There is no need, to check again and again, if the directory exists.
Once you've created it (as described in post #4), everything is fine!
And all files /etc/inittab.d/*.tab will be included in INIT-process.

Of course not. But when this is run once per booting, I think I have the computing resources to do that. Putting rests of configuration procedures into /etc/rc.local and editing it as necessary for new installs has saved me a lot of work over the years. Not the least having to remember this kind of details  ;-)

Indeed it is.
Your doing.  8^D

Well ...
Here's the thing.
Please correct me if I am wrong.

Debian uses systemd and Devuan uses sysvinit.
The /etc/inittab.d directory is directly related/of use to sysvinit and whoever puts *.tab files in it.

Unless I am mistaken, I understand that it is of no use to systemd (or Debian) because it is sysvinit-ware™.
ie: absolutely irrelevant to the systemd package, thus to Debian.

See https://github.com/systemd/systemd/releases/tag/v260

That being the case, it stands to reason that no Debian package will have a need to create the /etc/inittab.d directory.
So under no circumstance will it be 'normal' for a Debian package to create it, simply because it is written for Debian, a systemd enabled distribution.

Odds are that if a package required the use of /etc/inittab.d, it would not be allowed in the Debian repositories.

Don't have much else to add save a 'thank you' for the interesting exchange of ideas.

Best,

A.

As long as what was done was the right thing, good going.
Congratulations. 8^D

I would not know how to use /etc/inittab.d so will not opine any further than this:

- the warning existed because sysvinit was looking for the /etc/inittab.d directory and did not find it.
- the (to me) obvious solution was to fix the installer to create it at install time; optional or not, sysvinit was looking for it.
- if there was some issue as how the *.tab files worked or had to be written, add the corresponding documentation.
- I am not too sure it is the right solution because the problem was not the warning, in itself a consequence.
- (to me) problem was the /etc/inittab.d not being where it should have been which generated the warning.

The use of /etc/inittab.d is a feature of sysvinit, maybe very few use it, maybe not.
The solution has only obscured its existence, I only learnt about it because of the warning and fsmithred's post.

As always, just my $0.02

Best,

A.

And lo, it was done.

Oh ye of little faith!

The message will be removed and documentation added in the next upstream release. Expect in Freia. :-)

[ -d /etc/inittab.d ] || mkdir /etc/inittab.d

There is no need, to check again and again, if the directory exists.
Once you've created it (as described in post #4), everything is fine!
And all files /etc/inittab.d/*.tab will be included in INIT-process.

Best regards, FM_81

Happened to me once and have seen it happen to others.
But things sometimes change.

And nothing is certain exept death and taxes.

Waytogo !   8^)
Valiant effort, thanks for that.

Let's see what happens.

Best,

A.

[ -d /etc/inittab.d ] || mkdir /etc/inittab.d

I use this mechanism for other 'non-default' directories that are useful or necessary for me.

I would not be so sure!

Watch this space: https://bugs.debian.org/1131137

Yes ...
But sysvinit asks for it (see link) so it should be there.

Hmm ...

Added via Debian?
Something that is directly related to sysvinit?   8^D !!!

Right ...
You could try filing a bug against INIT: no inittab.d directory found with the Debian devs.
But you will not like the reply.

TL;DR:
Something along these lines ...

"On your way, mate. This is not a bug, sysvinit is not part of the default Debian configuration".

Eventually and only for completeness' sake, someone with the required knowledge could volunteer to write a small metapackage to do that.

If not, just add it manually and get rid on an unneeded warning in syslog.

Best,

A.

It has probably been removed from the Debian installation process because systemd does not need it.

This functionality got added in 2020 so it probably never was used in Debian as standard!

https://github.com/slicer69/sysvinit/co … fcf8740c49

I had been wondering if we should get it added (via Debian) by default to encourage its use for drop-ins rather than maintscripts editing /etc/inittab.

My box runs on Devuan Daedalus ...

~$ uname -a
Linux devuan 6.1.0-44-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.164-1 (2026-03-09) x86_64 GNU/Linux
~$ 

... and I also get that printout in /var/log/syslog:

$ cat /var/log/syslog | grep inittab
Mar 15 04:08:48 localhost init: No inittab.d directory found
Mar 15 07:53:41 devuan init: No inittab.d directory found
Mar 15 10:10:50 devuan init: No inittab.d directory found
Mar 15 14:11:47 devuan init: No inittab.d directory found
$

That printout is not present in anywhere else.
ie: /boot, /debug, /kern.log, /messages or /user.log.

My system does not have a /etc/inittab.d directory but it is quite evident that some part of sysvinit routine is looking for it.

Indeed ...
See here: https://fossies.org/linux/sysvinit/src/init.c

I think the best is to create the directory and remove the warning from the log.
It has probably been removed from the Debian installation process because systemd does not need it.

But Devuan does not use systemd.

Best,

A.

mkdir /etc/inittab.d as root might get rid of the message.

What is the full text of the message? And what other messages came out at the same time? Those might tell us what is generating the message. Look for it in syslog, dmesg, boot.log etc.

Thank you very much! After creation of dir "/etc/inittab.d" no any warning messages at booting)

What is the full text of the message? And what other messages came out at the same time? Those might tell us what is generating the message. Look for it in syslog, dmesg, boot.log etc.