Re: [bess] A question regarding Single-Active ES redundancy mode and DF election in RFC 7432

[Alexander Vainshtein <[e-mail]>]

Krzysztof,
Lots of thanks for your email. If I understand you correctly, it actually replaces Single-Active Redundancy Mode with the Port-Active one as defined in the corresponding draft<https://tools.ietf.org/html/draft-brissette-bess-evpn-mh-pa-01> (it also uses the Preference-Based DF Election<https://tools.ietf.org/html/draft-ietf-bess-evpn-pref-df-01> mechanism but tweaks it in such a way that it selects the same DF for all VLANs).

Regards,
Sasha

Office: [phone]
Cell:      [phone]
Email:   [e-mail]

From: Krzysztof Szarkowicz [mailto:[e-mail]]
Sent: Tuesday, September 4, 2018 12:14 PM
To: Alexander Vainshtein <[e-mail]>
Cc: Ali Sajassi (sajassi) <[e-mail]>; Michael Gorokhovsky <[e-mail]>; Rotem Cohen <[e-mail]>; Shell Nakash <[e-mail]>; [e-mail]; Ron Sdayoor <[e-mail]>
Subject: Re: [bess] A question regarding Single-Active ES redundancy mode and DF election in RFC 7432

Hi Sasha,

To use LAG in A/S mode on per VLAN basis, you can use ‘hack’, where you use preference-based DF election (draft-ietf-bess-evpn-pref-df) to ensure with the configuration that *all* VLANs on given ESI use PE1 as DF, and *all* VLANs on that ESI use PE2 as non-DF. Then, you need to have a mechanism to signal from PE2 to the CE the link standby status (e.g. via LACP, as used in MC-LAG deployments; or, I could even imagine per link CFM/LFM/E-LMI), so that CE doesn’t use link to PE2.

Thanks,
Krzysztof


On 2018-Sep-04, at 10:39, Alexander Vainshtein <[e-mail]<mailto:[e-mail]>> wrote:

Ali,
Lots of thanks for a prompt and detailed response.
It matches my understanding of the situation with Single-Active Redundancy Mode of Ethernet Segments in EVPN.
In particular, your confirmation that “You cannot use LAG to do active/standby on a per VLAN basis (aka EVPN single-active)” was quite important.

I have also noticed that Single-Active is not mentioned at all  in RFC 8388<https://tools.ietf.org/html/rfc8388>. I wonder what this means with regard to actual deployment of this mode.

Last but not least, I wonder if the expired draft<https://tools.ietf.org/html/draft-brissette-bess-evpn-mh-pa-01> on Port-Active multi-homing mode for EVPN will be refreshed and if, as part of such refresh, any details on the control plane of EVPN would be provided.

Regards, and, again, lots of thanks,
Sasha

Office: [phone]
Cell:      [phone]
Email:   [e-mail]<mailto:[e-mail]>

From: Ali Sajassi (sajassi) [mailto:[e-mail]]
Sent: Tuesday, September 4, 2018 8:00 AM
To: Alexander Vainshtein <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]>; Michael Gorokhovsky <[e-mail]<mailto:[e-mail]>>; Shell Nakash <[e-mail]<mailto:[e-mail]>>; Ron Sdayoor <[e-mail]<mailto:[e-mail]>>; Rotem Cohen <[e-mail]<mailto:[e-mail]>>
Subject: Re: A question regarding Single-Active ES redundancy mode and DF election in RFC 7432

Hi Sasha,

I don’t see any contradiction between the two statements from RFC 7432 that you mentioned below. For All-Active, DF election is for BUM traffic of a given VLAN (or group of VLANs in case of VLAN bundling) in the egress direction toward an ES. For Single-Active, DF election is for all traffic of a given VLAN (or group of VLANs …) in both directions of an ES. Now with respect to notification of active VLANs to a CE device: MVRP mechanism that is mentioned in the RFC is an IEEE standard way of doing such thing. However, if the CE support E-LMI, then that protocol can be used as well. Regarding LAG, it can be used to connect a CE in an active/standby mode where one link is active and another link in standby mode (assuming two-link bundle). You cannot use LAG to do active/standby on a per VLAN basis (aka EVPN single-active).

