Skip to main content

You're viewing an archived page. It is no longer being updated.

RIPE 71

18 November 2015, 11:00-12:30

Co-Chairs: Remco van Mook and Filiz Yilmaz (standing in for Florence Lavroff)
Scribe: Marco Hogewoning

Remco welcomed the attendees and apologised on behalf of Florence, explaining she was on maternity leave and participating remotely. Filiz Yilmaz stepped forward to assist Remco with chairing this particular session.

(Matthew Walster's presentation was moved back on the agenda)

Study About Internet Interconnections: Preliminary Results – Uta Meier-Hahn, Humboldt Institute for Internet Studies

The presentation is available at: 

https://ripe71.ripe.net/presentations/96-uta-interconnection-survey.pdf

Uta was unable to join the meeting in person and sent a recording of her presentation that was played to the audience.

Salem Yamout, RIPE NCC Executive Board, asked whether Internet Exchange Points and regulation were being included in Uta's report.

BGP Communities for Blackholing Traffic – Thomas King, DECIX

The presentation is available at:
https://ripe71.ripe.net/presentations/99-e-TK-20151118-RIPE_Connect_WG_Blackholing.pdf

There were no questions

Follow Up: Working With CDNs: Towards BCOP – Florence Lavroff,

The presentation is available at:
https://ripe71.ripe.net/presentations/83-RIPE-71-Connect-Working-Group-November-18th-2015-Closure-Working-with-CDNs-towards-BCOP..pdf

Remco replaced Florence and gave an overview of the history and current status of this BCOP project. He suggested closing this work item until there were other volunteers willing to pick it up again.

There were no objections to his suggestion and the work item was closed.

Update on MANRS - Andrei Robachevsky, ISOC

The presentation is available at:

https://ripe71.ripe.net/presentations/78-201511-MANRS-RIPE71.pdf

Martin Levy (Cloudfare) referred to an incident in Asia the week before and asked the audience if anyone noticed the leaks and if their own routes were being leaked from this network. He asked who wanted networks to take action.

Randy Bush (IIJ) referred to the testimonials stating that some of the organisations that signed up to MANRS have had big incidents over the past months.

Andrei responded that work was going on to implement compliance checks, but that it poses a challenge and often non-compliance is only noted when things go wrong.

Randy commented that he was not asking for the logos to be removed, but instead suggested to develop tools to help to better implement MANRS.

Andrei pointed to work ongoing in IETF and the initiatives to develop a BCOP, especially to guide the smaller networks in implementation.

Blake Willis (L33 Networks) suggested engaging with the equipment manufacturers regarding outreach to their customers on how to properly configure their networks.

Remco rounded off the discussion asking for a show of hands on who signed up, which saw four responses. Randy Bush added that he had implemented all recommendations but did not see the value in signing up to MANRS itself.

Traffic Volume Dependencies of Large IXPs – Thomas King, DECIX

The presentation is available at:

https://ripe71.ripe.net/presentations/100-e-TK-20151116-RIPE-Traffic_Volume_Dependencies_Between_IXPs-Neutral.pdf

Ruediger Volk (Deutsche Telekom) pointed to BGP convergence time as another potential reason for traffic loss. Thomas responded that the associated article on RIPE Labs had more details on the BGP effects.

Will Hargrave (LONAP) asked Thomas whether he agreed that remote peering is bad for the Internet. Thomas responded that it is not bad when it is set up correctly.

Remco suggested this was a nice topic for a future meeting.

Countries, IXPs and RIPE Atlas – Emile Aben, RIPE NCC

The presentation is available at:

https://ripe71.ripe.net/presentations/95-2015-11.ixp-country-jedi.emileaben_V2.pdf

Kurtis Linqvuist (Netnod) suggested extending this research to all countries and highlighting significant changes over time. Randy Bush responded pointing out that the problem with time-based measurements lies in the changes in the RIPE Atlas probe topology.

Emile Aben said they could try and avoid this bias.

(The presentation by Nick Hilliard was cancelled)

Tryst / RIPE Atlas – Martin Levy, CloudFlare

The presentation is available at:

https://ripe71.ripe.net/presentations/102-RIPE71_ASNtryst_RIPE_Atlas_Hackathon_Project.pdf

For time management reasons the chair did not open the floor for questions

So, You've Decided to Peer… - Matthew Walster, Fastly

The presentation is available at:
https://ripe71.ripe.net/presentations/22-RIPE71-Peering-Safety.pdf

There were no questions.

Remco van Mook asked the audience for feedback on the working group and its topics, suggesting they contact the chairs via email.

He thanked all participants and closed the session.