With a path that long, I find it unlikely that there isn't more than one link at one point or another. How do you figure that there is only one link? Do you have back-to-back dark fiber or lambda? Because if you have virtualized packet transport, it will most definitely be carried over multiple paths. E.g. One mpls vpn link does not mean that mpls transport is not redundant somewhere.
If there is really ony one L1/L2 path, it could be caused by the packets falling into different queues, yet this is also unlikely, because all packets should have the same markings and same L3/L4 information.
What type of transport do you have between Sao Paulo and NY?
On 12.9.2011, at 14:54, Persio Pucci <persio_at_gmail.com> wrote:
> Hi folks,
>
> I am having some problems trying to figure out what could be causing UDP
> packets get out-of-sequence on some multicast streams (market data) between
> Sao Paulo and New York.
>
> There is no part of this path with parallel, load-balanced connections,
> which could be a obvious cause. What else could I check? The packets do
> arrive, so they are not being dropped on the way, but they arrive out of
> sequence, being useless to the application.
>
> Also, we are monitoring on both ends sniffed data so I know I receive them
> in order in Sao Paulo, but not in New York. This is affecting a really small
> percentage of the data (less than 0,001%), but I believe this should be 0,
> and so do my customers (with reason).
>
> Any ideas or tips?
>
>
> 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 Mon Sep 12 2011 - 15:05:50 ART
This archive was generated by hypermail 2.2.0 : Sat Oct 01 2011 - 07:26:25 ART