Re: BFD stability follow-up from IETF-91

[Manav Bhatia <[e-mail]>]

Hi Vengada,

Let me at the outset say that i am in favor of such a draft. I think
its an important and a useful work that should be taken up by the WG.

However, i have issues in how it current stands. I think it needs a
wee bit more polishing.

>   Often (definitely more than once), we have felt the need to carry sequence numbers in
> BFD packets for debugging packet losses/ delayed delivery etc. This extension will
> increase the ease of debugging BFD flaps caused due to delayed delivery, silent drops etc.

There are ways to detect BFD drops but then those are implementation
specific and i will refrain from going there. So i concede that it
would help in packet losses. However i dont see this helping much in
case of delayed delivery (please see my earlier email).

Can the authors add some text on how this debugging mechanism would
work if somebody employs BFD authentication?

Cheers, Manav

> Thanks
> Prasad
>
> -----Original Message-----
> From: Rtg-bfd [mailto:[e-mail]] On Behalf Of Mach Chen
> Sent: Thursday, November 27, 2014 2:13 PM
> To: Marc Binderberger
> Cc: [e-mail]
> Subject: RE: BFD stability follow-up from IETF-91
>
> Hi Marc,
>
>> -----Original Message-----
>> From: Marc Binderberger [mailto:[e-mail]]
>> Sent: Thursday, November 27, 2014 1:43 AM
>> To: Mach Chen
>> Cc: Manav Bhatia; [e-mail]
>> Subject: RE: BFD stability follow-up from IETF-91
>>
>> Hello Mach,
>>
>> > This triggers me think out there should be another solution for
>> > getting the Tx and Rx timestamps without encoding the timestamps in
>> > the BFD
>> packets.
>> > For example, the Tx and Rx systems could just save timestamps
>> > locally or send them to a centralized entity and then use the
>> > sequence numbers to correlate them for further analyzing.
>>
>> I remember some discussion on NVO3 about how many bits it takes ;-) -
>> could you send the links/draft names you are working on to this list?
>> May be useful for further discussions.
>
> Sure, here is the link(http://tools.ietf.org/html/draft-chen-ippm-coloring-based-ipfpm-framework-02) for the reference.
>
> But here I want to say is that since we have sequence number, we may not need the marking based solution. Suppose that someone want to monitor the delay of a BFD packet , just record and save the timestamp at the Tx side, which indexed by the sequence number. Similarly, do the same at the Rx side. Then based on the timestamps from both Tx and Rx, and using the sequence number to correlate the timestamps, it can also provide a way to monitor the delay of the BFD packet.
>
> That means, only if there is sequence number, even if without carrying the timestamp in the BFD packet, BFD packet delay can be measured.
>
> Best regards,
> Mach
>
>>
>>
>> Thanks & Regards,
>> Marc
>>
>>
>>
>> On Wed, 26 Nov 2014 09:17:32 +0000, Mach Chen wrote:
>> > Hi Marc and Manav,
>> >
>> >> -----Original Message-----
>> >> From: Rtg-bfd [mailto:[e-mail]] On Behalf Of Marc
>> >> Binderberger
>> >> Sent: Wednesday, November 26, 2014 4:50 PM
>> >> To: Manav Bhatia
>> >> Cc: [e-mail]
>> >> Subject: Re: BFD stability follow-up from IETF-91
>> >>
>> >> Hello Manav,
>> >>
>> >>> I believe the work is important and addresses something thats
>> >>> really required (spent too much time debugging why BFD flapped!).
>> >>
>> >> agree :-) we should keep the discussion alive.
>> >>
>> >>
>> >>> side Time stamping would have helped in debugging whether the BFD
>> >>> packet was sent late, or whether the packet was sent on time and
>> >>> also arrived on time but was delayed when passing it up the BFD
>> >>> stack/processor (lay in the RX buffer for tad too long)
>> >>
>> >> well, I can see a point in having the Tx timestamps in the packet
>> >> mainly for the purpose of knowing "this" packet was okay/not okay
>> >> on the Tx side and to correlate it with your local Rx measurement.
