- ripe-atlas - mailman.ripe.net

Atlas probe offline
by Steve Hill 11 Jan '16

11 Jan '16

My Atlas probe (ID 14587) has been showing as offline since December 16th. I can see DHCP requests/replies, it ARPs for the gateway and I can see DNS requests and pings from the probe, and successful replies to all of this traffic being sent back to the probe. e.g.: 12:13:38.790096 IP 192.168.0.115.35404 > 217.146.115.154.53: 2+ AAAA? ctr-nue15.atlas.ripe.net. (42) 12:13:38.797337 IP 217.146.115.154.53 > 192.168.0.115.35404: 2 1/6/12 AAAA 2a01:4f8:120:14ee::2 (482) 12:13:38.798179 IP 192.168.0.115.57785 > 217.146.115.154.53: 3+ A? ctr-nue15.atlas.ripe.net. (42) 12:13:38.799649 IP 217.146.115.154.53 > 192.168.0.115.57785: 3 1/6/12 A 78.46.87.51 (470) 12:14:58.990977 IP 192.168.0.115.37934 > 217.146.115.154.53: 2+ AAAA? IUSB-READONLY.U36.M6466B3B0EC98.sos.atlas.ripe.net. (68) 12:14:59.060132 IP 217.146.115.154.53 > 192.168.0.115.37934: 2 1/1/2 AAAA 2001:67c:2e8:11::c100:1337 (164) 12:14:59.060983 IP 192.168.0.115.57523 > 217.146.115.154.53: 3+ A? IUSB-READONLY.U36.M6466B3B0EC98.sos.atlas.ripe.net. (68) 12:14:59.114915 IP 217.146.115.154.53 > 192.168.0.115.57523: 3 1/1/2 A 193.0.19.55 (152) 12:14:59.115925 IP 192.168.0.115 > 193.0.19.55: ICMP echo request, id 954, seq 0, length 64 12:14:59.142586 IP 193.0.19.55 > 192.168.0.115: ICMP echo reply, id 954, seq 0, length 64 12:15:00.116800 IP 192.168.0.115 > 193.0.19.55: ICMP echo request, id 954, seq 1, length 64 12:15:00.143773 IP 193.0.19.55 > 192.168.0.115: ICMP echo reply, id 954, seq 1, length 64 I can't see any other IPv4 traffic, but I do see IPv6 HTTPS traffic. I'm not sure where to go next with figuring out why the probe seems to have vanished. Any ideas would be greatly appreciated. Many thanks. -- - Steve

6 8

Examination of ARM based Atlas telemetry nodes
by Michael Schloh von Bennewitz 10 Jan '16

10 Jan '16

Hello developers, I'm scheduled to talk at the forthcoming MINIX conference, and was hoping to have the chance to examine your ARM based Atlas telemetry system. I only know of the Raspberry Pi devices loaned to the Tor project, and this could be relevant to my IoT work (leveraging IPv6 and ARM.) Andrew Tanenbaum recommended I write to Philip, is he here? Cheers, Michael -- Michael Schloh von Bennewitz Software Development Engineer Europalab Networks R&D, Munich Office: +49(89)44239885 UTC+2 Mobile: Same as 'Office' VoIP: sips:michael@schloh.com Web: http://michael.schloh.com/

3 2

Availability report accuracy
by Lorenzo Colitti 03 Jan '16

03 Jan '16

Hi, mostly out of curiosity: what sort of accuracy are the numbers in the monthly availability report? For example: the report says that my probe was "connected" 99.95% of the time (down for 20 seconds) in December. What does that mean? That the SSH control connection was not up? That the probe didn't have link? That it couldn't reach any targets? Something else? Cheers, Lorenzo

4 4

Seven new RIPE Atlas anchors have joined the community
by Michela Galante 31 Dec '15

