- ripe-atlas - mailman.ripe.net

IPv4 Built-ins fail while IPv6 work
by RIPE 26 Jun '24

26 Jun '24

Hi, My old hardware probe seems to have finally given up. So I installed a soft probe on my openwrt edgerouter-X, using the openwrt opkg packages. That seems to work except for all the IPv4 built-ins. I have a hurricane.net tunnel as my ISP ( Odido) is stuck in the past and does not provide IPv6. The IPv6 built-ins all show data but the IPv4 ones show ‘No data available for the period’ From the shell I can ping the addresses used by the built-ins, the names also resolve. Any idea what the cause could be ? Or how to find the cause. Probe ID = 1008458 Running OpenWrt 23.05.3, r23809-234f1a2efa Package: atlas-sw-probe Version: 5080-2 Regards, Paul

2 4

Atlas UI updated
by Stephen Suess 25 Jun '24

25 Jun '24

Hello Atlas fans, We have just released an update to the UI that includes: - NEW: download transactions in credit history - NEW: added toggles to allow hiding no reply/no report results in measurements details results - CHANGE: rework sorting in measurements results details (min to max rtt, no reply, no report) - CHANGE: map markers now sort from highest RTT on bottom to lowest RTT on top - CHANGE: added summary detail for RTT to DNS measurement detail overview 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/2024-06-25.html As always, we hope you find these changes useful and we welcome your feedback. Kind regards, Stephen Atlas UI

1 0

multiple probes down since Jun 24th?
by Andreas S. Kerber 25 Jun '24

25 Jun '24

Hi, our probe is reported down since yesterday: https://atlas.ripe.net/probes/7054/ Since I couldn't find a problem on our side and rebooting didn't change the status, I checked other probes (not mine) and they are reported down since the exact downtime as mine. Is there any kind of known problem at the moment? Other probes with the same downtime: https://atlas.ripe.net/probes/7040 https://atlas.ripe.net/probes/7052 https://atlas.ripe.net/probes/7063 https://atlas.ripe.net/probes/7067 https://atlas.ripe.net/probes/7068 -- Andreas Kerber (Senior IT-Engineer/Consultant) ask(a)idkom.de _______________________________________ IDKOM Networks GmbH Dieselstraße 1 * D-87437 Kempten (Allgäu) Tel.: 0831 59090-200 Fax: 0831 59090-90 www.idkom.de Registergericht: Amtsgericht Kempten (HRB 6359) Geschäftsführer: Benjamin Mayer * Klaus Giehl Diese E-Mail und alle mitgesendeten Dateien sind vertraulich und ausschließlich für den Gebrauch durch den Empfänger bestimmt!

6 5

repetitive 2fa authentication on every login
by ripe＠brite.cz 24 Jun '24

24 Jun '24

Hi team, https://atlas.ripe.net/probes/mine requires new login upon every browser open. It even requires 2FA every time. Is it possible to keep the session active for maybe a month, or at least do not ask 2FA every single time? I would think as services like gmail.com, portal.azure.com, account.synology.com, etc., do not require 2fa on every browser close and reopen, it may not be as crucial to perform 2fa every time I need to check if my probes are online? It becomes a bit annoying, especially me using mobile browser. Please consider. Thank you. Jiri

3 2

How long is a year?
by Edward Lewis 19 Jun '24

19 Jun '24

On my probe’s status page, there is an “All Time” “Time Connected” of 8y 343d 15h 30m. How many minutes are in a “y”? Context: in trying to illustrate uptime (number of 9’s) I wanted to show how long it would take to achieve different number of 9’s based on my probe’s history. (It’s only at one ‘9’ - 97%.)