>> >
>> > Yes, this is one solution if people think BFD delay is needed. If
>> > allow to have Tx timestamps to be carried in the packets, seems it
>> > should be no problem to leave a seat for the Rx timestamps as well
>> > :-). After all, with both Tx and Rx timestamp, it may simplify the
>> implementation.
>> >
>> >>
>> >> And even this point is less relevant with sequence numbers as this
>> >> number allows the identification of packets and thus the
>> >> correlation of information from the Tx and Rx system.
>> >
>> > Indeed, the sequence number helps a lot for the correlation between
>> > the Tx and Rx system.
>> >
>> > This triggers me think out there should be another solution for
>> > getting the Tx and Rx timestamps without encoding the timestamps in
>> > the BFD
>> packets.
>> > For example, the Tx and Rx systems could just save timestamps
>> > locally or send them to a centralized entity and then use the
>> > sequence numbers to correlate them for further analyzing.
>> >
>> > Best regards,
>> > Mach
>> >
>> >>
>> >>
>> >> Regards, Marc
>> >>
>> >>
>> >>
>> >>
>> >>
>> >>
>> >> On Wed, 26 Nov 2014 12:26:41 +0530, Manav Bhatia wrote:
>> >>> Hi Jeff,
>> >>>
>> >>> I vividly remember the original intent of the stability draft was
>> >>> to help debug BFD failures -- to isolate the issue at the RX or
>> >>> the TX side Time stamping would have helped in debugging whether
>> >>> the BFD packet was sent late, or whether the packet was sent on
>> >>> time and also arrived on time but was delayed when passing it up
>> >>> the BFD stack/processor (lay in the RX buffer for tad too long),
>> >>> etc. But then time stamping came with its own set of issues, and
>> >>> was hence dropped from the original draft.
>> >>>
>> >>> Can the authors send a summary on the list on why time stamping
>> >>> was dropped so that we're all clear on that one.
>> >>>
>> >>> The current proposal does help but is not complete.
>> >>>
>> >>> Assume that the RX end loses a BFD session and learns later that
>> >>> it did eventually receive the missing BFD packets (based on the seq #).
>> >>> How would it know which end was misbehaving? Was it a delay at the
>> >>> TX side, or was it the RX that delayed passing the packets to the
>> >>> BFD process(or). This is usually what we want to debug and i want
>> >>> to understand how this draft with sequence numbers can
>> >>> unequivocally tell me that.
>> >>>
>> >>> I believe the work is important and addresses something thats
>> >>> really required (spent too much time debugging why BFD flapped!).
>> >>> Clearly what would help is putting a small section that describes
>> >>> how we can use the sequence numbers to debug what and where things went wrong.
>> >>>
>> >>> Cheers, Manav
>> >>>
>> >>>
>> >>> On Wed, Nov 26, 2014 at 5:49 AM, Jeffrey Haas <[e-mail]> wrote:
>> >>>> draft-ashesh-bfd-stability-01 was presented again during IETF-91
>> >>>> in Honolulu.  The slides can be viewed here:
>> >>>>
>> >>>> http://www.ietf.org/proceedings/91/slides/slides-91-bfd-4.pptx
>> >>>>
>> >>>> To attempt to simplify the presentation, the contentious portion
>> >>>> of the timers were removed from the proposal, leaving only the
>> >>>> sequence numbering for detecting loss of BFD async packets.
>> >>>>
>> >>>> When the room was polled to see whether the draft should be
>> >>>> adopted as a WG item, the sense of the room was very quiet.  As
>> >>>> promised, this is to inquire for support for this draft on the WG
>> >>>> mailing list to make sure the whole group has a voice.
>> >>>>
>> >>>> It should be noted that post-meeting discussion on the fate of
>> >>>> this draft noted that BFD authentication code points are
>> >>>> plentiful and are available with expert review.  Should the draft
>> >>>> authors wish to continue this work as Experimental, that is an option.
>> >>>>
>> >>>> -- Jeff
>> >>>>
>> >>>
>> >
>