From: Nick Shah (nshah@connect.com.au)
Date: Mon Sep 16 2002 - 23:09:19 GMT-3
Sean
Most real world/most labs specify to bring up second channel upon a detected
load.Thats where you will need the dialer-load threshold, once again that
means change in bandwidth (regardless of Virtual Intf, because lets not
forget that VI is a bundle of links, whether 1 or 2) resulting in eventual
change LSA being generated upon change in cost.
So, In that case you will need to explicitly nail the bw. or cost.
What was observed earlier was that even with your config (always bring up 2
channels, disregarding the load) a change LSA was generated every few
seconds or so, after various tries it was determined that the only way to
keep the line quiet was to specify bw or ospf cost.
rgds
Nick
----- Original Message -----
From: "Sean" <sean_ccie@yahoo.com.cn>
To: "Nick Shah" <nshah@connect.com.au>; "Castelino, Flavian"
<Flavian.Castelino@nexinnovations.com>; "Frank Maisano"
<FrankM@netarch.com>; "Jim Brown" <Jim.Brown@caselogic.com>;
<ccielab@groupstudy.com>
Cc: <ccielab@groupstudy.com>
Sent: Tuesday, September 17, 2002 11:53 AM
Subject: Re: OSPF Demand-Circuit Oddity (How do I keep it quiet)
>
> Nick,
>
> See my config below, and a few show command dumps.
> I don't think this is a IOS issue.
>
> Thanks,
>
> Sean
>
>
> ******************************************
>
> w2 -- ISDN -- w5,
>
> There are some other routers connected with w2 and w5,
> I just skip them.
>
> Both w2 and w5 are in area 0. w2 is 12.1(15.7), w5 is 12.2(6g).
>
> w5#sh run int bri0
> Building configuration...
>
> Current configuration : 333 bytes
> !
> interface BRI0
> ip address 192.168.100.2 255.255.255.252
> encapsulation ppp
> ip ospf demand-circuit
> no ip mroute-cache
> dialer idle-timeout 60
> dialer map ip 192.168.100.1 name w2 broadcast 20408
> dialer load-threshold 1 outbound
> dialer-group 1
> isdn switch-type basic-5ess
> ppp authentication chap
> ppp multilink
> end
>
>
> w2#sh run int bri 0
> Building configuration...
>
> Current configuration : 354 bytes
> !
> interface BRI0
> ip address 192.168.100.1 255.255.255.252
> encapsulation ppp
> ip ospf demand-circuit
> no ip mroute-cache
> dialer idle-timeout 90
> dialer fast-idle 30
> dialer map ip 192.168.100.2 name w5 broadcast 20508
> dialer load-threshold 1 outbound
> dialer-group 1
> isdn switch-type basic-5ess
> ppp authentication chap
> ppp multilink
> end
>
>
> w2#sh ip ospf neighbor
>
> Neighbor ID Pri State Dead Time Address
Interface
> 192.168.254.5 1 FULL/ - - 192.168.100.2 BRI0
> 192.168.254.3 1 FULL/DROTHER 00:00:38 200.100.100.3
Ethernet0
> 192.168.254.7 1 FULL/DR 00:00:30 200.100.100.7
Ethernet0
>
> w5#sh ip ospf neighbor
>
> Neighbor ID Pri State Dead Time Address
Interface
> 192.168.254.7 1 FULL/ - 00:00:31 192.168.100.129 Serial0
> 192.168.254.2 1 FULL/ - - 192.168.100.1 BRI0
> 192.168.254.6 1 FULL/DR 00:00:37 100.100.100.6
Ethernet0
>
> w5#sh ip ospf int bri0
> BRI0 is up, line protocol is up (spoofing)
> Internet Address 192.168.100.2/30, Area 0
> Process ID 1, Router ID 192.168.254.5, Network Type POINT_TO_POINT,
Cost:
> 1562
> Configured as demand circuit.
> Run as demand circuit.
> DoNotAge LSA allowed.
