Re: Understanding RSTP - Please can someone explain this?

From: ALL From_NJ (all.from.nj@gmail.com)
Date: Fri Feb 06 2009 - 15:59:22 ARST

 Good topic and discussion.

People can get pretty 'enthused' with STP and how they have seen it / done
it. As usual, there may be more than one way to skin a cat ...

With this said however, a few points that I would suggest / recommend.

We know that STP keeps loops, mistakes, etc ... from causing additional
issues. I have heard people suggest disabling it for voice, or servers, or
whatever links. When L1 and L2 problems occur, in technical terms, these
problems suck.

Many times it is not the senior net folk making these mistakes, ... but
atlas these problems do occur. STP can keep these mistakes or problems from
spreading and becoming a wide spread problems. How do you t-shoot a wide
spread problem ... not easily. I would suggest to always keep STP turned
on.

All_CCIE Question related to this conversation:

- Configure a port to come up immediately when a device is plugged in, and
bypass the learning and listening phases of STP. If a BDPU is received, the
port should revert back to it's default STP process and not be able to
bypass the learning and listening phases of STP. Furthermore, whenever a
new port is configured to come up immediately, this behavior will also apply
to this newly configured port as well. (2 points)

My 2 cents ...

Andrew Lee Lissitz , TGIF (not .. a new certification)

On Fri, Feb 6, 2009 at 11:37 AM, Ravi Singh <way2ccie@googlemail.com> wrote:

