I was using 2 physically separate layer 3 interfaces on my router, no secondary IP addresses here (only way I could get around the Cisco next-hop limitation). Maybe the route server was only sending you the lower IP address path (step 13 of BGP selection process)??? Or maybe the original peer was up longer (step 10)?? ? It would be good to have route server access to see what the route server was seeing and sending. I can only tell what I am sending. Is there a web interface (or plan to implement one)? Sent from my iPhone On Dec 16, 2012, at 21:11, "David Farmer" <farmer@umn.edu> wrote:

Well, here is what I was seeing earlier;

* 63.127.10.0/23 69.147.218.14 180 0 25694 18606 i *> 69.147.218.14 180 0 25694 18606 i * 64.244.48.0/20 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 69.67.16.0/20 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 71.5.104.0/21 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 74.202.238.0/24 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 74.202.239.0/24 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 199.66.72.0/22 69.147.218.14 0 180 0 25694 i *> 69.147.218.14 0 180 0 25694 i * 204.17.210.0 69.147.218.14 180 0 25694 40796 i *> 69.147.218.14 180 0 25694 40796 i

And, here is what I am seeing now;

* 63.127.10.0/23 206.108.255.14 180 0 25694 18606 i *> 206.108.255.14 180 0 25694 18606 i * 64.244.48.0/20 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 69.67.16.0/20 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 71.5.104.0/21 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 74.202.238.0/24 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 74.202.239.0/24 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 199.66.72.0/22 206.108.255.14 0 180 0 25694 i *> 206.108.255.14 0 180 0 25694 i * 204.17.210.0 206.108.255.14 180 0 25694 40796 i *> 206.108.255.14 180 0 25694 40796 i

What every you did its announcing the new next-hop addresses now. Maybe you were having a similar issue to what the CDW/Berbee guys were. I apologize, and hopefully everyone's issues are as simple.

Thanks.

On 12/16/12 20:40 , Larry Patterson wrote:

...

David,

We've been announcing on both ranges for quite some time. We only left the old ones up because many bi-lateral's had not changed to our new peers yet. I've just turned down all of our old peering sessions now. I think that you are incorrect that we were announcing via only our old next-hop address. We were announcing via both, I can assure you of this. Just to be sure (and to be compliant with the schedule), I've turned down all of our old IPv4 and IPv6 sessions. Please let me know if you see any issues.

Thanks, Larry Patterson ­ Chief Technology Officer Atomic Data 615 North 3rd Street Minneapolis, MN 55401 612.466.2000 Main Line 612.466.2080 Direct Dial Twitter <http://twitter.com/AtomicData> | Facebook <http://www.facebook.com/AtomicDataCenters> | LinkedIn <http://www.linkedin.com/companies/atomic-data-centers> Simple. Safe. Smart. www.atomicdata.com <http://www.atomicdata.com/> This message and any information or attachments included therewith is intended only for the individual or entity named above. If the reader is not the intended recipient, you are hereby notified that any use, dissemination, distribution or copy of this message or any information attached or contained therein is strictly prohibited. If you have received this message in error, please notify the sender by return e-mail and destroy all copies of the message and any attachments. Thank you.

Please consider the environment before printing this email.

On 12/16/12 8:22 PM, "David Farmer" <farmer@umn.edu> wrote:

...

I apologize for implying you hadn't done you part and thanks for fixing it, I now see you announcing the new next-hop address.

On 12/16/12 20:11 , James Stahr wrote:

...

I can't speak for anyone else, but I thought I had completed the changes by swapping my $C secondary addresses around a few weeks ago and thought I was done. But it looks like I did the swap too quickly and even though 69.147.218.26 was now a secondary address and all of my BGP sessions were using 206.108.255.26 as the source IP, I was still advertising the next-hop using the old block. Turns out if the BGP sessions don't drop, the next-hop won't change. As I've previously stated, my $C equipment cannot advertise routes with BOTH next-hops.

-James

-----Original Message----- From: David Farmer [mailto:farmer@umn.edu] Sent: Sunday, December 16, 2012 7:46 PM To: MICE Discuss; peering@wiktel.com; James Stahr; mwilker@jagcom.net; servicenotices@atomicdata.com Cc: farmer@umn.edu Subject: MICE Renumbering

OK, I think it is now time for a little public shaming; the following MICE participants are only announcing their prefixes using the old next-hop addresses through the route servers. We were all suppose to be using only the new addresses as of two weeks ago. There is one other participant announce with both the old and the new next-hop address, we (Northern Lights GigAPOP) just stopped doing this earlier today, and are now announcing using only the new next-hop addresses. All other participants are announcing to the route servers are using only the new next-hop addresses.

AS33362 Wikstrom Telephone Company (Wiktel) AS25694 Atomic Data AS15011 Jaguar Communications AS3599 CDW/Berbee

Please convert your peering session to announcing with the new next-hop addresses, ASAP. If for some reason you can't do this in the next week, please send a note to the MICE-DISCUSS mailing list detailing when you will be converting.

As a reminder, here is the original cutover schedule that was emailed out several months ago:

Addition of Address Secondary: before 10/1/2012 New Prefix Announcements: after 10/1/2012 Final use of old IP address next-hop: before 11/30/2012 Removal of old IP address range: after 12/1/2012 Project End: 12/31/2012

Thanks. -- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

-- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

########################################################################

To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

-- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

On Sun, Dec 16, 2012 at 09:11:00PM -0600, David Farmer wrote:

What every you did its announcing the new next-hop addresses now. Maybe you were having a similar issue to what the CDW/Berbee guys were. I apologize, and hopefully everyone's issues are as simple. ...

There is an oddity in the route servers, in such that, even though they are configured to present the router-ID in the new session as the new IP, that they have been seen to be using the old IP address to connect to the remote host for the new session. It could be this sort of oddness that triggers some sides to be announcing out the old next-hop for their blocks. This is what I accounted for on one announcement previously. Digging into the route tables on the route servers, only one AS isn't dual announcing, and since you already named it, we could do the push on them. AS 15011. Although they did respond to me privately a couple weeks ago that they know they need to get this done ASAP. The others that are showing up with 69.147.218/24 next-hops are all dual-announcing as far as I can tell, and a few blocks just happen to have choosen the announcement with the old-next hop in there (especially on the 2nd RR), but all others are dual-announcing. It is almost to the point where we should clean up the route servers. We are sitting pretty good for the renumbering project for the final stretch. I think everybody but AS15011 is ready to go and drop the old block, although the schedule allows that to have happened on the 1st of the month already. -- Doug McIntyre <merlyn@iphouse.net> -- ipHouse/Goldengate/Bitstream/ProNS -- Network Engineer/Provisioning/Jack of all Trades ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1