init 5 from single user mode fails

jd1008 jd1008 at gmail.com
Sun Mar 20 16:20:30 UTC 2016



On 03/20/2016 10:14 AM, James Hogarth wrote:
>
>
> On 20 March 2016 at 16:08, jd1008 <jd1008 at gmail.com 
> <mailto:jd1008 at gmail.com>> wrote:
>
>
>
>     On 03/19/2016 08:37 PM, g wrote:
>
>
>         On 03/19/16 20:52, g wrote:
>
>             On 03/19/16 20:38, jd1008 wrote:
>             <>
>
>                 SO how can I debug this, or is this the normal behavior?
>                 If normal, since which release?
>
>             --
>
>             because you booted to level 1, not all that is needed for
>             xwindows
>             has been loaded.
>
>             if you init 3, note list of additional progs that are
>             initialize.
>
>             after your login, run startx or init 5, depending on which
>             state
>             you want.
>
>         --
>
>         and i just might be incorrect on that recall from 'chemo
>         brain' memory
>         because i did a quick run thru and did not have problem
>         issuing 'init 5'
>         or 'startx' as root user at level 1.
>
>         as for diff of what is initialized from 'level 1' and 'level
>         3', it
>         just may be appear that they appear diff.
>
>         my bad. please excuse.
>
>
>     Well, I am mostly concerned with finding out the why of it.
>     What is NOT BEING DONE or STARTED which prevents going to full
>     init level 5?
>
>
>
> So init 1 is synonymous with the rescue target in systemd terms (see 
> man systemd.special)
>
> If you look at what is available at that level it does not surprise me 
> that startx doesn't work.
>
> Carrying out init 5 is synonymous with systemctl isolate 
> graphical.target (the old runlevel 5 translating to the graphical 
> target in systemd terms)
>
> Related to this the old init 3 is roughly equivalent to systemctl 
> isolate multi-user.target
>
> I say roughly equal as targets and isolate is far more precise, 
> powerful and capable than the old init levels and the symlinks that 
> run to Kill or Start services so behaviour from days of yore may 
> differ slightly ...
>
> It would be useful if you are seeing issues transitioning from one 
> target to another to get the journal from the attempts to do so.
>
OK. So, when it happens, you would like the output of journalctl -xb?  
There are quite a options to journalctl,
and I am not sure which ones will reveal the info you want.
Pls. let me know.

Cheers,

JD


More information about the users mailing list