- ripe-atlas - mailman.ripe.net

Re: [atlas] [ripe.net #1133019] Re: Your RIPE Atlas Probe (ID: 16784) is not connected to our network
by Daniel Suchy 01 Aug '14

01 Aug '14

It's surprising, that you can't manage probe remotelly when it isn't "connected" (even you can directly reach it on L3). Technically, probe runs on Linux, so you can have SSH daemon with restricted IP access running here (listenning daemon has minimal performance overhead) - by current "behavior" you simply loose any way to diagnose problem - and dependence of established tunnel is another layer, which can fail... As stated before - simple rebooting is just a *workaround* of any issue, not a solution (but it seems that's your basic recomendation). I think you should work on improvements on probe diagnostics in such cases - new hardware versions of probes seems to be more problematic compared to older probes, as I can see also on atlas mailing list. I remember, that I had to plug out/in USB flash as some step with probe before (also your recomended "solution"). Also on probe should run some kind of watchdog, which reboots probe in case of similar problem automatically - probably it doesn't run now... With regards, Daniel On 31.7.2014 17:31, Philip Homburg wrote: > Hi Daniel, > > On Tue Jul 29 19:29:20 2014, danny(a)danysek.cz wrote: >> What checks did you performed at your side before sending such email? >> You don't have any kind of "emenergency" probe management, like SSH >> shell access directly to the device? What do you see in *your* logs >> about affected probe before it was disconnected? > > The mail gets sent when the is not connected for some time. Nothing more. > > There are no open ports on the probes. The probe can only be controlled when it is connected. > > There is some evidence in the logs that there was already something wrong with the probe when it disconnected. Most likely there is something wrong with the filesystem on the USB flash drive, but that is hard to say. > > Philip >

6 6

Re: [atlas] Paris-traceroute variations
by FANOU Roderick 31 Jul '14

31 Jul '14

Hi all, Just a quick question. What is the real difference between a TCP paris-traceroutes and an UDP paris-traceroute ? How do the probes perform each of them? Thanks, Best regards, Roderick On 26.06.2014 13:13, Philip Homburg wrote: > Hi Juan, > > On 2014/06/19 16:59 , Juan Antonio Cordero Fuertes wrote: >> Not sure this is the right place to ask this... sorry if it is not. >> >> I'm trying to configure Paris-traceroute measurements, and it is not >> clear for me what is the meaning of the /paris/ parameter. In >> https://atlas.ripe.net/docs/udm/ it is said that it corresponds, for >> values from 1 to 16, to "the number of variations to be used for a Paris >> traceroute <http://www.paris-traceroute.net/>". What is this? Does it >> correspond to the number of initial probes to be used by >> paris-traceroute? I am unable to figure it out from the RIPE Atlas >> docs... any indication would be appreciated. > > Paris-traceroute tries to make sure that all packets of a traceroute > take the same route through a load balancer. This in contrast to > traditional traceroute where packets from different hops typically take > different routes when load balancers are involved. > > However, in the case of paris-traceroute it is still interesting to find > out if there are multiple routes or not. For this reason, the traceroute > measurement creates different variations that may take a different route. > > Each interval, it will try one variation. So if you select 16 variations > then it will take 16 intervals before you get back to the first one. > > Philip

2 1

Re: [atlas] [ripe.net #1133019] Re: Your RIPE Atlas Probe is not connected to our network
by Daniel Suchy 29 Jul '14

29 Jul '14

There's no firewall / ACL, anything on the way, just pure L3 device. Probe worked properly, then disconnected without any logical reason with that notification received. Probe is pingable on both IPv4 and IPv6 from outside, proper (so there's no L3 issue on the way). Probe itself resets all incoming TCP connections from my IP address located in another ASN (probably due to firewall at the probe). On traffic mirrored from the probe I see ocasional NTP queries to some servers, so "something" on probe runs. Nothing else. You do not provide any access to probe for hosts to check anything directly, so I expect you can manage fully (as this is something running any kind of Linux?). As probe has public IPv4/IPv6 address, you should not have problem reach it. Unfortunatelly, probe is located on remote datacenter with limited physical access. Also this is not a solution of problem - with reboot you probably loose anything to diagnose causes of this problem. What checks did you performed at your side before sending such email? You don't have any kind of "emenergency" probe management, like SSH shell access directly to the device? What do you see in *your* logs about affected probe before it was disconnected? With regards, Daniel On 22.7.2014 11:17, Alastair Strachan wrote: > Hello Daniel, > > Thanks for your email. > > It would appear there is something on your network blocking the probe from connecting outside your network. Could I ask you to double check to ensure there is no security/firewall rule that has been updated or needs updating to allow the probe a clean connection outside your network. > > I would also recommend trying a complete power recycle on the probe. > > Thank you for taking the time to host a RIPE Atlas probe. > > If you have any further questions, please let me know > > Kind Regards, > > Alastair Strachan > > RIPE Atlas Team > > RIPE NCC Customer Satisfaction Survey > Tell us about your customer services experience by filling out the anonymous, three-question RIPE NCC customer satisfaction > > survey:https://www.ripe.net/contact/survey/satisfaction-cs/ > >> >> On 18.7.2014 05:40, RIPE Atlas wrote: >>> Dear RIPE Atlas probe host, >>> >>> Thank you for participating in RIPE Atlas! >>> >>> We have noticed that your probe is not connected to our infrastructure and has been >> disconnected since 2014-07-11 03:15 UTC. Could you please check >> whether everything is set up correctly on your side? If it is, and >> your probe is still reported as disconnected when you log in at >> https://atlas.ripe.net, please reply to this email so that we can >> follow up. >>> >>> You can review the instructions for setting up your probe at >> https://atlas.ripe.net/getting-started >>> >>> If you have any questions, please contact us at atlas(a)ripe.net >>> >>> Best regards, >>> >>> RIPE Atlas Team >>> >>> ********** >>> You are receiving this message because you applied for or were given >> a RIPE Atlas probe. We will only send messages when your probe >> becomes disconnected for an extended period of time. >>> >> > >

1 0

Fwd: Probe 2317 (home fibre) is disconnected
by Colin Johnston 29 Jul '14

29 Jul '14

strange, connectivity seems fine to me at least within BT network range and HSO network range ? Should I reset the probe or this is a bigger monitoring probe failure ? Colin Begin forwarded message: > From: RIPE Atlas <atlas(a)ripe.net> > Subject: Probe 2317 (home fibre) is disconnected > Date: 29 July 2014 12:30:05 BST > To: colin johnston <colinj(a)mx5.org.uk> > Delivered-To: colinj(a)mx5.org.uk > Received: by 10.70.41.145 with SMTP id f17csp332256pdl; Tue, 29 Jul 2014 04:30:32 -0700 (PDT) > Received: from koko.ripe.net (koko.ripe.net. [2001:67c:2e8:11::c100:1348]) by mx.google.com with ESMTPS id f10si19194016wic.74.2014.07.29.04.30.19 for <colinj(a)mx5.org.uk> (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 29 Jul 2014 04:30:20 -0700 (PDT) > Received: from janus.atlas.ripe.net ([193.0.19.13]) by koko.ripe.net with esmtps (UNKNOWN:AES256-GCM-SHA384:256) (Exim 4.72) (envelope-from <atlas(a)ripe.net>) id 1XC5bZ-0004TU-Nf for colinj(a)mx5.org.uk; Tue, 29 Jul 2014 13:30:08 +0200 > Received: from localhost.localdomain ([127.0.0.1] helo=janus.atlas.ripe.net) by janus.atlas.ripe.net with esmtp (Exim 4.72) (envelope-from <atlas(a)ripe.net>) id 1XC5bZ-00004y-IA for colinj(a)mx5.org.uk; Tue, 29 Jul 2014 11:30:05 +0000 > X-Received: by 10.180.75.230 with SMTP id f6mr41435469wiw.5.1406633431897; Tue, 29 Jul 2014 04:30:31 -0700 (PDT) > Return-Path: <atlas(a)ripe.net> > Received-Spf: none (google.com: atlas(a)ripe.net does not designate permitted sender hosts) client-ip=2001:67c:2e8:11::c100:1348; > Authentication-Results: mx.google.com; spf=neutral (google.com: atlas(a)ripe.net does not designate permitted sender hosts) smtp.mail=atlas(a)ripe.net > Content-Type: text/plain; charset="utf-8" > Mime-Version: 1.0 > Content-Transfer-Encoding: 7bit > Message-Id: <20140729113005.32636.40047(a)janus.atlas.ripe.net> > X-Ripe-Spam-Level: / > X-Ripe-Spam-Report: Spam Total Points: 0.4 points pts rule name description ---- ---------------------- ------------------------------------ -1.0 ALL_TRUSTED Passed through trusted hosts only via SMTP -0.7 RP_MATCHES_RCVD Envelope sender domain matches handover relay domain -1.9 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000] 4.0 DCC_CHECK Detected as bulk mail by DCC (dcc-servers.net) > X-Ripe-Signature: 407c82a9d3629e01ca957e4b84b7f67191521e4a20cb459dafae86d3fab252b9 > > Dear colin, > > Your probe 2317 has been disconnected since 2014-07-29 10:38:01 UTC. > > This is an automatically generated email from RIPE Atlas. It was sent to you because you asked to be notified if your probe becomes disconnected for more than 30 minutes. > > If you want to change this, or disable this notification altogether, please go to https://atlas.ripe.net/probes/2317/, and click the "edit" button under "Notifications". > > Kind regards, > > RIPE Atlas Team

4 6

probe selection criteria on "WW" measurements
by Paul Vlaar 25 Jul '14

25 Jul '14

Hi, I have two questions regarding probe selection for UDMs. When I specify for example that I want to run an ongoing measurement on 250 probes in the "WW" region, will these probes remain the same throughout the lifecycle of the measurement (unlimited) or will these be somehow randomized every time the measurement is run? Secondly, I am wondering about even spread of probe location in doing such a measurement. For instance I am as much interested in results from regions that have only a few probes, as I am interested in results from those regions that are better served. When I look at the results so far of a test measurement, it appears that I'm receiving results from 35 probes in US, but only 3 from FI and none from NO. The latter surprises me a bit since there are 100 probes reported in NO, so why aren't there any results from those coming back? While I understand that countries with more probes will likely yield more results, I am wondering if there is any method to get a better "country-balanced" result for measurements ran in "WW". Thanks, ~paul -- Paul Vlaar DNS Infrastructure Group Afilias e-mail: pvlaar(a)afilias.info phone: +1-416-673-4078 mobile: +31-6-506-306-35 mobile USA: +1-602-410-4148

5 5

Support RIPE Atlas with RIPE NCC Board
by Daniel Karrenberg 25 Jul '14

25 Jul '14

The RIPE NCC Executive Board is discussing RIPE Atlas as documented in the latest meeting minutes that have been published recently. https://www.ripe.net/lir-services/ncc/executive-board/minutes/2014/minutes-… The relevant part is: --------- 3.3.1 ATLAS EB286 01/04/2014 AP The Board asked Axel to produce a break down of the costs for ATLAS over a period of  3 years and an outlook beyond 2014 to see the development on OPEX and CAPEX in the coming years. Furthermore the Board asked to see a comparison between the current figures and a case where TTM and DNSMON services were continued in their current form in 2014 and beyond. Status 19/06/2014: DONE There was discussion about the value of ATLAS to the various stakeholders and the associated maintenance cost. After discussion the Board asked for clear documentation of the results of the activities so that they could review and ultimately justify the cost of the project overall. The Board discussed the ongoing cost of the ATLAS project. The Board agreed that the value of the project needed to be packaged correctly. ATLAS now has many other projects wrapped into it and the activity needed proper and concise documentation as well as a solid marketing plan moving forward. ACTION EB295: The Board asked Axel to produce a value proposition for the ATLAS service in preparation for the September meeting. --------- It looks to me that the board needs to be re-assured that RIPE Atlas is valuable enough for the membership to justify the cost. Between the lines I read basic support but a desire to hear a consistent story and success stories of benefit to the membership. Of course the RIPE NCC staff will produce an answer to the board and I will do my best to help with that. But ultimately the board is responsible to the membership. So it would be very very helpful if we could get some users of RIPE Atlas to speak up both directly to the board and to help us prepare our answer to the board. Daniel

