[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Installing plymouth



On 1/30/19 3:13 PM, Bastian Blank wrote:
> On Wed, Jan 30, 2019 at 02:25:20PM +0100, Thomas Goirand wrote:
>> On 1/30/19 1:45 PM, Bastian Blank wrote:
>>> On Wed, Jan 30, 2019 at 12:47:01PM +0100, Thomas Goirand wrote:
>>>> To get systemd to log on all tty (ie: tty0 and ttyS0), plymouth is
>>>> needed. So, my proposal is to get plymouth installed by default on our
>>>> images, with the "nosplash" option on the kernel command line (so that
>>>> plymouth doesn't display a splash screen).
>>> I tried that and don't see any difference.  Please describe in more
>>> detail what the expected behaviour is.
>> If you put systemd.show_status=true in the kernel command line, and if
>> there's:
>> console=tty0 console=ttyS0
>> then systemd will only print to ttyS0 unless plymouth is installed
>> (though I am setting it up with "nosplash text" on the kernel command
>> line so it doesn't bother).
> 
> "systemd.show_status=true" modifies the status output, not the log.
> Also on is the default and only overriden by "quiet".

Which is why I don't have "quite" in my images.

> "nosplash" disables all plymouth services, see
> debian/patches/0005-cmdline.patch in the plymouth source.

Though with the package, systemd correctly outputs to both tty0 and ttyS0.

>>>> Any objection? Or would you guys want this only in the OpenStack image?
>>> Yes.  I don't see how it should do what you want.
>> And now?
> 
> Well, it would be nice if you tried to explain what you are trying to
> accomplish or wanted to see.

I already explained. It's probably my fault if you don't get it, but I
don't see how I can do it better without repeating myself. Which part
don't you understand?

> systemd got two output mechanisms: status and log.
> 
> log goes to the journal, but can also go via kmsg and the console.
> 
> status is only active on boot, before the gettys are active and the
> output is cleared.

Maybe reading this will help:
https://github.com/systemd/systemd/issues/3403

Though according to it, it's probably fixed in Buster.

Cheers,

Thomas Goirand (zigo)


Reply to: