From: afiddler (afiddler@xxxxxxxxx)
Date: Sun Sep 16 2001 - 15:32:39 GMT-3
There is a "cost" parameter, which works similarly to a floating static in
the world of routing protocols. The default is 3, so a cost of 5 would
cause the backup link to be used only if the primary link was down.
The following is taken directly from CCO (there looks to be more information
on CCO and CD-ROM on this topic):
When there are multiple peers to a given destination, use the cost keyword
to determine which router is preferred and which is capable. The cost
keyword only applies in fault tolerance mode.
In full-tolerance mode, the preferred path is always used unless it is
unavailable. The preferred path is either the path over which the first
response to an explorer was received, or, in the case of remote peers, the
peer with the least cost.
The cost keyword specified in a remote peer statement takes precedence over
the cost learned as part of the capabilities exchange with the remote peer.
The cost keyword is relevant only in fault tolerance mode.
----- Original Message -----
From: "Bob Chahal" <bob.chahal@ntlworld.com>
To: "Fear, Russell H" <Russell.Fear@capgemini.co.uk>
Cc: <ccielab@groupstudy.com>
Sent: Sunday, September 16, 2001 10:07 AM
Subject: Re: DLSW back-up peer slow
> Not sure what you can do because when the primary goes down circuits have
to
> re-establish across the backup peer anyway so it doesn't matter how
quickly
> the backup peer takes to come up. That's what I think, but I'm not 100%
> sure.
>
> ----- Original Message -----
> From: "Fear, Russell H" <Russell.Fear@capgemini.co.uk>
> To: "'Bob Chahal'" <bob.chahal@ntlworld.com>
> Cc: <ccielab@groupstudy.com>
> Sent: Sunday, September 16, 2001 1:33 PM
> Subject: RE: DLSW back-up peer slow
>
>
> > I think you're right. Is it correct to leave the connection to timeout
in
> > this way or should I be adjusting something to reduce this ? The lab
> > question I was working on said 'ensure no bridged traffic is lost as a
> > result of the primary connection failing.' I havent got any traffic
> > travelling over the link but assume it would be lost in the long
> fail-over.
> > Am I right ?
> >
> > Thanks for your help,
> >
> > Russell
> >
> >
> > -----Original Message-----
> > From: Bob Chahal [mailto:bob.chahal@ntlworld.com]
> > Sent: 16 September 2001 13:01
> > To: Fear, Russell H; ccielab@groupstudy.com
> > Subject: Re: DLSW back-up peer slow
> >
> >
> > Isn't this just the time the primary peer will take to timeout the DLSW
> peer
> > connection in the absence of DLSW keepalives.
> > ----- Original Message -----
> > From: "Fear, Russell H" <Russell.Fear@capgemini.co.uk>
> > To: <ccielab@groupstudy.com>
> > Sent: Sunday, September 16, 2001 11:55 AM
> > Subject: DLSW back-up peer slow
> >
> >
> > > I am setting up a DLSW back-up peer in my lab and notice that it takes
a
> > > long time to swap over on failure of the primary peer. The switch back
> is
> > > fairly quick but the swap over on failure of the primary is very slow.
> > >
> > > Is there something such as a timer or time-out parameter that I should
> be
> > > tweaking to reduce this lag ?
> > >
> > > TIA
> > >
> > > Russell
> > >
> > >
> > > * 700 2201 Internal
> > > * +44 ( 0 ) 870 238 2201 External
> > > * russell.fear@capgemini.co.uk
> > >
> > >
> > >
> > >
> > >
> > >
> >
>
****************************************************************************
> > ****************
> > > " This message contains information that may be privileged or
> confidential
> > and
> > > is the property of the Cap Gemini Ernst & Young Group. It is intended
> only
> > for
> > > the person to whom it is addressed. If you are not the intended
> recipient,
> > you
> > > are not authorized to read, print, retain, copy, disseminate,
> distribute,
> > or use
> > > this message or any part thereof. If you receive this message in
error,
> > please
> > > notify the sender immediately and delete all copies of this message ".
> > >
> >
>
****************************************************************************
> > ****************
> > > **Please read:http://www.groupstudy.com/list/posting.html
> > **Please read:http://www.groupstudy.com/list/posting.html
> **Please read:http://www.groupstudy.com/list/posting.html
**Please read:http://www.groupstudy.com/list/posting.html
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:32:18 GMT-3