New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Systemd 256 loginctl changes #20400
Systemd 256 loginctl changes #20400
Conversation
test/common/testlib.py
Outdated
raise | ||
|
||
# terminate all systemd user services for users who are not logged in | ||
self.machine.execute("systemctl stop user@*.service") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This change changes behaviour, as it used to stop non web console
sessions. Is that wanted?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In theory, loginctl kill-user
should do that as well, but we don't do that. Let's see what the tests say -- if it goes green, then ok. It may also only affect e.g. c-podman, but we can put it back if needed. This cleanup is unfortunately incredibly finnicky and brittle 😢
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah, you get errors like this from stale sessions 😢
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So hmm should we generalize this? Or add another for loop to iterate over "normal sessions" except where uid == 0
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmm, I just saw that we already do the loginctl kill-user
thing in terminate_sessions()
. So I don't know, perhaps reproduce one of the failures locally and see how you can clean up after e.g. the "anton" test?
test/common/testlib.py
Outdated
raise | ||
|
||
# terminate all systemd user services for users who are not logged in | ||
self.machine.execute("systemctl stop user@*.service") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In theory, loginctl kill-user
should do that as well, but we don't do that. Let's see what the tests say -- if it goes green, then ok. It may also only affect e.g. c-podman, but we can put it back if needed. This cleanup is unfortunately incredibly finnicky and brittle 😢
d53d53e
to
daa397c
Compare
In systemd 256 list-session now also includes a manager session: 4 1000 admin - 538 user web console no - 5 1000 admin - 544 manager - no -
daa397c
to
c40c9e6
Compare
This now broke fedora-rawhide on TF so we should fix this now :) and this also needs cockpit-project/bots#6385 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Would be nice to know what these "manager" sessions are.. but if this works, it works. :-)
Agreed, created systemd/systemd#32845 |
rawhide tests run again, so I consider this a success! |
No description provided.