Implementaton report:draft-ietf-l3vpn-ospf-2547bis 
------------------------------------------------------

There are at least two independent implementations of draft-ietf-l3vpn-ospf-
2547bis, one  by Cisco and one  by Juniper.  Below  is the list of  the main
features of the draft, along with information as to whether the features are
supported by the implementations. 

All  of the  procedures  which  are supported  by  both implementations  are
believed  to be  interoperable.   However, these  implementations have  been
deployed for  so long  that it  isn't really possible  to cite  the specific
interoperability testing  that was  done.  At the  current time, we  are not
aware of any interoperability problems. 

1. Separate and independent instance of OSPF for each VPN. 

   Cisco: yes
   Juniper: yes

2. Proper handling of Domain Identifiers.  The use of these is optional, but
   an implementation which  doesn't use them must still  properly handle the
   reception of routes which carry domain identifiers.

   Sub-features: 

   a. Ability  to assign a Domain  Identifier to each OSPF  instance, and to
      carry  it   in  the   OSPF  Domain  Identifier   Extended  Communities
      attribute.   This must  be configurable  per OSPF  instance,  and must
      default to NULL.

      Cisco: yes
      Juniper: yes

   b. Ability  to  properly  encode  the  Domain  Identifier  as  a  Domain
      Identifier Extended Communities attribute.  

      Cisco: yes
      Juniper: yes

   c. Interpretation  of Communities  0005, 0105, 0205,  and 8005  as Domain
      Identifier Extended Communities. 

      Cisco: yes
      Juniper: yes

   d. Proper handling  of received  Domain Identifier  Extended Communities
      attributes, including  the specified rules for equality  of two domain
      identifiers, and  using Domain Identifier to help  determine whether a
      particular route is an external route.

      Cisco: yes
      Juniper: yes

   e. Recognition  that  the   absence  of  a  Domain  Identifier  Extended
      communities attribute  is functionally identical to the  carrying of a
      NULL attribute.

      Cisco: yes
      Juniper: yes


3. The ability to assign any link to any area, including area 0. 

   Cisco: yes
   Juniper: yes

4. The PE functions as ABR for any non-0 area to which it is attached. 

   Cisco: yes
   Juniper: yes

5. The OSPF  decision process installs best OSPF-distributed  route into the
   VRF. 

   Cisco: yes
   Juniper: yes

6. Translation  of OSPF  routes  into BGP-distributed  VPN-IPv4 routes  with
   following attributes:  

   * Route Type Extended Communities attribute
   * Domain Identifier (optional)
   * OSPF metric carried in MED as per specification

   Cisco: Yes
   Juniper: Yes

7. Proper setting and processing of the DN bit

   Cisco: yes
   Juniper: yes

8. Proper  setting and  processing  of  the OSPF  route  tag, including  the
   default value. 

   Cisco: yes
   Juniper: yes

9. Proper procedures for importing BGP routes into the VRF: 

     a. Eligibility for import based on RT
     b. BGP decision process selects best BGP route
     c. Preference for OSPF-distributed route over BGP-distributed route. 

    Cisco: yes
    Juniper: yes

10. BGP routes advertised  by OSPF if and only if they  are installed in the
    VRF. 

    Cisco: yes
    Juniper: yes

11. Proper procedures for translating  BGP-distributed OSPF routes into OSPF
    routes for readvertisement into OSPF.  

    a. Proper handling of BGP-distributed routes which were originally intra-
       area.  

    Cisco: yes
    Juniper: yes
 
    b. Proper handling of BGP-distribute routes which were originally inter-
       area. 

    Cisco: yes
    Juniper: yes
 
    c. Proper  handling of BGP-distribute  routes which were  originally AS-
       external.

    Cisco: yes
    Juniper: yes
 
    d. Proper handling of BGP-distribute routes which were originally NSSA..

    Cisco: yes
    Juniper: yes
 
12. Sham links (optional)
 
    a. Treated as demand circuits (optional)

    Cisco: yes
    Juniper: yes
 
    b. Sham Link Endpoint Address advertised by BGP, but NOT by OSPF.

    Cisco: yes
    Juniper: yes
 
    c. Configurable metric for sham link

    Cisco: yes
    Juniper: yes
 
    d. Proper procedures  for sending OSPF messages on  sham link, including
       the default intervals for control messages. 

    Cisco: yes
    Juniper: yes


    e. Proper procedures for forwarding data on sham links. 

    Cisco: yes
    Juniper: yes