- ripe-atlas - mailman.ripe.net

Requested and participating probes
by Huseyn Gambarov 13 Dec '24

13 Dec '24

Hello. I am currently working on a project that heavily relies on Ripe Atlas Ping measurements. During the development, I noticed that sometimes when, let’s say, I request 3 probes, only 2 are participating in the measurement. In the Ripe Archive, I found that there was a similar discussion here - https://mailman.ripe.net/archives/list/ripe-atlas@ripe.net/thread/D6QZS4WC5… I have two questions, that I could not find answers to, and I would be very grateful if you could help me with them. First, in the mentioned discussion it is said that there are multiple reasons why fewer probes are participating and there is ongoing work to allow users to obtain information about specific conditions that occurred during the measurement creation. Can you tell, if there is a way to get that information via API? Second, do you know if there is a reasonable timeout to wait before a new attempt to reschedule the probe that failed to participate in the first measurement? Thank you for your help. With regards, Huseyn Gambarov.

2 1

Understanding how to use the `paris` parameter
by Bhaskar, Abhishek Vasisht 13 Dec '24

13 Dec '24

Hello y'all! I'm a PhD student from Georgia Tech and had a question about the "paris" field when performing traceroute measurements. The RIPE documents states that ```Paris - A number greater than 0 will indicate the number of variations to be used for a Paris traceroute. Setting the value to 0 makes the measurement use the standard traceroute``` But regardless of what Paris value I give to the traceroute measurement, I only get one path with a single paris_id. I would expect that if I gave a paris value of 10, I should see 10 potential variations of the path between A and B, correct? Please correct me if I'm wrong. Here is an example measurement https://atlas.ripe.net/measurements/84562046 Thanks! -- Abhishek Vasisht Bhaskar Student, PhD in Computer Science Georgia Institute of Technology https://abhishekvasishtb.github.io/

1 0

MTU tracepath in RIPE atlas
by Marco Moock 12 Dec '24

12 Dec '24

Hello! Maybe some people here know the Linux tracepath application. It tests the PMTU to a destination. Could such a feature be implemented in RIPE Atlas? -- kind regards Marco Send unsolicited bulk mail to 1734020167muell(a)stinkedores.dorfdsl.de

1 0

Join Us for the "Exploring the RIPE Atlas UI" Webinar on 17 December 2024
by Michela Galante 11 Dec '24

11 Dec '24

Dear all, We are excited to invite you to the upcoming Exploring the RIPE Atlas UI Webinar, scheduled for Tuesday, 17 December 2024. Join Stephen Suess for a 45-minute webinar showcasing the key features and tools of the RIPE Atlas UI. This interactive session is designed to enhance your understanding and usage of this powerful platform. By the end of this webinar, you will be able to: • Identify and utilise key RIPE Atlas UI features to efficiently find and analyse existing Internet measurements. • Evaluate various data visualisation and analysis tools to gain meaningful insights from measurement results. • Customise the RIPE Atlas UI to suit specific usage patterns and prioritise the information you need. For more information and to register, please visit: https://learning.ripe.net/w/courses/64-exploring-the-ripe-atlas-ui/305 Kind regards, Michela Galante Senior Community Development Officer RIPE NCC

1 0

Re: Transfer of Atlas anchor ownership
by Robert Kisteleki 03 Dec '24

03 Dec '24

Hi, Can you please see to this change request? Thanks, Robert On Thu, Nov 28, 2024 at 4:05 PM Luuk Hendriks <luuk(a)nlnetlabs.nl> wrote: > Hi Robert, > > > We'd like to transfer the ownership of the Atlas anchor that I installed > at the University of Twente many moons ago. It's attached to my personal > account and we could not find a way to transfer it via the web > interface. > > It's anchor ID 6113, currently under ripe(a)luukhendriks.eu > and should go to Roland's account, roland(a)nlnetlabs.nl . > > > If there's anything more you need from me, please let me know. > > > Cheers, > luuk >

1 0

Probe #2413 somehow failing
by ripe＠brite.cz 29 Nov '24

29 Nov '24

Hi team, please have a look at Probe #2413, it shows connected but it seems it is not providing any measurements results. I tried different cables, power sources, connected to different switch, and finally move it to completely different location. It is v1 so it does not have USB memory stick. Please advise. Thanks Jiri

7 8

Streaming result data missing probe information with "enrichProbes: true"
by Wouter de Vries 27 Nov '24

27 Nov '24

Hi folks, When using the Streaming API there is an option called "enrichProbes" which should cause probe metadata to be added to every result. That seems to have worked for a while, but it has stopped working. Example code (in python) to reproduce (it's the example from the docs, with just enrichProbes set to true): import json import websocket ws = websocket.WebSocket() ws.connect("wss://atlas-stream.ripe.net/stream/") params = {"streamType": "result", "msm": 1001, "enrichProbes": True} ws.send(json.dumps(["atlas_subscribe", params])) for data in ws: print(data) Example result: ["atlas_result",{"fw":5080,"mver":"2.6.2","lts":66,"dst_name":"193.0.14.129","af":4,"dst_addr":"193.0.14.129","src_addr":"192.168.1.20","proto":"ICMP","ttl":56,"size":32,"result":[{"rtt":21.447667},{"rtt":21.226292},{"rtt":21.253375}],"dup":0,"rcvd":3,"sent":3,"min":21.226292,"max":21.447667,"avg":21.3091113333,"msm_id":1001,"prb_id":50274,"timestamp":1731917452,"msm_name":"Ping","from":"85.107.81.62","type":"ping","step":240}] As you can see there is no probe metadata present. I'm fairly sure I'm doing this right and I think this worked previously. Is something broken in the backend? Best regards, Wouter

2 2

UI feature wish: "Daily Credits Balance" pie chart widget
by ripe＠nurfuerspam.de 26 Nov '24

26 Nov '24

Hi, I very much like/liked the "Daily Credits Balance" pie chart widget from the old UI dashboard (https://atlas.ripe.net/my/) It gives an instant overview of the credit revenue and expenditure stream, not only the grand total, but also the _share_ of individual probes and measurements in the revenue and expenditure streams. With the actual number being shown when hovering over an individual slice of the pie. And clicking on a slice would take one to the page of the respective probe or measurement (though it seems currently broken for measurements, possibly because of the ongoing restructuring). Is there a way to get it back in the new dashboard, refreshed obviously, and integrated into the new overall site layout? Thanks! Kind regards, R.

2 1

Probe flash drive corrupted
by Michael J. Oghia 20 Nov '24

20 Nov '24

Hi team, all: I just checked my dashboard, and one of my probes is offline. The error message says: "Your probe was automatically untagged as "system: Flash drive filesystem corrupted." Does anyone know how to fix it? Or does it require me to change the probes? If the latter is the case, how can I go about doing that? Kind regards, -Michael

2 4

Feature idea: new status "Probe (temporarily) closed for new UDMs"
by ripe＠nurfuerspam.de 20 Nov '24

20 Nov '24

Hi, Obviously, measurements are the purpose of being of the RIPE Atlas system. Thus, e.g., I take greate care to maximize probe uptime so measurements scheduled on them don't get disrupted by local events that are likely not representative of what most measurements are looking at. However, sometimes, temporary disruption may be known to come up, but is unavoidable, e.g., for maintenance of, or around the probe. Or a probe needs to be taken out of service "for good". In those cases, it is obviously hardly possible to avoid long-running measurements from being impacted, either by a short-term disruption, or by the removal of a probe altogether. And measurement owners obviously are aware of that, and need to take that into account in a way specific to each measurement's purpose. And it is obviously possible to remove, e.g., defunct probes from a running measurement and replace them with new ones. However, when a probe host knows _ahead of time_ that their probe will experience some event potentially disrupting measurements (like during planned maintenance activities), or will have to be decommissioned at a specific time in the future, how about if they could signal that condition to the Atlas infrastructure, and the infrastructure would at least not schedule _new_ measurements on such probes? E.g., by having a new probe "status" that can be explicitly set by the probe host to say, "this probe shall (perhaps just temporarily) not get any _new_ UDMs scheduled at this time (while the status is set)". Just a thought, not sure how much effort that would be to implement, and whether that would be worth the benefit of disrupting "fewer" measurements in case of foreseeable issues with a probe (or anchor), especially also upon an upcoming need to decommission one entirely. Thanks! Kind regards, R.

1 0