Skip to content
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.

Sign up for GitHub

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

Jump to bottom

--show-boot breaks the interactive shell #66

Closed
gportay opened this issue Sep 9, 2021 · 16 comments
Closed

--show-boot breaks the interactive shell #66

gportay opened this issue Sep 9, 2021 · 16 comments

Comments

@gportay
Copy link
Contributor

gportay commented Sep 9, 2021

Hello,

fakemachine --show-boot execs an interactive shell but most characters typed-in get eaten by another process.

Things work correctly if --show-boot is not set.

To reproduce it, simply run

$ fakemachine --show-boot

Welcome to Arch Linux!

(...)
[  OK  ] Started fakemachine runner.
Running /bin/bash using kvm backend
bash-5.1# 
(most of characters are eaten...)

tty appears to be /dev/console.

As you see, I am running this on Arch Linux. It may come from this; I wonder if Debian users reproduce the same issue.

Any idea?
@obbardc
Copy link
Member

obbardc commented Sep 27, 2022

The same thing happens in debos; see go-debos/debos#191

@obbardc
Copy link
Member

obbardc commented Sep 27, 2022

Also the console output seems to clear itself, seems like it's since we start Plymouth during the machine boot...

[  OK  ] Started Show Plymouth Boot Screen.
[  OK  ] Started Forward Password R…s to Plymouth Directory Watch.

@obbardc
Copy link
Member

obbardc commented Sep 27, 2022

Also, this seems to only affect KVM/QEMU backends. UML (fakemachine --show-boot --backend=uml) is unaffected.

@obbardc obbardc added this to the v0.0.3 milestone Oct 12, 2022
@obbardc obbardc modified the milestones: v0.0.4, v0.0.5 Jan 28, 2023
obbardc added a commit that referenced this issue Jul 26, 2023
@obbardc
qemu: Append nosplash to kernel arguments to stop plymouth from starting
e1f3788 
The plymouth service starts when running a qemu/kvm guest, plymouth clears
the serial console during boot which in turn causes the host's shell to be
cleared. Add nosplash to the kernel argument to stop plymouth from
starting during the machine's boot.

This parameter is already present in the uml backend.

Fixes: #66
Signed-off-by: Christopher Obbard <[email protected]>
@obbardc obbardc mentioned this issue Jul 26, 2023
Merged
obbardc added a commit that referenced this issue Jul 27, 2023
@obbardc
qemu: Append nosplash to kernel arguments to stop plymouth from starting
8dbae9d 
The plymouth service starts when running a qemu/kvm guest, plymouth clears
the serial console during boot which in turn causes the host's shell to be
cleared. Add nosplash to the kernel argument to stop plymouth from
starting during the machine's boot.

This parameter is already present in the uml backend.

Fixes: #66
Signed-off-by: Christopher Obbard <[email protected]>
@evelikov
Copy link

I was able to reproduce the issue without plymouth. If you'd like some output logs I'm happy to share next time I play with debos/fakemachine.

@evelikov
Copy link

When the issue was reported there was no plymouth in Arch, so odds are Gaël and I are hitting the same not plymouth related issue.

@obbardc
Copy link
Member

obbardc commented Jul 27, 2023

Okay, logs or reproduction method would be great. It doesn't happen in Debian. I will unlink the other PR from closing this issue.

@obbardc
Copy link
Member

obbardc commented Nov 8, 2023

@evelikov I am still waiting for some instructions on the above. From my testing #162 fixes the remaining shell issues with --show-boot on Debian.

@evelikov
Copy link

evelikov commented Nov 8, 2023

We very rarely use --show-boot, so this is the bottom of my go-debos todo list. There are bunch of blockers (as mentioned previously) and until those are sorted I won't be getting to this one.

@obbardc
Copy link
Member

obbardc commented Nov 8, 2023 •
edited
Loading

There are bunch of blockers (as mentioned previously) and until those are sorted I won't be getting to this one.

Could you point out what those blockers are? Can we move the discussion about that to a separate ticket under https://github.com/go-debos/debos/issues as I don't think it belongs as comments here?

@evelikov
Copy link

evelikov commented Nov 9, 2023

Right, I had them listed in the internal task as per your request a while back. Will cross check and copy/paste as time permits.

@obbardc
Copy link
Member

obbardc commented Nov 9, 2023

Right, I had them listed in the internal task as per your request a while back. Will cross check and copy/paste as time permits.

On a side note, I'm trying to move the Debos roadmap to GitHub in the clear. It will allow for community to understand and help with blockers.

@evelikov
Copy link

evelikov commented Dec 5, 2023

Here's a list go-debos/debos#456.

Can you please let us know how community can help? Thanks

@obbardc obbardc removed this from the v0.0.6 milestone Jan 17, 2024
@obbardc
Copy link
Member

obbardc commented Oct 30, 2024

Did anyone test this again under arch after the fix #162 was merged ?
cc @evelikov @gportay

I don't know how to use arch so I cannot test there.

@gportay
Copy link
Contributor Author

gportay commented Oct 30, 2024

Did anyone test this again under arch after the fix #162 was merged ?

I am not using debos anymore, I am sorry and it is so old for me to remember the exact issue.

@evelikov
Copy link

Not using debos any more either - migrated to mkosi instead.

@obbardc
Copy link
Member

obbardc commented Oct 30, 2024

Let's consider it closed, then. If others have issues we can reopen it.

@obbardc obbardc closed this as completed Oct 30, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@evelikov @obbardc @gportay