-
Notifications
You must be signed in to change notification settings - Fork 1.1k
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
lxc-console broken for systemd containers under hosts without systemd #1704
Comments
Did you recently upgrade your lxc version? |
I think it was the latest update that broke things. Centos / Ubuntu containers were working on May 1st - 3rd while I was using them to test some scripts
|
looking at the PR's the containers were working on 7th June too |
Can you show me the containers config file, please? |
__________________________________________________-
Alpine Linux containers on the same I use a |
Can you show what happens in the terminal when you do |
Ok, can you ease start a container that has this problem with |
|
for comparison these are similar logs from an Alpine Linux container that starts ok: |
Cheers, not that I'm currently in the middle of a big chunk of work in LXD so this will have to sit for a bit. Sorry about that. I'll definitely come back to this! |
ok no problem |
Facing the same issue |
@itoffshore, @Chetan21, can you try and reproduce this behavior with |
NVM.. I could not reproduce. But, in general, there seems to be a lot of problems with CentOS LXC containers. Switched to debian containers, works well! |
Steps followed for that error, were exact as mentioned by @itoffshore in previous comments on updated |
@Chetan21, but you were unable to reproduce? |
I still have the same errors with |
Right, I think the explanation is rather simple. Can you show me the full process tree for one of those containers that don't allow for proper |
I'm able to
|
That looks like you're not using Can you please also show:
from the host. |
|
For the systemd container the
and start the systemd container again. |
Yes that fixed it - thanks ever so much for looking at this issue. Should this functionality be part of |
liblxc doesn't care what runs inside the container at all so it won't know whether it needs to create the cgroup in question or not. openrc will likely not want to have a |
* systemd based containers require the /sys/fs/cgroup/systemd cgroup for lxc-console and services inside the containers to work * systemd_ctr() based on recommendations from lxc: lxc/lxc#1704 (comment) Committer note: I did few minor modifications.
Having the same issue with systemd cgroup present
|
The $PATH seams okay from the lxc-console
|
Same "group error" output with the prebuilt Ubuntu image using the download template. Systemd started normally on the Ubuntu container and nothing seams broken but yet, interfacing with the containers from python seams fine. I cant figure out why the group ID 19 cannot be found or need
|
What happens if you do sudo lxc-attach --clear-env -P /home/madara/containers -n ubuntu |
@brauner same "group error"
|
The warning about the gid is not really an issue. I might send a small fix though once I've convinced myself that this is the right thing to do. :) |
Closes lxc#1704. Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Closes #1704. Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Closes lxc#1704. Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Closes lxc#1704. Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Distributions: Alpine Linux (OpenRC) / Manjaro (OpenRC) / Antix Linux (Sysvinit)
Distribution versions: Alpine 3.6 / Manjaro 17.0.2 / Antix 17-b1
lxc-start --version
=2.0.8
Issue description
lxc
containers under nonsystemd
hosts were working until about 4 weeks ago. At the momentlxc-console
connects to atty
but does not give a login prompt. Containers are started ok withlxc-start
but cannot be stopped withlxc-stop
. Usinglxc-attach
to connect to a container shows an error of:groups: cannot find name for group ID 19
(this is thelog
group on thelxc
host).Inside the container
$PATH
does not seem to work (calling binaries stopped working) &openssh-server
has stopped working in the containers. (I have been using ubuntu / centos containers created withlxc-create -t download
).Steps to reproduce
lxc-create -t download -n <container>
lxc-start -n <container>
lxc-console -n <container>
Information to attach
lxc-start -n <c> -l <log> -o DEBUG
- this gives a337meg
core dumplxc-start -n ubuntu2 -l trace -o ubuntu.lxc.log
:The text was updated successfully, but these errors were encountered: