Please create user "bridgestrap" on polyanthum

We are going to set up the tool bridgestrap on polyanthum and it would be desirable to do so using a separate user account.

mentioned in issue tpo/anti-censorship/bridgestrap#5 (closed)

mentioned in issue #40080 (closed)

added Sponsor 30 - FINISHED label

added Next label

assigned to @anarcat

added Doing label and removed Next label

the bridgestrap user has been created, and has "lingering" rights to run systemd services, like bridgescan.

let me know if you need anything else.

closed

To administer BridgeDB, I log in as phw and then run "sudo -u bridgestrap -s". When doing the same for the bridgestrap user, I'm told that "user phw is not allowed to execute '/bin/bash' as bridgestrap." Can you please fix that? Or suggest an alternative for setting up and taking care of the bridgestrap process?

reopened

you're totally right of course, sorry about that. should be fixed in puppet's 1362d2c4 and deployed on polyanthum already.

closed

I ran into one more issue: while configuring my systemd user script, I got the error Failed to connect to bus: No such file or directory after running systemctl --user daemon-reload. As you point out in your guide, I did first run export XDG_RUNTIME_DIR=/run/user/$(id -u) but the directory /run/user/2150 does not exist. Can you please help with that?

reopened

I ran into one more issue: while configuring my systemd user script, I got the error Failed to connect to bus: No such file or directory after running systemctl --user daemon-reload. As you point out in your guide, I did first run export XDG_RUNTIME_DIR=/run/user/$(id -u) but the directory /run/user/2150 does not exist. Can you please help with that?

did you try logging out and back in again?

Yes, I did. (I logged in by running sudo -u bridgestrap -s.)

uh! indeed, something is off here. i'll see what i can do.

can you try again? i reran the "enable-linger" magic bit and that seems to have started the process. not sure wtf is going on here.

i figured it out: the puppet configuration was incorrect and fixed (see #33449 (closed)). this should work again now. sorry for the confusion.

closed

mentioned in issue #33449 (closed)