Re: Max routes in VRF?

Hi Jacques,

   For Question 2 ...I *think* TCAM usage of vpnv4 routes is double than the
plain ipv4 route,

Thanks

Regards
Anantha Subramanian Natarajan

On Fri, Jul 3, 2009 at 1:21 AM, Jacques Allison <jacques_at_zallison.com>wrote:

> Thanks for the valuable feedback.
>
> As below, it's a sup720-3bxl (Show modules)
>
> 5 2 Supervisor Engine 720 (Active) WS-SUP720-3BXL
>
> And this is the max routes,
> Host# show mls cef maximum-routes
> FIB TCAM maximum routes :
> =======================
> Current :-
> -------
> IPv4 + MPLS - 512k (default)
> IPv6 + IP Multicast - 256k (default)
>
>
> Also, each line card has "L3 Forwarding Resources" that can be used:
> L3 Forwarding Resources
> Module FIB TCAM usage: Total Used
> %Used
> 4 72 bits (IPv4, MPLS, EoM) 1048576 7625
> 1%
> 144 bits (IP mcast, IPv6) 7755 135
> 2%
>
> detail: Protocol Used
> %Used
> IPv4 3
> 1%
> MPLS 7622
> 1%
> EoM 0
> 0%
>
> IPv6 5
> 1%
> IPv4 mcast 130
> 2%
> IPv6 mcast 0
> 0%
>
> Adjacency usage: Total Used
> %Used
> 1048576 7755
> 1%
>
> Some follow up question:
> 1. has anyone imported the global routing table (GRT) into a VRF without
> any
> problems?
> 2. having GRT routes in IPv4 or VPNv4 VRF has the same TCAM usage?
> 3. I assume that all PE devices that import the vrf routes need enough FIB
> TCAM for MPLS.
> 4. same with each line card should have enough FIB TCAM available.
> 5. with "mls cef maximum-routes" , if you increase the mpls allocation it
> would decrease ip automatically ?
> 6. how does mls cef maximum-routes relate to module FIB TCAM usage?
> 7. what issues would you foresee if I have a transit AS with internet VRF
> with all global routes?
>
> Regards,
> J
>
>
> -----Original Message-----
> From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
> Alex
> H. Ryu
> Sent: 02 July 2009 11:12 AM
> To: Jacques Allison
> Cc: ccielab_at_groupstudy.com
> Subject: Re: Max routes in VRF?
>
> If you are talking about SUP720-3BXL, probably not.
>
> You can check current TCAM usage using following command.
>
> sh platform hardware capacity forwarding
>
> Also, you can check maximum values.
>
> show mls cef maximum-routes
>
>
> If you want to change TCAM partition, use following command from
> configuration mode.
>
> (config)#mls cef maximum-routes ?
> ip number of ip routes
> ip-multicast number of multicast routes
> ipv6 number of ipv6 routes
> mpls number of MPLS labels
>
> (config)#mls cef maximum-routes ip ?
> <1-1007> number of 1K entries
>
> (config)#mls cef maximum-routes ip-multicast ?
> <1-503> number of 1K entries
>
> (config)#mls cef maximum-routes ipv6 ?
> <1-503> number of 1K entries
>
> (config)#mls cef maximum-routes mpls ?
> <1-1007> number of 1K entries
>
>
> Since global full BGP table is too large - over 280k nowadays -, if you
> want to keep it in VRF, single instance will be 280k or so.
> That means that if you wants to keep 3 VRFs with full table, it will be
> 280k * 3 , which is 880k entries.
> It is beyond the limit.
>
>
> Alex
>
>
> Jacques Allison wrote:
> > Would it be possible to run eBGP to multiple upstream ISP from with VRF
> > address-family and get all global Internet routes? I would like to know
> the
> > maximum limits of routes within a VRF. I know it would depend on hardware
> -
> > let say it's a SUP720BXL with specs of 1 million routes. would this also
> fit
> > into VRF?
> >
> > Thanks
> > J
> >
> >
> > 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
>
>
> 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
Received on Sun Jul 05 2009 - 22:07:04 ART