VM connection issue

Gautam · August 9, 2025, 5:21pm

Hi ERC team,
We added a few new PCs to our husarnet group (and removed previous ones) and are currently unable to ping or ssh to the VM on all but one of them. We request you look into this at the earliest.
Thanks,
IITB Mars Rover Team

robocon_ekam · August 10, 2025, 4:25pm

Hi,

We are facing the same issue and are not able to test any communication nodes or the new ros2router. Also the Husarnet Dashboard Not Working issue still persists.
Moreover, we are unable to connect new computers to the husarnet group.

Team Robocon IITR

pidpawel · August 11, 2025, 11:37am

Does the issue still persist? I’ve just run our typical test suite and everything seems nominal (there may have been some leftovers from the Dashboard outage which sould have resolvd themselves after about 10 minutes).

Couple of remarks just in case though:

please update the Husarnet on your machines, if you’re using the nightly builds - we’ve fixed some bugs at the end of last week
same goes for ros2router - some time ago we updated it’s docker image and at the end of the week we’ve published the updated tutorial (Ros2router short tutorial - #12 by pidpawel), alongside with updated scripts erc2025/ansible/roles/teams/files/ros2router at main · husarion/erc2025 · GitHub

Gautam · August 11, 2025, 12:33pm

Hi,
After updating the nightly build, husarion-ops is now peer-to-peer and can be pinged while the VM is still unreachable. This issue is on 3 of the 4 devices we have in the group.

pidpawel · August 11, 2025, 2:41pm

Are all of these machines on the same Internet connection? Do they have same permissions (especially in the network)? Are they using the same operating system? What are the differences between them?

Gautam · August 11, 2025, 3:53pm

The main issue was that on running husarnet status on the VM the new devices weren’t getting updated. This has been fixed now and the new devices can ping and ssh successfully. Though we do not know exactly what triggered this, these are the specifications of the PCs - 2 running on Ubuntu 22.04.5 LTS (Jammy Jellyfish) and 2 running Ubuntu 24.04.3 LTS (Noble Numbat). We did try using different internet connections (using the same ones that worked during previous tests). All run on the nightly build.
On our end (the PCs), all changes in the dashboard were reflected within a minute or two but didn’t on the VM until about half an hour ago, the VM was stuck in a state that was around 2 weeks ago.

pidpawel · August 11, 2025, 4:05pm

Thank you for the report. We will have a look into that. Glad that it’s working now.