Just to clarify a bit:

Before we can go to such low timers we need to put in place a patch that
does something like this:

https://github.com/CumulusNetworks/quagga/commit/cb1faec92248549307ffcfd93b24713d339cb8b0

Then we can safely add in whatever MRAI timers we want.  As such I intend
to take these two MRAI patches and insert them into the take-X stream after
this above patch is ported over.

donald

On Tue, Dec 8, 2015 at 7:59 AM, Daniel Walton <[email protected]>
wrote:

>
>
> On Tue, Dec 8, 2015 at 4:40 AM, Paul Jakma <[email protected]> wrote:
>
>> On Mon, 7 Dec 2015, Daniel Walton wrote:
>>
>> FWIW IOS has used 0 for iBGP and 5 for eBGP since ~2004.  We (cisco at
>>> the time) never had any complaints when we changed to these values (they
>>> were both 30s before). The only reason we didn't do 0 for eBGP was because
>>> dampening was still used a fair amount back then (not so much now).
>>>
>>
>> With the MRAIs set to 0 you can get some bursts of path hunting but they
>>> tend to be short lived because bgp ends up converging so fast (sub-second
>>> is typical).  Given the lack of complaints when the iBGP timer was changed
>>> to 0, I don't think increased path hunting is much of a concern. Path
>>> hunting is worse as meshiness increases and an iBGP core tends to be more
>>> meshy and eBGP connections.  I think we could use 0 for both without any
>>> path hunting issues.
>>>
>>
>> Interesting.
>>
>> I'd probably keep eBGP !0, just for the sake of being conservative for
>> now. BGP can produce /many/ rounds of messages, and the literature shows
>> that as you decrease the MRAI past the optimum the number of messages and
>> go toward 0 the number of messages increases very quickly again (but
>> linearly). Even just 1s, just to avoid causing bad storms somewhere between
>> different ASes.
>>
>>
> I can get on board with 0 for iBGP and 1s for eBGP :)  Maybe someday we
> can look into this draft as a means of making it 0 for ebgp:
>
> https://tools.ietf.org/html/draft-keyupate-bgp-rcn-00
>
> Daniel
>
>
>
>> Path hunting is indeed made much more worse by high MRAIs.
>>
>> iBGP, sure go to 0. Same operator, and the network will be smaller in
>> size compared to the Internet.
>>
>> Important bit: we're all agreed the standards specified values and
>> Quagga's current defaults are _way_ too high. ;)
>>
>> regards,
>> --
>> Paul Jakma      [email protected]  @pjakma Key ID: 64A2FF6A
>> Fortune:
>> Romeo wasn't bilked in a day.
>>                 -- Walt Kelly, "Ten Ever-Lovin' Blue-Eyed Years With Pogo"
>>
>
>
_______________________________________________
Quagga-dev mailing list
[email protected]
https://lists.quagga.net/mailman/listinfo/quagga-dev

Reply via email to