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
Work bsc#1103199 and bsc#1112109 around #6027
Conversation
We have a bug, which interferes with usage of serial console. On qemu backend it's enough to wait for welcome message on tty, which means that service is running. On Xen HVM service gets restarted constantly, so the only possible workaround is to stop it ASAP. See [poo#42359](https://progress.opensuse.org/issues/42359).
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.
I wonder if as a follow-up to this, we could simply do similar things for other big modules exporting a lot of things... :)
@@ -456,6 +457,8 @@ sub wait_boot { | |||
|
|||
# on s390x svirt encryption is unlocked with workaround_type_encrypted_passphrase before here | |||
unlock_if_encrypted if !get_var('S390_ZKVM'); | |||
# On remote backends we sync on Welcome message already where possible, mitigating bsc#1112109 | |||
wait_serial(get_login_message()) unless is_remote_backend; |
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.
Oh, so this is why we have in so many tests this failing step. The return value is ignored. Does this wait_serial actually work in your local tests?
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.
Now I have actually noticed that it failed on ppc64. I will create new PR to remove this one and addressing you comment below, as reverting export changes will create many conflicts.
@@ -57,8 +61,6 @@ sub run { | |||
assert_script_run 'less /var/log/YaST2/y2log*|grep "Automatic DHCP configuration not started - an interface is already configured"'; | |||
} | |||
|
|||
# Stop serial-getty on serial console to avoid serial output pollution with login prompt | |||
systemctl "stop serial-getty\@$testapi::serialdev"; | |||
# Mask if is qemu backend as use serial in remote installations e.g. during reboot | |||
systemctl "mask serial-getty\@$testapi::serialdev" if check_var('BACKEND', 'qemu'); |
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.
Why not also move this call?
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.
ping
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.
It's already done: 9c0cca4
Regression introduced in os-autoinst#6027, which doesn't work in many scenarios. Also, addressing second comment from the initial PR. See [poo#42359](https://progress.opensuse.org/issues/42359).
We have a bug, which interferes with usage of serial console. On qemu
backend it's enough to wait for welcome message on tty, which means that
service is running. On Xen HVM service gets restarted constantly, so the
only possible workaround is to stop it ASAP.
See poo#42359.
Verification runs
default ppc64le
default 64bit
opensuse gnome