- ripe-atlas - mailman.ripe.net

Another probe disconnected
by Ernst J. Oud 23 Aug '24

23 Aug '24

Hi, And just now probe 1004467 also has been disconnected. What is going on? Regards, Ernst J. Oud

6 9

Re: Auto reconnect?
by Ernst J. Oud 23 Aug '24

23 Aug '24

Ok. Hope so. But it is weird that probes try to connect to a controller that clearly is not responding. I have six probes, 5 work fine, 1 is now disconnected for 34 hours. If it would connect to the controller it always used to connect to - ctr-nue14- all would be well but it appears it persists to connect to a non functioning controller, probably ctr-dub-hw01. Or something else is going awry in fail over between controllers perhaps? I know this is a free non-critical service but the lack of communication here wonders me a bit. Especially since the downtime is longer than ever before. Ernst > On 23 Aug 2024, at 17:56, Nick Hilliard (INEX) <nick(a)inex.ie> wrote: > > Ernst J. Oud wrote on 23/08/2024 16:54: >> Do we need to take actions to have disconnected probes reconnect “over the weekend”? Or is a reconnect in this case automatically? > probes automatically reconnect, as far as I'm aware. > > Nick

1 0

Auto reconnect?
by Ernst J. Oud 23 Aug '24

23 Aug '24

I asked before but this list appears to be not for support questions? Try again… Do we need to take actions to have disconnected probes reconnect “over the weekend”? Or is a reconnect in this case automatically? Regards, Ernst J. Oud

1 0

RIPE NCC Atlas probes are going offline: `controller INIT exit with error`
by Marek Küthe 22 Aug '24

22 Aug '24

Hello, I am running two RIPE NCC Atlas Probes [1] [2] and both went offline a few hours after each other and a `controller INIT exit with error` appears in the log [3]. What can I do about this? The controller also seems to be somehow not pingable [4]. I use Debian 12 with the self-compiled .deb package (version 5090) and use a slightly hardened systemd unit [5]. [1] https://atlas.ripe.net/probes/1008669/ [2] https://atlas.ripe.net/probes/1008711/ [3] https://paste.i2pd.xyz/?2c936299e0b7473f#G9caVaD8bxLiDvSmHF4HEsZAEjMrW7Gchv… / https://paste.burble.com/?cecb8def804b373d#GyNsT22QBmii4PGZFnRnVP6mLQC9vhsv… [4] https://i.ping.pe/o/L/img_oL3R8bjI.png [5] https://codeberg.org/mark22k/dn42-router/src/commit/5c855adbbca45263268ba5d… I would appreciate an answer! -- Marek Küthe m.k(a)mk16.de er/ihm he/him

3 2

Also probe 60200 is disconnected (4G-T-Mobile)
by Ernst J. Oud 22 Aug '24

22 Aug '24

Hi, Yesterday (SW) probe 1005072 disconnected and this morning (HW) probe 60200 disconnected. This also happened on July 18th with exactly these same two probes, then it apparently was a problem in the Atlas infra. Please advice. Regards, Ernst J. Oud Begin forwarded message: > From: RIPE Atlas <no-reply(a)ripe.net> > Date: 22 August 2024 at 08:30:02 CEST > To: ernstoud(a)euronet.nl > Subject: Probe 60200 is disconnected (4G-T-Mobile) > Reply-To: RIPE Atlas <no-reply(a)ripe.net> > > Dear Ernst J. Oud, > > Your probe 60200 has been disconnected from the RIPE Atlas infrastructure since 2024-08-22 05:52:00 UTC. > > You may be able to learn more about the possible cause from the "Status" tab on your probe's information page: > > https://atlas.ripe.net/probes/60200/#!tab-status > > Probes are tagged with suspected problems every four hours, so if there is nothing listed under the Status tab and your probe has just gone offline, please check back in a few hours, as more information may be available. > > If you need further assistance or have any questions, please contact us at atlas(a)ripe.net. > > Kind regards, > > RIPE Atlas Team > > ********** > 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/60200/, and click the "edit" button under "Notifications".

1 0

RIPE NCC Atlas probes are going offline: `controller INIT exit with error`
by Marek Küthe 21 Aug '24

21 Aug '24

Hello, I am running two RIPE NCC Atlas Probes [1] [2] and both went offline a few hours after each other and a `controller INIT exit with error` appears in the log. What can I do about this? The controller also seems to be somehow not pingable [3]. I use Debian 12 with the self-compiled .deb package (version 5090) and use a slightly hardened systemd unit [4]. [1] https://atlas.ripe.net/probes/1008669/ [2] https://atlas.ripe.net/probes/1008711/ [3] https://i.ping.pe/o/L/img_oL3R8bjI.png [4] https://codeberg.org/mark22k/dn42-router/src/commit/5c855adbbca45263268ba5d… I would appreciate an answer! -- Marek Küthe m.k(a)mk16.de er/ihm he/him

1 0

software probe DNS resolving not via OS functions
by Marco Moock 21 Aug '24

21 Aug '24

Hello! I have a software probe and it does many DNS requests. I installed the nscd cache on my machine, but the probe itself doesn't use the operating system's libc functions to resolve names. Is there a special reason for that? I am aware I can install a real local caching DNS server on the machine like dnsmasq, but I would like to ask the question. :-) -- kind regards Marco

2 1

Some RIPE Atlas calls take a long time
by Ben Cartwright-Cox 20 Aug '24

20 Aug '24

Hey folks, Anyone else getting issues with some RIPE Atlas API calls taking a very long time for calls that seem like they should be simple, like a credit check? # time curl https://atlas.ripe.net/api/v2/credits/ -H 'Authorization: Key xxx-xxxx' -o /dev/null real 0m6.394s user 0m0.063s sys 0m0.005s

2 1

RIPE Atlas Infra IP Rages
by Milad Afshari 20 Aug '24

20 Aug '24

Dear Atlas Team, I hope this message finds you well. Could you please provide the exact IP range(s) used by the RIPE Atlas infrastructure? This information is crucial for our ongoing network configuration. Thank you for your assistance. Best regards, Milad

6 7

Atlas UI update 2024-08-19
by Stephen Suess 19 Aug '24

19 Aug '24

Hello Atlas community, We have just released an update to the UI that includes (among other things): - NEW: Advanced search for probes and measurement listing pages - CHANGE: Header components with new icons to match ripe.net - CHANGE: Anchors cache updated on demand (instead of 24 hours) - CHANGE: Improved debounce, reduced error messaging on measurement form If you want to see the full list of what has changed, you can check the release notes here: https://atlas.ripe.net/docs/releases/website/2024-08-19.html As always, we hope you find these changes useful and we welcome your feedback. Kind regards, Stephen Atlas UI

1 0

Strange behavior on UI
by Victor TALVAZ 09 Aug '24

09 Aug '24

Hi, I'm not sure but there is something incorrect on the UI; In Network tab => Connection History in the website, I have the feeling that the column Disconnected for/Connected for has been inverted. For instance, I was connected for 20hours, and now the UI displays that I was Disconnected for 20hours. Dunno if you see the same issue. Regards, Victor

2 5

[atlas]Recent UI changes
by Stephen Suess 08 Aug '24

08 Aug '24

Hello Atlas Users, We have released several updates to the UI over the past month. Here are some of the highlights: - Measurement overview tab: improved summary info for various measurement types and (in the case of dns) most observed answers filtering - Measurement results tab: multiple, non-modal, resizable traceroute popups, toggles for filtering of no reply and no report items, and better error messaging when there are multiple sub queries If you wish to consult an exhaustive list of changes, you can see everything at https://atlas.ripe.net/docs/releases/website.html We hope you find these changes useful and welcome any feedback or suggestions you have. Kind regards, Stephen Atlas UI

2 5

Re: v1 probe loses connection after a while (and webui needs more love)
by Sulev-Madis Silber 07 Aug '24

07 Aug '24

i meant there were recent updates to ripe atlas i noticed the recent power outage is separate issue but i noticed problems appearing at same time. as it is always powered, i'm suspecting it now had to reconnect or maybe it also reflashes new updates on boot, so now it's different. but i don't know what's happening here. hence the question!

3 4

Cannot download periodic measurements
by Shihan Lin 07 Aug '24

07 Aug '24

Hi all, I recently tried to download some public periodic measurements that I launched in May, but I find I cannot download the full result by API or by web interface. One is example is measurement 72054485, which is launched on May 20th. I transformed the start time and end time to timestamp, which is 1716250200 (2024-05-21 00:10) and 1716336540 (2024-05-22 00:09), and when I used the following API to download the results, it responds me with empty file. https://atlas.ripe.net/api/v2/measurements/72054485/results?start=171625020… I cannot download it either when I use the web interface. Any idea about how to solve it? Thank you! Best, Shihan

2 3

doctoral/postdoctoral scholarships at HPI, Germany
by Vaibhav Bajpai 06 Aug '24

06 Aug '24

Dear colleagues, Here (below) is an opportunity to apply for a doctoral/postdoctoral scholarship at the Hasso Plattner Institute in Potsdam, Germany. Deadline: 15.8.2024 At our chair, we do systems research, and publish papers at the intersection of Internet measurements, security and privacy - that are directly relevant for the network operations community. Folks interested in working at our chair can indicate so in the application. Kind regards, — Vaibhav ------------------------ Prof. Dr. Vaibhav Bajpai, www.vaibhavbajpai.com ----------- Hasso Plattner Institute in Potsdam, Germany "Data Science and Engineering" https://hpi.de/en/research/cooperations-partners/research-schools.html The Hasso Plattner Institute (HPI) at the University of Potsdam is one of Germany’s top-ranked computer science institutes. Its PhD school "Data Science and Engineering" currenty supports 20 PhD students. Annually, we seek to add talented young researchers and offer 10 full Ph.D. and Postdoc Scholarships for a support of up to four years in our Potsdam location, on the border of Berlin. Students and Postdocs enjoy a great research environment, close mentorship by professors and postdocs, interaction with many peers, enormous computing power with HPI’s Data Lab, and significant travel funding. HPI faculty and their research groups ensure high quality research and will supervise you. Applications must be submitted at https://hpi.de/en/research/cooperations-partners/research-schools/scholarsh… by August 15, 2024. Please include the following documents: - Curriculum vitae and copies of certificates/transcripts - Brief research proposal, ideally mentioning the suitable HPI research groups you would like to work with - Work samples/copies of relevant scientific work (e.g., master's thesis or scientific papers) - Letter(s) of recommendation and any other supporting documents More information about the program and the application process is at https://hpi.de/en/research/cooperations-partners/research-schools/scholarsh… -----------

1 0

Lot of disconnects for Probe #27262
by Victor TALVAZ 05 Aug '24

05 Aug '24

Hi, Since some days i suffer for highly disconnects, connections to RIPE Atlas are disconnected every 10 minutes, or 50 but nothing more. I tried to change USB stick, and tried to put it on another switch/router. Seems the same. I have tools to monitor my FTTH line from another place, and it looks stable on both IPv4/IPv6. I see the disconnects but there is no alert sent in SOS History and the Probe doesn't reboot. I wonder what more I can do to see the issue ? Can it be an issue on Atlas side ? Maybe somebody from the RIPE see more infos that what I see. PS: I can see dates on connect/disconnect that change sometimes, it's really weird. Thanks for your help, Victor Talvaz

1 2

Re: v1 probe loses connection after a while (and webui needs more love)
by Sulev-Madis Silber 04 Aug '24

04 Aug '24

and, it also seems to try on v6, gets reply but doesn't seem to get connection up for some reason

1 0

v1 probe loses connection after a while (and webui needs more love)
by Sulev-Madis Silber 04 Aug '24

04 Aug '24

i have a old probe here. it's v1 or maybe it was v2. anyway... there were recent updates to system and i had to take everything off power for a few hours. after i put i back and it really is always on here, despite being home. the probe didn't reconnect. i don't have it on pdu. i have never needed it to cycle it before in previous 11 years. taking power off fixed it. now it's down again while on power and network. i went to looked what my atlas vlan does and found probe having both v4 and v6. and looks like it periodically tries to reconnect to ssh, gets reply and then... what happens next? it reports fw issue and trying to reconnect. but what kind of issue suddenly appeared when i have done none of fw changes. it's connection is via fbsd box i have full control over. can somebody check wtf is it and maybe fix it if needed. or maybe it just is too old? it works tho. i'm amazed how long it has been running. before next power cycle, can somebody diagnose it. it seems to connect out well? but just v4 i see. v4 is natted there while v6 is native and routed directly. probe id is 4415 p.s.: new webui tiles are really easy to accidentally mess up with finger on phone screen. never know when you scroll page and when you drag and drop. it's annoying

1 0

Probe #7030 may have specified the wrong location
by Marek Küthe 04 Aug '24

04 Aug '24

Hello, I noticed the probe #7030 [1] during the measurement #76522525 [2]. It has China as its location. However, the probe itself probably belongs to AS23961 [3]. You can see their locations on their website [4]. However, the probe does not appear to be in China, but in Hong Kong. Hong Kong is listed as an extra country/region in the RIPE NCC Atlas and I think this should be used accordingly so as not to falsify results. Could someone adjust the location? Regards Marek Küthe [1] https://atlas.ripe.net/probes/7030 [2] https://atlas.ripe.net/measurements/76522525 [3] https://bgp.tools/as/23961 [4] https://www.misaka.io/services/mc2 -- Marek Küthe m.k(a)mk16.de er/ihm he/him

1 0

REST API: Optional Fields for DNS Measurements
by Richter, Robert 03 Aug '24

03 Aug '24

Dear fellow RIPE Atlases, I am looking for the full definition of all optional field of the RIPE Atlas REST API for creating measurements. Especially the one for creating a DNS measurement. I only find the one defining the required attributes for a DNS measurement. But is there also a list constituting the optional fields? Best regards, Robert

2 1

RIPE Atlas data backend maintenance on 29 July 2024
by Francesco Iannuzzelli 29 Jul '24

29 Jul '24

Dear RIPE Atlas users, On 29 July 2024, between 9:00-18:00 (CEST), we will carry out planned maintenance on the data storage backend of RIPE Atlas. During this time, RIPE Atlas will be available but will serve stale data until work will be completed. Results for existing measurements will still be collected and stored. Measurement scheduling will be available but new measurements will be executed after the maintenance window. Results streaming will continue unaffected. We apologise for the inconvenience this may cause. Best regards, Francesco RIPE Atlas Team

1 1

[atlas]Installing Software Probe
by Milad Afshari 22 Jul '24

22 Jul '24

Dear RIPE Atlas Team, I encountered the following error while trying to download the RIPE Atlas software probe from the RIPE NCC GitHub repository: fatal: remote error: upload-pack: not our ref d72872cc6bb9ed5da47dcf1dc726fd8a36c6058f fatal: Fetched in submodule path 'probe-busybox', but it did not contain d72872cc6bb9ed5da47dcf1dc726fd8a36c6058f. Direct fetching of that commit failed. Can you please assist me with this issue? Kind regards, Milad Afshari

3 4

[atlas]API Not Working
by MARC BOSCH MARTINEZ 21 Jul '24

21 Jul '24

Hi all! Is anyone having problems issuing SSL measurements from the API? Seems that issuing them via the website works fine, but I've been having issues lately with the ones that I created using the API, leading to measurements in which all or most probes state "No Report Available". Best, Marc

2 1

[atlas]results page empty
by Marco Moock 16 Jul '24

16 Jul '24

Hello! https://atlas.ripe.net/measurements/75882685/results If I try to access the results, the page is empty. I tried Pale Moon and MS Edge. In Tor Browser, I can access it. Is something faulty at my side or is that a general problem? -- kind regards Marco

3 2

[atlas]Can not target google.com with periodic measurement
by Malte Tashiro 15 Jul '24

15 Jul '24

Hi, there seems to be something wrong (?) with the measurement creation process, or maybe some rules have changed? I'm trying to schedule a toy measurement to google.com, however, as soon as I set the timing to periodic, I get the error message: "We do not allow more than 25 concurrent measurements to the same target: google.com" I am not running any measurements at the moment. This behavior is independent of the measurement type (ping, traceroute, SSL) and the number of probes (error appears with one probe requested). It seems to solely be a problem of the timing. So far this message only appears for google.com and 8.8.8.8, other targets work. If I schedule the measurement in the future, I can actually submit it, but get a "Scheduling denied" measurement, for example: https://atlas.ripe.net/measurements/75418006/details I don't know if this is related, but it seems like somebody is hammering Atlas with a large number of geolocation measurements (500 probes per measurement) and DNS lookups (1 probe) at the moment. In the last hour 5582 measurements were scheduled, which is 1.5 measurements per second! Best, Malte

5 7