> As I said . I stand to be corrected .. Thanks for correcting me operator
> sid
> . I haven't labbed up much on switching yet, I think. Between RSTP & STP ,
> PVST simply slipped out of my mind. Here are a few clarifications
>
> You can enable RSTP in a Cisco switch by using the *spanning-tree mode
> rapid-pvst *global command. Alternatively, you can simply enable 802.1s
> MST,
> which by definition uses 802.1w RSTP
> For the command to change the link type I just read a command
> *spanning-tree
> link-type point-to-point* . I have never tested this, so I am not sure if
> this really is the command.
>
> Do you have the book Certification guide 3rd edition by Wendell Odom. there
> is a bit of more (different)stuff on this .
>
> I have a few rack sessions scheduled this weekend and I will definitely try
> these things out. Thanks for the eye opener , yet again ;-)
>
>
> Ravi
>
> On Fri, Feb 6, 2009 at 3:31 PM, operator sid <ccie1@live.co.uk> wrote:
>
> >
> > Thank Ravi
> >
> > But few things dont seem correct.
> >
> > You said
> >
> > "RSTP does not require any additional configuration by itself , though
> the
> > features which make it RSTP , obviously need to be configured . For
> example
> > you won't explicitly enable RSTP but you would enable portfast,
> uplinkfast
> > or backbonefast with their relevant commands."
> >
> >
> > However Cisco doc says, you have to enable RSTP by using rapid-pvst mode
> > instead of the old pvst+ mode as shown below: (For MST, RSTP is built
> > within, so you just need to enable MST mode)
> >
> > Services1(config)#spanning-tree mode rapid-pvst
> >
> > Further Cisco doc says:
> >
> > "Spanning Tree UplinkFast and BackboneFast features are PVST+ features.
> > These are disabled when you enable rapid-PVST+ because those features are
> > built within rapid-PVST+. Therefore, during the migration you can remove
> > those commands. "
> >
> > "However, The configuration of the features such as PortFast, BPDUguard,
> > BPDUfilter, root guard, and loopguard are applicable in rapid-PVST+
> mode."
> >
> > Regarding the questions i asked
> >
> > 1) you seem to agree with Cisco doc on this, however i agree with what
> Fred
> > Quoted in his email from his ARCH book. I Quote again:
> >
> > "Even if the recommended design does not depend on STP to resolve link
> > or node failure events, STP is required to protect against user-side
> > loops. There are many ways that a loop can be introducted on the
> > user-facing access-layer ports. Wiring mistakes, misconfigured end
> > stations, or malicious users can create a loop. STP is required to
> > ensure a loop-free topology and to protect the rest of the network
> > from problems created in the access layer."
> >
> > This is very contradictory, as this is also from cisco source.
> >
> > 2) yOU mentioned change the Duplex setting, in many implementations i
> have
> > seen Auto rather then Full-duplex configured on links, and link may
> > negotiate as half 100 (although this may not be ideal) does that mean
> RSTP
> > will not work and will revert back to legacy STP. How many people
> actually
> > notice this? Seems like something you can easily get wrong inadvertently.
> >
> > Regards...
> >
> >
> > ----------------------------------------
> > > Date: Fri, 6 Feb 2009 14:54:22 +0000
> > > Subject: Re: Understanding RSTP - Please can someone explain this?
> > > From: way2ccie@googlemail.com
> > > To: ccie1@live.co.uk
> > > CC: ccielab@groupstudy.com
> > >
> > > Well .. The document really has everything you need to know. Anyhow , I
> > will
> > > try to clear your doubts in simpler terms .
> > >
> > > RSTP does not require any additional configuration by itself , though
> the
> > > features which make it RSTP , obviously need to be configured . For
> > example
> > > you won't explicitly enable RSTP but you would enable portfast,
> > uplinkfast
> > > or backbonefast with their relevant commands.
> > > RSTP is called Rapid only because of these features hence these are a
> > part
> > > of RSTP.
> > >
> > > coming to your questions
> > > 1) should STP be disabled on edge ports all together ? As a good
> practice
> > ,
> > > Yes. Your explanation has the answer "If RSTP is not disabled on edge
> > ports,
> > > convergence times will be excessive for packets traversing those ports
> "
> > .
> > > There is no use for STP to be enabled to a port connected to a PC. so
> why
> > > wait for 30 seconds for the port to realise not to run STP on it.
> > >
> > > 2)What is the configuration to explicitly configure RSTP link types?
> > Again
> > > your explanation has the answer "A port that operates in fullduplex is
> > > assumed to be point-to-point, while a half-duplex port is considered as
> a
> > > shared port by default". So change the duplex settings explicitly and
> you
> > > can change the link type .
> > >
> > > HTH, and I definitely stand to be corrected so anyone with any comments
> > is
> > > most welcome.
> > >
> > > Regards,
> > > Ravi
> > > On Fri, Feb 6, 2009 at 2:37 PM, operator sid wrote:
> > >
> > >>
> > >> I have read it,and unfortunately its not a good document, i was hoping