I will be travelling over next few days with limited email access, so please expect some delay for my responses.

Cheers,
Ali

From: Alexander Vainshtein <[e-mail]<mailto:[e-mail]>>
Date: Sunday, September 2, 2018 at 6:09 AM
To: Cisco Employee <[e-mail]<mailto:[e-mail]>>
Cc: "[e-mail]<mailto:[e-mail]>" <[e-mail]<mailto:[e-mail]>>, Michael Gorokhovsky <[e-mail]<mailto:[e-mail]>>, Shell Nakash <[e-mail]<mailto:[e-mail]>>, Ron Sdayoor <[e-mail]<mailto:[e-mail]>>, Rotem Cohen <[e-mail]<mailto:[e-mail]>>
Subject: A question regarding Single-Active ES redundancy mode and DF election in RFC 7432

Ali and all,
I have a question regarding one of the aspects of RFC 7432, namely operation of the default Designated Forwarder (DF) election process on an Ethernet Segment (ES) that operates in the Single-Active Redundancy Mode.

RFC 7432 defines the Single-Active Redundancy Mode in Section 3 as following:
“Only a single PE, among all the PEs attached to an Ethernet segment, is allowed to forward traffic to/from that Ethernet segment for a given VLAN”.

The same RFC in Section 8.5 also specifies that the DF for a specific VLAN on a multi-homed Ethernet segment (ES) is the only PE attached to this segment that is responsible for sending BUM traffic for this VLAN to the CE. It also defined the default DF election procedure that elects a single “live” PE on the specific ES as the DF for each specific EVI that is represented on this ES.

These two definitions look contradictory to me, because:

  1.  The default DF election procedure only involves the PEs attached to the specific ES
  2.  In the Single-Active Redundancy mode the elected DF for a specific VLAN must also be the only PE that is allowed to forward traffic received with this VLAN from the CEs to the peer PEs. It is not clear to me, how this can be achieved.

     *   The RFC mentions MVRP as a possible method to notify the attached CEs that a specific PE is NOT a DF for a specific VLAN in the case of an ES that operates in the Single-Active Redundancy Mode. Does this mean that CEs that are attached to a multi-homed ES operating in Single-Active Redundancy Mode SHOULD support MVRP?
     *   Are there any alternatives to MVRP that can be used for this purpose. In particular, is it possible to use Ethernet Local Management Interface (E-LMI) as defined in MEF-16<http://www.mef.net/resources/technical-specifications/download?id=42&fileid=file1> for this purpose?
     *   The RFC mentions LAG as the method to connect the CE to a multi-homed ES operating in the All-Active Redundancy Mode. Is it possible to connect a CE that uses LAG to a multi-homed ES operating in the Single-Active Redundancy Mode?

Your feedback would be highly appreciated.

Regards, and lots of thanks in advance,
Sasha

Office: [phone]
Cell:      [phone]
Email:   [e-mail]<mailto:[e-mail]>


___________________________________________________________________________

This e-mail message is intended for the recipient only and contains information which is
CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this
transmission in error, please inform us by e-mail, phone or fax, and then delete the original
and all copies thereof.
___________________________________________________________________________

___________________________________________________________________________

This e-mail message is intended for the recipient only and contains information which is
CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this
transmission in error, please inform us by e-mail, phone or fax, and then delete the original
and all copies thereof.
___________________________________________________________________________
_______________________________________________
BESS mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/bess


___________________________________________________________________________

This e-mail message is intended for the recipient only and contains information which is 
CONFIDENTIAL and which may be proprietary to ECI Telecom. If you have received this 
transmission in error, please inform us by e-mail, phone or fax, and then delete the original 
and all copies thereof.
___________________________________________________________________________