31 Dec '15

Dear RIPE Atlas users, Since the previous announcement, seven new RIPE Atlas anchors have been activated bringing the total number to 164. - nc-nou-as56055, hosted by Micro Logic Systems (MLS) in Nouméa, New Caledonia. - bg-sof-as47872, hosted by Sofia Connect in Sofia, Bulgaria. - bg-sls-as198228, hosted by Silistra Telecom Solutions in Silistra, Bulgaria. - fr-ysd-as201958, hosted by Claranet in Saint-Denis, France. - de-erl-as680, hosted by Friedrich-Alexander-Universität Erlangen-Nürnberg in Erlangen, Germany. - cr-sjo-as263779, hosted by NIC Costa Rica in San José, Costa Rica and sponsored by LACNIC. - nl-roo-as43350, hosted by NFOrce Entertainment in Roosendaal, Netherlands. The map of all anchor locations is available at: https://atlas.ripe.net/anchors/map/ You can access the anchoring measurements for all anchors at: https://atlas.ripe.net/anchors/list/ And here are the logos and links to hosting companies: https://atlas.ripe.net/get-involved/community/#!tab-anchor-sponsors We are still accepting new applications from organisations interested in hosting a RIPE Atlas anchor. We are particularly interested in deploying anchors in the Middle East, Western Asia, Latin America and Africa, in order to add more geographical diversity to the measurement data. There are more organisations that want to host a RIPE Atlas anchor but don’t have the necessary funding. To learn more about sponsoring an anchor, please contact us at mcb(a)ripe.net To apply for your own RIPE Atlas anchor: https://atlas.ripe.net/get-involved/become-an-anchor-host/ For any other questions, please contact us at atlas(a)ripe.net We’d like to thank all the anchor hosts and the organisations sponsoring anchors for their support in 2015. Thanks to you it was possible to expand the coverage and reach this amazing goal. Wishing all of you the best for 2016. Michela Galante on behalf of the RIPE Atlas Team RIPE NCC

1 0

How to deal with inaccurate geocoding of probes
by Klaus Darilion 30 Dec '15

30 Dec '15

Hi! Inspired by https://perso.telecom-paristech.fr/~drossi/index.php?n=Dataset.Anycast I wanted to detect my Anycast locations using the Atlas probes. The problem is, that my results are only as accurate as the geocoding of the Atlas probes. For example probe #19273 is reported to be somewhere in Kansas. Bit the probe is only 5ms away from my LAX server and also traceroute indicates the probe is in LAX (traceroute to 63.84.150.1). So, now I am seeking some advice how to handle such problems. Is there some tag that indicates "verified location"? Any other ideas? Thanks Klaus

2 1

Ticket response times
by Micha Bailey 30 Dec '15

30 Dec '15

Hi, what's the average time for responses to atlas(a)ripe.net tickets? I've been waiting almost 2 months to hear back on 1193254, and was wondering if that's a typical wait time or if it fell between the cracks...

3 2

Anchor probes with broken lat/lon
by Klaus Darilion 29 Dec '15

29 Dec '15

https://atlas.ripe.net/api/v1/probe/6155/ "latitude": 2228.7695, "longitude": -4412.6105, https://atlas.ripe.net/api/v1/probe/6166/ "latitude": 2840.3705, "longitude": 631.8105, regards Klaus

2 1

Exploring the Idea of RIPE Atlas Virtual Probes
by Vesna Manojlovic 29 Dec '15

29 Dec '15

Dear colleagues, Some of you have asked about Virtual Probes on this list. Although we don't plan to make virtual probes available in 2016, we do plan to investigate this idea and develop some prototypes. Please find more details on RIPE Labs: https://labs.ripe.net/Members/suzanne_taylor_muzzin/exploring-the-idea-of-r… Kind regards, Vesna Manojlovic RIPE NCC

5 7

Support for arbitrary DNS queries
by Shumon Huque 26 Dec '15

26 Dec '15

I'd like to request the ability to have RIPE Atlas probes be able to issue DNS queries for arbitrary record types, and have that capability exposed in the API. I'm particularly interested in newer DANE record types like TLSA, OPENPGPKEY, SMIMEA, etc but a general purpose mechanism would be useful. Is this feasible? Are others interested in this capability? Thanks! Shumon Huque

11 19

13 new zones added to DNSMON
by Vesna Manojlovic 22 Dec '15

22 Dec '15

2 2

DNS measurement failed
by Maciej Andziński 16 Dec '15

16 Dec '15

Hi, Since yesterday I've been encountering a problem, many of the DNS measurements end with "failed" status. It happens for every measurement that tries to query 8.8.8.8, 8.8.4.4 or when using "Use the Probe's Resolver(s)" feature. However, everything works fine when querying a-dns.pl (194.181.87.156) server. Sample measurements: 3082732 3083505 3083581 Do you observe any issues related to the lack of available probes / exceeded limits? Many greetings, Maciej Andziński, NASK

3 4

Number of packets now ignored for ICMP tests?
by Stephane Bortzmeyer 16 Dec '15

16 Dec '15

I asked just one packet: {'definitions': [{'description': 'Ping 2001:67c:217c:6::2 from AS #3215', 'af': 6, 'packets': 1, 'type': 'ping', 'is_oneoff': True, 'target': '2001:67c:217c:6::2'}], 'probes': [{'requested': 500, 'type': 'asn', 'value': '3215', 'tags': {'include': ['system-ipv6-works']}}]} But, for the nine probes that were selected, I got 27 tests, which is the default of 3 packets per probe. Strange. Measurement #3083491

2 2

[request] AD bit in DNS query
by Maciej Andziński 15 Dec '15

15 Dec '15

Hi, Apropos of the discussion about arbitrary QTYPE in DNS measurement, I would like to make a request for another small improvement. It would be nice if a probe had the ability to send a DNS query with AD bit set. Such query would be useful in some cases (for instance when using "Use the Probe's Resolver(s)" feature) and would reduce data volume. According to RFC6840 section 5.7: 5.7. Setting the AD Bit on Queries The semantics of the Authentic Data (AD) bit in the query were previously undefined. Section 4.6 of [RFC4035] instructed resolvers to always clear the AD bit when composing queries. This document defines setting the AD bit in a query as a signal indicating that the requester understands and is interested in the value of the AD bit in the response. This allows a requester to indicate that it understands the AD bit without also requesting DNSSEC data via the DO bit. Many greetings, Maciej Andziński, NASK

1 0

more than 10 concurrent measurements to the same target
by Klaus Darilion 12 Dec '15

12 Dec '15

Hi! When scheduling checks I often get the error: "We do not allow more than 10 concurrent measurements to the same target." Imo this is not correct. Yes, I do schedule more then 10 measurements to the same target, but they are not concurrent. They are sequential - one ofter the other. Thus, it would be great if this check could be fixed to consider start and stop times. Thanks Klaus

4 4

DNSmon "not indicative of what happens to normal traffic" claims the root ops
by Stephane Bortzmeyer 11 Dec '15

11 Dec '15

http://root-servers.org/news/events-of-20151130.txt

6 11

Probe Address Discovery
by Mike 10 Dec '15

10 Dec '15

On the Network tab for my probe, there is a section named, "Probe Address Discovery". It shows the addresses on my probe, placing a green check mark next to the one(s) it thinks are the public address(es) for the probe. For my probe, the IPv4 address shown in the IP Echo Service row has a green check mark. Unfortunately, that IPv4 address was replaced about a month ago by another IPv4 address, and it can no longer be considered to be an address for my probe. Additionally, the IPv6 address for the IP Echo Service is also incorrect, but that one doesn't have a green check mark next to it. Is it a problem that these addresses are incorrect? Is there a way to refresh them, or have them refreshed, with the correct IP addresses? thanks.

1 0

What is the Aggregator on Map of DNS Measurements
by Klaus Darilion 09 Dec '15

09 Dec '15

Hi! The MAP of DNS Measurements contains an "Aggregator" option. Unfortunately the online help (Enter a regular expression ....) does not help much. Is this feature somewhere described in detail? I do not understand it. Thanks Klaus

2 2

CAIDA BGP Hackathon 6-7 Feb 2016 -- Call for Participation
by Vesna Manojlovic 09 Dec '15