> Transmit Delay is 1 sec, State POINT_TO_POINT,
> Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
> Hello due in 00:00:04
> Index 1/1, flood queue length 0
> Next 0x0(0)/0x0(0)
> Last flood scan length is 1, maximum is 3
> Last flood scan time is 0 msec, maximum is 8 msec
> Neighbor Count is 1, Adjacent neighbor count is 1
> Adjacent with neighbor 192.168.254.2 (Hello suppressed)
> Suppress hello for 1 neighbor(s)
>
> =================================================================
> Now after I pinged 192.168.100.2 from w2, I saw the followings:
> =================================================================
>
> w5#
> 1d17h: %LINK-3-UPDOWN: Interface BRI0:1, changed state to up
> 1d17h: %LINK-3-UPDOWN: Interface Virtual-Access1, changed state to up
> Vi1 DDR: Dialer statechange to up
> Vi1 DDR: dialer protocol up
> 1d17h: %LINEPROTO-5-UPDOWN: Line protocol on Interface BRI0:1, changed
state to
> up
> 1d17h: %LINEPROTO-5-UPDOWN: Line protocol on Interface Virtual-Access1,
changed
> state to up
> 1d17h: %LINK-3-UPDOWN: Interface BRI0:2, changed state to up
> 1d17h: %ISDN-6-CONNECT: Interface BRI0:1 is now connected to 20408 w2
> 1d17h: %LINEPROTO-5-UPDOWN: Line protocol on Interface BRI0:2, changed
state to
> up
> 1d17h: %ISDN-6-CONNECT: Interface BRI0:2 is now connected to 20408 w2
> w5#sh ip ospf int bri0
> BRI0 is up, line protocol is up (spoofing)
> Internet Address 192.168.100.2/30, Area 0
> Process ID 1, Router ID 192.168.254.5, Network Type POINT_TO_POINT,
Cost:
> 1562
> Configured as demand circuit.
> Run as demand circuit.
> DoNotAge LSA allowed.
> Transmit Delay is 1 sec, State POINT_TO_POINT,
> Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5
> Hello due in 00:00:02
> Index 1/1, flood queue length 0
> Next 0x0(0)/0x0(0)
> Last flood scan length is 1, maximum is 3
> Last flood scan time is 0 msec, maximum is 8 msec
> Neighbor Count is 1, Adjacent neighbor count is 1
> Adjacent with neighbor 192.168.254.2 (Hello suppressed)
> Suppress hello for 1 neighbor(s)
>
> ================================================================
> See ospf cost on bri 0 not changes, and interface vi1 is up now
> ================================================================
>
>
>
> --- Nick Shah <nshah@connect.com.au> 5DU}ND#:> Sean
> >
> > I had labbed it up a while ago, and every time the second channel kicked
in
> > (depending on load-threshold), we had a change in bandwidth, so
'nailing'
> > the bandwidth seemed a good idea (we can use b/w or cost to accomplish
> > this).
> > There was a long thread a while ago, check it out.
> >
> > However, now that I see that you have done your tests, I would think
that it
> > would more likely be an IOS issue.
> >
> > rgds
> > Nick
> > ----- Original Message -----
> > From: Sean <sean_ccie@yahoo.com.cn>
> > To: Nick Shah <nshah@connect.com.au>; Castelino, Flavian
> > <Flavian.Castelino@nexinnovations.com>; Frank Maisano
<FrankM@netarch.com>;
> > Jim Brown <Jim.Brown@caselogic.com>; <ccielab@groupstudy.com>
> > Cc: <ccielab@groupstudy.com>
> > Sent: Tuesday, September 17, 2002 6:49 AM
> > Subject: Re: OSPF Demand-Circuit Oddity (How do I keep it quiet)
> >
> >
> > > Nick,
> > >
> > > I doubt about this. In my lab, I did not use "ip ospf cost #" to
specify
> > > the cost under bri int, and I used ppp multilink. It keeps quiet w/o
any
> > > problem. When I ping through the isdn bri, I can see both bri0:1 and
> > bri0:2
> > > up, and yet the "show ip ospf int bri 0" still show the cost of 1564
(for
> > > single B channel of 64kbps).
> > >
> > > I don't think ppp multilink will change ospf cost over bri line since
when
> > > multilink is enabled and up, a virtual-access interface is dynamically
> > created
> > > to bundle the traffic, so the original bri cost is not touched.
> > >
> > > I suggest Jim do a "debug ip ospf monitor" to see whether there is any
> > change
> > > LSA when the isdn line is in active.
> > >
> > > Thanks,
> > >
> > > Sean
> > >
> > >
> > >
> > >
> > > --- Nick Shah <nshah@connect.com.au> 5DU}ND#:> When using MLPP, a
change
> > in
> > > bandwidth (when 2 links come up) is considered
> > > > a change in OSPF cost of an interface, hence a change LSA is
generated.
> > What
> > > > you need to do in this case is to either nail the bandwidth as 128
or
> > > > explicitly specify a cost 9999 (for historic reasons, specify 9999,
even
> > > > though in this case since there are no parallel paths you can
specify
> > actual
> > > > cost)
> > > >
> > > > Nick
>
> _________________________________________________________
> Do You Yahoo!?
> PBOJ5=5W,Si@V5=<R - QE;"MF3vCb7QSi@V5gWSV\1(!
> http://cn.ent.yahoo.com/newsletter/index.html
This archive was generated by hypermail 2.1.4 : Mon Oct 07 2002 - 07:43:54 GMT-3