Replies: 17 comments
-
Please give us exact reproducer. I would figure this is either SELinux or User Namespace issue. |
Beta Was this translation helpful? Give feedback.
-
I don't have SELinux, just usernamespace. Just start the tuxnvape/multicms container in and out a pod and watch the logs (userns rootless) |
Beta Was this translation helpful? Give feedback.
-
Please submit the Podman command. |
Beta Was this translation helpful? Give feedback.
-
Don't use a podman command, sent it with ansible
|
Beta Was this translation helpful? Give feedback.
-
Same test with httpd official image from docker hub AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using 10.0.2.100. Set the 'ServerName' directive globally to suppress this message |
Beta Was this translation helpful? Give feedback.
-
Strange Could you try those commands with podman. podman run --pod new:item.website_name docker.io/tuxnvape/multicms:latest And see if this blows up? BTW It worked fine on my Fedora 36 laptop. |
Beta Was this translation helpful? Give feedback.
-
It's working fine But why it fails when I manage it with ansible (it's RedHat product too). Where I can reprot the bug ? There is not other containers fails... |
Beta Was this translation helpful? Give feedback.
-
If it's an ansible module bug, this is a really strange bug |
Beta Was this translation helpful? Give feedback.
-
I guess I would need to see what actual Podman command ansible is executing, to unerstand what is going on. One problem might be that the tty for stderr under ansible might be very different, If you run the ansible script on a rootful podman, does it work? |
Beta Was this translation helpful? Give feedback.
-
BTW is their a debug mode in ansible to show the command being executed? |
Beta Was this translation helpful? Give feedback.
-
BTW If you are a Red Hat customer, you should be reporting this in a bugzilla so that we could get ansible people to look at it. |
Beta Was this translation helpful? Give feedback.
-
This is the debug `pod:
TASK [Run multicms container] ********************************************************************************************************
|
Beta Was this translation helpful? Give feedback.
-
Same bug in other distros, and in other architecture: arm64 |
Beta Was this translation helpful? Give feedback.
-
After debug if I use:
Containers fail like the ansible method |
Beta Was this translation helpful? Give feedback.
-
This is working for me on F36 |
Beta Was this translation helpful? Give feedback.
-
Okay, I solved the problem by myself with
and
|
Beta Was this translation helpful? Give feedback.
-
Hit this while working through #20573. using Unit file as follows, exactly as produced by
|
Beta Was this translation helpful? Give feedback.
-
Apache containers log error when starting in pod (not in container), userns rootless mode
AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 174.10.0.20. Set the 'ServerName' directive globally to suppress this message
(13)Permission denied: AH00091: apache2: could not open error log file /var/log/apache2/error.log.
AH00015: Unable to open logs
Tried to add group_add tty without success
podman version 4.1.1
Archlinux kernel 5.18.15-arch1-1 #1 SMP PREEMPT_DYNAMIC Fri, 29 Jul 2022 22:52:39 +0000 x86_64 GNU/Linux
Beta Was this translation helpful? Give feedback.
All reactions