09 Dec '15

Hi everyone, as a continuation of our own successful RIPE Atlas events, we are now taking part in CAIDA's BGP "live" hackahton, by providing data, measurements platform, expertise -- and you ;-) RIPE NCC's services, such as RIS & RIPE Atlas, will be "remixed" with other BGP-related data sources & tools, to create wonderful new hacks. As participants in RIPE Atlas community, you are very much invited to take part in this event. Details are below. Regards, Vesna -------- Forwarded Message -------- Subject: [mat-wg] CAIDA BGP Hackathon 6, 7 Feb 2016 -- Call for Participation Date: Tue, 8 Dec 2015 20:37:49 -0800 From: Alberto Dainotti <alberto(a)caida.org> To: mat-wg(a)ripe.net What: CAIDA BGP Hackathon, Call for Participation (CFP) When: Feb 6-7, 2016 Where: the San Diego Supercomputer Center (SDSC) on the University of California at San Diego (UCSD) campus. Who: Parties interested in hacking on live BGP data. Inquiries: bgp-hackathon-info(a)caida.org Important Deadline: Dec 22, 2015 for travel grant requests Apply to attend: http://www.caida.org/workshops/bgp-hackathon/1602/ CAIDA, in cooperation with CSU, USC, UFMG, FORTH, Route Views, RIPE NCC, will host a BGP Hackathon on the 6-7th of February 2016 at UC San Diego (La Jolla, CA, USA). The theme of the hackathon is “live BGP measurements and monitoring”. We will provide participating teams with access to data sources and a toolbox: live streaming of BGP data, the new BGPMon interface, BGP processing tools and APIs such as the opensource BGPStream software framework, the PEERING testbed, RIPE RIS, visualization tools, and data-plane active measurement platforms such as CAIDA Ark and RIPE Atlas. Participating teams will work on "challenges" that extend, integrate and demonstrate the utility of these platforms/data for understanding or solving practical problems (e.g., detecting BGP prefix hijacking, evaluating anycast performance, effectively visualizing phenomena). The hackathon will be held in San Diego the weekend immediately preceding the NANOG conference and the AIMS academic Internet measurement workshop. The hackathon aims to: - bring together these different communities, e.g., to discuss problems operators face that academics may want to research; - advertise tools (e.g., PEERING peering.usc.edu, BGPStream bgpstream.caida.org) to the communities and get people familiar with using them, and encourage further use in the future; - get people working together on interesting/important problems, hopefully spurring further collaboration on these problems; - provide extra incentive for students to attend NANOG. ### FORMAT Each team of 2-4 people will work on a "challenge". The organization committee will propose a set of challenges to bootstrap feedback and refinement based on community input: participants can propose totally new challenges, modifications to existing ones, and express their preferences. The set of potential challenges is described in the event website at https://github.com/CAIDA/bgp-hackathon/wiki/List-of-Challenges and will continuously evolve in the days preceding the hackathon. On Saturday morning, participants will very briefly introduce their interests and ideas and teams will be officially formed. On Saturday and Sunday, participants will work together in teams to hack and develop their ideas, culminating in very short presentations to the jury on Sunday evening and a party to announce the winning teams and celebrate everyone's participation. During the event, domain-experts will provide support to the teams. Participants are free to work on drafted challenges and on their own ideas in the days preceding the hackathon. A mailing list and documentation will provide support on the platforms/tools/data used in the hackathon. However, for the teams to compete in the hackathon for prizes, they will have to demonstrate that substantial work was done during the two-day event. ### PARTICIPATION AND TRAVEL GRANTS Application to participate in the hackathon is open from December 8, 2016. Application does not guarantee participation. There is no application fee, and no participation fee. We will accept applications until one week before the hackathon, or until capacity is reached. However, the deadline to apply for a travel grant (through the hackathon application form) is 22 December 2015. Travel grants reimburse flights and hotel expenses. Food and drinks will be provided throughout the two-day event. http://www.caida.org/workshops/bgp-hackathon/1602/

1 0

Map-based visualisations for Atlas data
by Baptiste Jonglez 08 Dec '15

08 Dec '15

Hi, Is there a way to reuse and adapt the code for the maps available here? https://atlas.ripe.net/results/maps/ It could be interesting to have a generic map-based visualisation, where all probes matching a certain set of conditions would be displayed. Possible applications include: - monitor IPv6 deployment for an AS, by displaying all probes matching "asn_v6 = XX". It would probably be useful to also display probes matching "asn_v4 = XX", as a baseline, using a different color - monitor DNS resolver outage geographically, by matching on system tags like "resolving A doesn't work". - match on user tags, like "IXP" or "core", to monitor deployment of Atlas at specific locations/types of networks. In all those examples, the "time slider" functionality would be extremely useful, to see the evolution over time. It could be even nicer if the slider could move automatically, so as to provide nice animations. Is all this currently feasible, and if not, would it be hard to implement? Thanks, Baptiste

3 2

Turris Omnia as probe Platform?
by Kessler, Thomas 08 Dec '15

08 Dec '15

Dear all, some of you are probably aware of the Turris Omnia project by CZ.NIC ( https://www.indiegogo.com/projects/turris-omnia-hi-performance-open-source-… ). This might be an interesting platform for running an VM probe?! Just thought it might be worth brining it up here for discussion. Cheers Thomas