3 4

Software probe keeps going to disconnected
by vom513 19 Jun '24

19 Jun '24

Hello all, I have 2 software probes. One is on a datacenter VPS and the other at home (runs on Proxmox). The one at home keeps going to disconnected. The probe is Debian 12. Probe version is 5080-1. This machine has v6 as well - so I was thinking maybe something with temp/privacy addresses getting reaped but no, I only have one good old EUI-64 address on there. (Knock on wood) - no other VMs on my Proxmox machine have any issues like this. The only thing that looks remotely related (and this happened a bit before Nagios sent out an alert) is this: Jun 17 12:34:41 atlas-probe ATLAS[123749]: /usr/local/atlas/bin/ATLAS: 99: kill: No such process Jun 17 12:34:41 atlas-probe ATLAS[123749]: no ssh client matching . cleanup state files. for next restart Jun 17 12:34:41 atlas-probe ATLAS[123749]: RESULT 9006 done 1718642081 bc2411ca94c6 no reginit.vol start registration Jun 17 12:34:41 atlas-probe ATLAS[123749]: /var/atlas-probe/status/reginit.vol does not exist try new reg Jun 17 12:34:45 atlas-probe ATLAS[123749]: Ping works Jun 17 12:34:45 atlas-probe ATLAS[123749]: /usr/local/atlas/bin/ATLAS: 92: kill: No such process Jun 17 12:34:45 atlas-probe ATLAS[123749]: start reg Jun 17 12:34:45 atlas-probe ATLAS[144349]: ATLAS registration starting Jun 17 12:34:45 atlas-probe ATLAS[144349]: REREG_TIMER 1718416507 expired now is 1718642085 Jun 17 12:34:45 atlas-probe ATLAS[144349]: REREG_TIMER_EXPIRED go re register REREG_TIMER 1 , now is 1718642085 Jun 17 12:34:45 atlas-probe ATLAS[144349]: REGHOSTS reg03.atlas.ripe.net 193.0.19.246 2001:67c:2e8:11::c100:13f6 reg04.atlas.ripe.net 193.0.19.247 2001:67c:2e8:11::c100:13f7 Jun 17 12:34:45 atlas-probe ATLAS[144349]: ssh -p 443 atlas(a)reg03.atlas.ripe.net INIT Jun 17 12:34:48 atlas-probe ATLAS[144349]: Got good controller info Jun 17 12:34:48 atlas-probe ATLAS[144349]: check cached controller info from previous registration Jun 17 12:34:48 atlas-probe ATLAS[144349]: NO cached controller info. NO REMOTE port info Jun 17 12:34:48 atlas-probe ATLAS[144349]: Do a controller INIT Jun 17 12:34:48 atlas-probe ATLAS[144349]: Controller init -p 443 atlas(a)ctr01.atlas-prod.aws.ripe.net INIT I believe I should see an persistent ssh outbound ? The only ESTAB is me ssh’ing in, and a 443 to RIPE: root@atlas-probe:/home/vom# ss -planto | grep ESTAB ESTAB 0 0 192.168.69.111:22 192.168.64.21:65409 users:(("sshd",pid=145083,fd=4),("sshd",pid=145059,fd=4)) timer:(keepalive,115min,0) ESTAB 0 0 [2603:6011:6300:1f02:be24:11ff:feca:94c6]:37176 [2a13:27c0:1c:d01:cc61:f3a3:920d:cd88]:443 users:(("ssh",pid=144380,fd=3)) timer:(keepalive,103min,0) Let me know what else I can look at. I assume there is some debug I can crank up ? Just going to reboot this for now to get it happy and get Nagios to shut up. Thanks.

2 2

[ anchor] Quota on the usage of an anchor in measurement.
by Huseyn Gambarov 19 Jun '24

19 Jun '24

Hello. In my research, I am conducting multiple RTT measurements against different targets. During the measurement process, a single anchor stops getting scheduled for a requested measurement after some number of times or returns -1 as a result. I want to ask if there is a limit on how many times a single anchor can participate in a measurement. And what is its cooldown period if such a limit exists? With regards, Huseyn Gambarov Research Assistant PhD Student in Computer Science Case Western Reserve University

2 1

Atlas UI updated
by Stephen Suess 10 Jun '24

10 Jun '24

Hi Atlas Users, Along with the aforementioned removal of old measurement detail page links, we have just released an update to the UI that includes: - NEW: Pref for using local time (vs UTC) everywhere - NEW: Added local toggle to QuickLook for using probes close to user location - CHANGE: Add traceroute table to msm overview map marker popups - CHANGE: Make clearer the difference between no report (yet) vs no reply (from target) - CHANGE: Map now resizes on larger screens in measurement overview 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/2024-06-10.html As always, we hope you find these changes useful and we welcome your feedback. Kind regards, Stephen Atlas UI

1 0

Removing links to legacy measurement detail pages
by Stephen Suess 07 Jun '24

07 Jun '24

Dear Atlas Users, Now that we have been up and running with the new measurement detail pages for a few weeks, we plan to remove the banner and links to the legacy pages next week. Please let us know if you still have any need for the old detail pages and why, and we will do our best to address any concerns. Kind regards, Stephen Atlas UI

1 0

RIPE Atlas infrastructure maintenance on 11 June 2024
by Robert Kisteleki 07 Jun '24

07 Jun '24

Dear RIPE Atlas users, On 11 June 2024, between 8:30-11:30 (CEST) we will carry out maintenance of the central RIPE Atlas infrastructure. The UI, APIs, access to stored results and metadata, as well as measurement scheduling will be unavailable. During this time results will still be still collected and stored. We apologise for the inconvenience this may cause. Regards, Robert Kisteleki, for the RIPE Atlas team

1 0