Bug#810018: procps: Please (re)consider shipping procps pidof

Gioele Barabucci gioele at debian.org
Tue Apr 28 08:39:13 BST 2026


Hi Craig,

On 28/04/26 08:58, Craig Small wrote:
> On Tue, 28 Apr 2026 at 06:51, Gioele Barabucci <gioele at svario.it 
> <mailto:gioele at svario.it>> wrote:
> 
>     while we try to reach consensus in #1131136 on the best way to remove
>     the Essential bit from `sysvinit-utils`, I think we can focus on a
>     simpler task: moving `pidof` from `sysvinit-utils` to `procps`.
> 
> I thought the Essential part had to happen before the switch the pidof part.
> Looking at this plan, it seems this part could be done first.

Removing Essential first was needed before we got rid of pidof in 
maintscripts. Now we can deal with that later.

>     * No maintainer script uses `pidof` anymore (the 4 low-popcon remaining
>     packages will be NMU'd soon; patches ready).
> 
> OK, when I checked a few months ago there were quite a lot.

The last "wave" of removals was three months ago. There are still 
occurrences of `pidof` in various maintscripts, but they are all guarded 
by `command -v` and similar constructs, so they turn into NOPs in case 
`pidof` is not available.

>     * Both implementations of `pidof` support these options, with the same
>     semantics/output.
> 
> Agreed, I think -z is the main difference now.  I'm not 100% if that 
> option is useful anymore.

And fortunately `-z` does not seem to be used by any Debian package. 
Nevertheless local scripts by our users may be impacted. We will need to 
write a NEWS entry and a release note warning about this change of 
implementation. (I forgot to put this in the plan.)

>     The plan:
> 
>   That seems ok to me (procps maintainer).

Thanks!

>     Does this sound right? Is anything missing?
> 
> It does sound right, but I have a nagging feeling something is missed.

Me too. :) Any transition like this is bound to break something. I think 
all the preparatory work that has been done will greatly reduce both the 
chances of this happening as well as the blast radius.

Regards,

-- 
Gioele Barabucci




More information about the Debian-init-diversity mailing list