7 9

New on RIPE Labs: DomainMON in RIPE Atlas
by Chris Amin 07 Dec '15

07 Dec '15

Dear colleagues, A few weeks ago we gave you a heads up about a new visualisation tool in RIPE Atlas called DomainMON, and we're pleased to announce that this is now available for all RIPE Atlas users with available credits. Please find more details on RIPE Labs: https://labs.ripe.net/Members/suzanne_taylor_muzzin/ripe-atlas-domainmon-is… Kind regards, Chris Amin RIPE NCC

1 1

Probe power down.
by Alistair Mackenzie 07 Dec '15

07 Dec '15

Hi all, I'm currently moving house and expect my probe to be offline for approximately one week, Is there any formal channels that I must notify ripe of this other than the location change once it's back online? Thanks, Alistair

3 3

Dead probes
by Marty Strong 02 Dec '15

02 Dec '15

Hi, I've got 1 or 2 dead probes, they power on but I see no mac address from them. They have the first 2 lights on, but none of the others. I don't see this combination listed on https://atlas.ripe.net/about/faq/#what-do-the-lights-on-the-side-of-the-pro… Has anybody else come across this issue? Anybody know what the light combination means? Marty

9 15

location field in probe API
by Bajpai, Vaibhav 01 Dec '15

01 Dec '15

Hello, TL;DR: The probe API does not expose the probe location even if I explicitly ask for it. location: A human readable location for this probe, either explicitly specified by the probe owner, or derived from the longitude and latitude. It appears the 'location' field in the probe API is hidden by default and it must be explicitly set in the request parameters to expose location field values. However, even if I do set it, I still don't get location entries. I made multiple permutations of query requests and also tried the usage example on the API documentation webpage [1]: Get the IPv4 addresses and the usually hidden human readable location string for public probes in Germany: https://atlas.ripe.net/api/v1/probe/?country_code=de&is_public=true&fields=… What could be wrong? -- Thanks! PS: It seems, this field is also missing in the probe archive API. [1] https://atlas.ripe.net/docs/rest/#probe Best, Vaibhav ===================================================== Vaibhav Bajpai Room 91, Research I School of Engineering and Sciences Jacobs University Bremen, Germany www.vaibhavbajpai.com =====================================================

1 0

Fwd: Re: Measurements delete and ping issue
by Daniel Karrenberg 30 Nov '15

30 Nov '15

I did not copy the list on this message and then realised that it is a generic answer that deserves to be seen and archived here. Daniel -------- Forwarded Message -------- Subject: Re: [atlas] Measurements delete and ping issue To: Andrea Consadori <andrea.consadori(a)hotmail.it> References: <DUB126-W888F5B62064D1CE913A99DE0000(a)phx.gbl> From: Daniel Karrenberg <daniel.karrenberg(a)ripe.net> Message-ID: <565C9760.2050704(a)ripe.net> Date: Mon, 30 Nov 2015 19:37:20 +0100 On 30.11.15 15:42 , Andrea Consadori wrote: > > > Hi all, > i just start using proble atlas and i found is not possible to delete > Stopped measurements from web, it's normal? > i try to search on old mailing list messages and seems to be and old > issue but i still cannot delete stopped masurements > (for example ids: 3050379 > <https://atlas.ripe.net/measurements/3050379/>, 3034900 > <https://atlas.ripe.net/measurements/3034900/>) > ... Andrea, An important idea behind RIPE Atlas is that all results are re-useable. Therefore we store the results of *all* measurements as long as we possibly can because those results can be useful for other researchers even if they did not run the measurements themselves. Each result says something about the packet flow in the Internet at the particular time it was obtained. There is no way we can foresee how these results can help to answer interesting research questions and help us to understand the Internet packet layer better than we do now. For a recent example of this, please see https://labs.ripe.net/Members/becha/ripe-atlas-tools-hackathon-results "ASN Tryst". Therefore we do not provide the functionality of deleting measurements. The only way we foresee that results can be deleted is if we run out of storage space. Of course that one has to be careful to select the measurements one wants to re-use according to the question one wants to answer. Your measurement shows this very well. I do not recommend using your results unless the question is about reachability with RTTs under 6ms. I hope this answers your question Daniel (co-inventor of RIPE Atlas)

1 0