Re: Spanning-Tree forward delay

A good question indeed! Adel - you rock man, always good stuff.

Just to add here a little ...

The best answer would be to use RSTP ;-) ... this speeds everything up and
allows for redundant links to forward quickly, and also indirect failures
ensuring faster convergence!

As for uplink fast, it is important to note that this feature is an 'edge'
feature. When you enable uplink fast, the switch priority is incremented to
make sure that this switch is not the root. Kind of makes sense ...

if you have a lab, enable uplink fast and then view the switch priority.

Portfast is always a great idea!

Portfast with bpduguard is even better ;-)

Here is the link to the docs (watch the ugly word wrap):
http://www.cisco.com/en/US/partner/docs/switches/lan/catalyst3560/software/release/12.2_52_se/configuration/guide/swstpopt.html#wp1031144

IMO, this documentation is fairly well written and should help you to 'fill
the holes' in your knowledge. Here is a search from youtube, plenty to
choose from:

http://www.youtube.com/results?search_query=cisco+spanning+tree+&search_type=&aq=f

HTH,

Andrew Lissitz

.
On Tue, Feb 2, 2010 at 3:07 AM, Adel Abouchaev <adel_at_netmasterclass.net>wrote:

> Uplink fast is a feature that allows faster convergence of the spanning
> tree
> when it has root port candidates in blocked state. It will start forwarding
> immediately on the new root port upon the failure of the uplink. The
> meaning
> of this is different from listening and learning, since the port is already
> known to be on the path to the root, but is currently blocked by the
> algorithm. The new convergency procedures will run as the port is already
> forwarding (configuration PDU, etc.).
>
> Portfast would completely skip listening and learning. You need to be
> careful there, and know what is connected to the port and what could be
> connected to the port.
>
> Forward delay would do the job, it needs an assessment upon how small the
> value can be based on the complexity of the network. This parameter should
> be set on a root bridge for the VLAN, otherwise it will not take effect.
>
> Cheers,
>
> Adel Abouchaev, CCIE# 12037, CISSP, MCSE
>
> Netmasterclass LLC, Cisco Learning Partner
> RFC821: adel_at_netmasterclass.net
> E.164: +18886772669
> HTTP: www.netmasterclass.net
>
>
>
> -----Original Message-----
> From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
> ccie_ka_at_gmx.de
> Sent: Monday, February 01, 2010 11:47 PM
> To: ccielab_at_groupstudy.com
> Subject: Spanning-Tree forward delay
>
> Hi group,
>
> I know that this subject are often discussed. But for me is not clear if a
> question ask:
> How to reduce the listening and learning state in spanning-tree.
> My first thing that I can do is "spanning-tree vlan <id> forward-delay
> <sec>
> and also spanning-tree portfast
> But what about uplink-fast ? Shuldn't this also a solution ?
>
> Regards
>
> Dennis
> --
> Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5
> -
> sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>

-- 
Andrew Lee Lissitz
all.from.nj_at_gmail.com
Blogs and organic groups at http://www.ccie.net