (diameter of the RSTP domain) is less than Max_Age (default=20, max of Max_Age=40). I think it is a quite big value.

But if we consider a mixed environment RSTP+STP, than we will be in problem in the STP part of the network. MST seems to be a better choice in mixed environments because the MST domains behaves like a pseudobridge and do not decrement Message Age.

(See, http://www.cisco.com/univercd/cc/td/doc/product/lan/c3550/12114ea1/3550scg/swmstp.htm#51869

"The IST and MST instances do not use the message-age and maximum-age information in the configuration BPDU to compute the spanning-tree topology. Instead, they use the path cost to the root and a hop-count mechanism similar to the IP time-to-live (TTL) mechanism.")

Do You see it similar?

best regards

VBala'zs

Re: Exceeding diamater 7 of STP

Hi,

it seems like max diameter beeing 40 in RSTP.

(The IEEE is very user-unfriendly, I consider impossible to understand it completely.)

Regarding MST: I'm not sure what Message Age value is the MST "pseudobridge" putting to BPDUs on the domain edge.

BTW, what is your network diameter, what switches are you using?

Regards,

Milan

New Member

Re: Exceeding diamater 7 of STP

Hi,

I have found the following:

802.1s: 13.15 Changing Spanning Tree Information

"The Spanning Tree Protocol (STP, Clause 8 of IEEE Std 802.1D, 1998 Edition) and the Rapid Spanning Tree Protocol (RSTP, Clause 17 of IEEE Std 802.1D, 1998 Edition) do not use a hop count and detect both circulating aged information and loss of connectivity to a neighboring bridge by means of Message Age and Max Age (maximum message age) parameters. To ensure compatibility MSTP increments Message Age for information received at the boundary of an MST Region, discarding the information if necessary."

802.1s: 13.5 Modelling an MST Region as a single RSTP Bridge

"The Port Path Cost (MSTP’s External Port Path Cost) is added to the Root Path Cost just once at the Root Port of the CIST Regional Root, the closest Bridge in the Region to the Root Bridge of the entire network. TheMessageAge used by STP and RSTP is also only incremented at this Port. If the CIST Root iswithin an MST Region it also acts as the Regional Root, and the Root Path Cost and Message Age advertised are zero, just as for a single Bridge."

Therefore MST has the feature described before.

Hop count in our Ethernet Based Access Network will be in the range 7-14. The topology is trees (max. 7 hops) connected to a ring (or a meshed topology). It will be used to collect the traffic of ADSL DSLAMs with Ethernet uplink and transport ot towards the B-RASs. Usage of 2950/3550/3750/4000 is planned. But interworking with other vendor equipment should be considered (!).

best regards

VBala'zs

Re: Exceeding diamater 7 of STP

Hi,

I think your STP understanding is deeper than mine.

But regarding your network topology and other vendors equipment - I'd be afraid of such a complex switched network.

Haven't you thought about involving L3 distribution layer?

Regards,

Milan

New Member

Re: Exceeding diamater 7 of STP

Hi,

L3 distribution layer can not be used because we use PPPoE so L2 connectivity should be ensured to the B-RAS. :-(((

Thanks for the discussion, it helped me a lot

best regards

VBala'zs

512
Views
0
Helpful
6
Replies
CreatePlease login to create content