> > >> someone could explain better...
> > >>
> > >>
> > >> ________________________________
> > >>> Date: Fri, 6 Feb 2009 14:09:48 +0000
> > >>> Subject: Re: Understanding RSTP - Please can someone explain this?
> > >>> From: way2ccie@googlemail.com
> > >>> To: ccie1@live.co.uk
> > >>> CC: ccielab@groupstudy.com
> > >>>
> > >>> This might help if you have not read it before
> > >>>
> > >>>
> > >>>
> > >>>
> > >>
> >
> http://www.cisco.com/en/US/tech/tk389/tk621/technologies_white_paper09186a0080094cfa.shtml
> > >>>
> > >>>
> > >>>
> > >>> Ravi
> > >>>
> > >>>
> > >>> On Fri, Feb 6, 2009 at 12:25 PM, operator sid> wrote:
> > >>>
> > >>> Hi Group
> > >>>
> > >>>
> > >>>
> > >>> I am still confused about how RSTP is implemented. From what i
> > understand
> > >> the major difference is that STP
> > >>>
> > >>>
> > >>> used Timers for Loop prevention whereas RSTP coordinates between
> > >> neighbors via messages (proposal/aggreement) to turn on links
> > >>>
> > >>> more quickly after topology changes and is "timer free".
> > >>>
> > >>>
> > >>>
> > >>>
> > >>> However, I have not noticed any difference in the configuration from
> > the
> > >> legacy STP
> > >>>
> > >>> configurations and the RSTP configuration on cisco devices. Or are
> > there
> > >> any differences??
> > >>>
> > >>>
> > >>>
> > >>> I have read in documentation that RSTP natively includes features
> > >>>
> > >>> like UplinkFast, BackboneFast and PortFast. So are these features now
> > >>> obsolete
> > >>> and not needed to be configured if you are running RSTP.
> > >>> (Although i have seen Portfast still configured along with RSTP on
> many
> > >>> switches)
> > >>>
> > >>>
> > >>>
> > >>>
> > >>> Also can someone explain the below Points from Cisco Documentation
> > >>>
> > >>>
> > >>>
> > >>> 1) should STP be disabled on edge ports all together as suggested
> > below?
> > >>>
> > >>>
> > >>>
> > >>> "STP edge ports are bridge ports that do not need STP enabled, where
> > loop
> > >> protection is not needed out
> > >>>
> > >>>
> > >>> of that port or an STP neighbor does not exist out of that port. For
> > >> RSTP, it is important to disable STP
> > >>>
> > >>> on edge ports, which are typically front-side Ethernet ports, using
> the
> > >> command bridge
> > >>>
> > >>> bridge-group-number spanning-disabled on the appropriate interface.
> If
> > >> RSTP is not disabled on edge
> > >>>
> > >>>
> > >>> ports, convergence times will be excessive for packets traversing
> those
> > >> ports."
> > >>>
> > >>>
> > >>>
> > >>> 2) It seems RSTP relies on duplex setting to determine inter-switch
> > >> links. What is the configuration to explicitly
> > >>>
> > >>>
> > >>> configure RSTP link types? (I couldnt find this in the documentation)
> > >>>
> > >>>
> > >>>
> > >>> "RSTP can only achieve rapid transition to the forwarding state on
> edge
> > >> ports and on point-to-point links.
> > >>>
> > >>> The link type is automatically derived from the duplex mode of a
> port.
> > A
> > >> port that operates in fullduplex
> > >>>
> > >>>
> > >>> is assumed to be point-to-point, while a half-duplex port is
> considered
> > >> as a shared port by
> > >>>
> > >>> default. This automatic link type setting can be overridden by
> explicit
> > >> configuration. In switched
> > >>>
> > >>> networks today, most links operate in full-duplex mode and are
> treated
> > as
> > >> point-to-point links by RSTP.
> > >>>
> > >>>
> > >>> This makes them candidates for rapid transition to the forwarding
> > state."
> > >>>
> > >>>
> > >>>
> > >>> Also i am a bit rough on my RSTP knowledge even after skimming a
> > >>> few Cisco documents. Can someone please explain this in simple way.
> > >>>
> > >>>
> > >>>
> > >>> Thanks in advance
> > >>> _________________________________________________________________
> > >>> Check out the new and improved services from Windows Live. Learn
> more!
> > >>> http://clk.atdmt.com/UKM/go/132630768/direct/01/
> > >>>
> > >>>
> > >>>
> > >>> Blogs and organic groups at http://www.ccie.net
> > >>>
> > >>>
> _______________________________________________________________________
> > >>> Subscription information may be found at:
> > >>>
> > >>> http://www.groupstudy.com/list/CCIELab.html
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>
> > >> _________________________________________________________________
> > >> Love Hotmail? Check out the new services from Windows Live!
> > >> http://clk.atdmt.com/UKM/go/132630768/direct/01/
> > >
> > >
> > > Blogs and organic groups at http://www.ccie.net
> > >
> > > _______________________________________________________________________
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> >
> > _________________________________________________________________
> > Windows Live Messenger just got better .Video display pics, contact
> updates
> > & more.
> > http://www.download.live.com/messenger
>
>
> 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@gmail.com

Blogs and organic groups at http://www.ccie.net

This archive was generated by hypermail 2.1.4 : Sun Mar 01 2009 - 09:44:10 ARST