3 2

New on RIPE Labs: Midsummer Update for RIPE Atlas
by Vesna Manojlovic 24 Jul '14

24 Jul '14

Dear colleagues, in these hot and slow summer days, we give you an overview of the current status of RIPE Atlas, new features since the last update during RIPE68, and an insight into past and future community activities: https://labs.ripe.net/Members/fatemah_mafi/ripe-atlas-midsummer-update-2014 Regards, Vesna -- Measurements Community Building team <MCB(a)ripe.net>

2 1

some thoughts and question regrding probe "stability"
by Wilfried Woeber 22 Jul '14

22 Jul '14

Hi Folks, triggered by the discussion related to DNSMON, and an issue (power, resolved) with one of my V1 probes, I'd like to get some input or start a disussion or an investigation. To start with, I am not very clear what the term "stability" w/should mean in this context, as the probes are supposed to buffer measurment data locally, at least for a while (true?). So, here goes... Obviously, looking at some Atlas Stat pages, there are probes with a 100% uptime. Now, looking a the 3 under my supervision (2x V1, 1 Anchor), ref "Connected" and "Disconnected", there's no chance to get near that value, as all of them tend to topple over on a regular basis, mostly for a *short* period of time in the range of 0m(!) to some 30+m. With respct to the bahaviour of the Anchor, which is mounted in the same rack as the backbone router it connects to, in a Data Center, we tried to correlate the (reported) disconnection events with the router and interface logs for the probe. No luck there, also, no maint works or the like, so I presume the Anchor didn't reboot or that there were "real" network problems. Let's compare the most recent dis/connection logs for my 3 pets: ID 6009 2014-07-14 03:58:03 3d 8h 16m Still Connected 2014-05-27 03:03:54 48d 0h 46m 2014-07-14 03:50:47 0h 7m 2014-05-20 15:19:02 6d 11h 37m 2014-05-27 02:57:00 0h 6m 2014-05-14 21:16:56 5d 17h 59m 2014-05-20 15:16:22 0h 2m 2014-04-08 16:03:21 36d 5h 1m 2014-05-14 21:05:17 0h 11m ID 0466 2014-07-13 23:31:05 3d 12h 45m Still Connected 2014-07-09 23:05:40 3d 23h 54m 2014-07-13 22:59:49 0h 31m 2014-06-16 10:53:21 23d 11h 55m 2014-07-09 22:49:04 0h 16m 2014-05-25 09:03:06 22d 1h 38m 2014-06-16 10:42:00 0h 11m 2014-05-24 20:34:50 11h 54m 2014-05-25 08:29:12 0h 33m ID 0414 2014-07-07 23:41:23 9d 12h 35m Still Connected 2014-07-02 03:58:45 5d 19h 31m 2014-07-07 23:29:54 0h 11m 2014-06-13 09:37:50 18d 18h 7m 2014-07-02 03:45:08 0h 13m 2014-06-08 13:22:14 4d 20h 7m 2014-06-13 09:29:38 0h 8m 2014-05-21 08:29:23 18d 4h 45m 2014-06-08 13:15:11 0h 7m Again, I fail to see some obvious correlation, what am I missing? Does anyone else see a similar pattern? How to start debugging, if there's anythig that needs debugging? Thanks for your ideas and help! Wilfried

9 13

Actual public measurements
by Brian Rak 22 Jul '14

22 Jul '14

Is there any way to link someone to a public measurement, without requiring that they go and sign up for a RIPE account? It's an annoying extra step when I'm just trying to show someone the results of a measurement. I'm aware I could grab that raw data via the API, but then I'd have to re-implement the UI... which I don't have time for.

3 2

Questions about paris traceroute
by Thomas Holterbach 18 Jul '14

18 Jul '14

Hi all, I would like to start some Paris-traceroute measurements from RIPE Atlas probes, with a specific frequency and a control over the paris id. In particular, I am wondering if it is possible to start a traceroute measurement with all the 16 paris id at the same time. Because so far, I only find out how to start periodically a traceroute with a different paris id. I am also a little bit confused about the interval between two measurements. In the doc, it is written that the default value (900s) is the minimum, but when I create a UDM, the minimum interval seems to be 60s. I also tried 10 secondes with Restful API and the measurement started, but the real interval seems to be far upper than 10s. Can you give me more informations about this interval ? Thank you for your help, Thomas.

5 6