| 1 |
On Thu, 2019-04-25 at 17:42 -0400, Michael Orlitzky wrote: |
| 2 |
> On 4/25/19 5:23 PM, Michał Górny wrote: |
| 3 |
> > > What's wrong? You only need the effect of tmpfiles_process() if you're |
| 4 |
> > > running systemd or OpenRC. If the user is running SysV-init and if the |
| 5 |
> > > package also installs a SysV-init script, then that init script is going |
| 6 |
> > > to have to create any temporary directories it needs, because SysV-init |
| 7 |
> > > doesn't know anything about tmpfiles. |
| 8 |
> > > |
| 9 |
> > |
| 10 |
> > This isn't about init scripts. This is about arbitrary directories that |
| 11 |
> > need to be created at runtime and potentially cleaned up or recreated |
| 12 |
> > after reboot, such as cache directories. |
| 13 |
> > |
| 14 |
> |
| 15 |
> It's very much about init scripts. We have at least five service |
| 16 |
> managers in the tree, |
| 17 |
> |
| 18 |
> * sys-apps/openrc |
| 19 |
> * sys-apps/systemd |
| 20 |
> * sys-apps/sysvinit |
| 21 |
> * sys-process/runit |
| 22 |
> * sys-process/daemontools(-encore) |
| 23 |
> |
| 24 |
> And the only reason we would need a transient directory created and/or |
| 25 |
> cleaned-up is because one of those service managers is going to start a |
| 26 |
> program that needs it. Two of them can use the tmpfiles mechanism, but |
| 27 |
> the others must handle it on their own: in particular, they don't need |
| 28 |
> tmpfiles_process() to do anything. |
| 29 |
> |
| 30 |
|
| 31 |
No. tmpfiles is also used for programs started directly by user, such |
| 32 |
as eix. |
| 33 |
|
| 34 |
-- |
| 35 |
Best regards, |
| 36 |
Michał Górny |
Archives