Bugzilla will be upgraded to version 5.0 on December 2, 2018. The outage period for the upgrade will start at 0:00 UTC and have a duration of 12 hours
Bug 131963 - init script missing
init script missing
Product: Red Hat Enterprise Linux 3
Classification: Red Hat
Component: initscripts (Show other bugs)
i386 Linux
medium Severity medium
: ---
: ---
Assigned To: Bill Nottingham
Brock Organ
Depends On:
  Show dependency treegraph
Reported: 2004-09-07 10:44 EDT by Berthold Cogel
Modified: 2014-03-16 22:48 EDT (History)
3 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2005-09-20 17:24:41 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Berthold Cogel 2004-09-07 10:44:11 EDT
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; de-DE; rv:1.7.2)

Description of problem:
keytables are not loaded at system startup if serial console is
activated. /etc/rc.d/init.d/keytable is missing in package and older
version is removed during update.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1.Install fresh system from Taroon Update 2 CD set with keyboard other
than qwerty (for example a german keyboard). Or install system with
previous CD sets and update kbd to kbd-1.08-10.2 .    
2.Activate a serial console
3.Login on both serial and local console and try to type a 'z' or 'y'
on both screens.

Actual Results:  keytable for (german) keyboard is not loaded if
kernel parameter console=ttyS0,... is used. Keyboard mapping for local
console is not set.

Expected Results:  Local console should use the selected keytable
(german, ...).

Additional info:
Comment 1 Berthold Cogel 2004-09-08 09:40:52 EDT
Copying the initscript from an older version solves the problem.
Comment 2 Bill Nottingham 2004-09-28 21:48:24 EDT
What happens if you use console=ttyS0 console=tty0?
Comment 3 Berthold Cogel 2004-09-29 06:30:31 EDT
We use console=tty0 console=ttyS0,57600n8r. With only console=tty0,
everything works as expected.  
When we use the keytable initscript from a previous version of the kbd
package and console=tty0 console=ttyS0,57600n8r, everything works as

The solution for the problem is simple. Provide a kbd package with a
working initscript.

Comment 4 Bill Nottingham 2004-09-29 15:57:38 EDT
That's superfluous for 99% of cases, and actually causes bugs in the
presence of some serial consoles.
Comment 5 Berthold Cogel 2004-10-08 05:18:27 EDT
In our case we have problems with the local Console (not the serial
console) if /etc/rc.d/init.d/keytable is missing or not executed
during bootup. The keyboard layout is not set correctly. 
OK: If keytable is called during bootup, the characterset for the
serial console is mixed up and the local terminal has to be reset. 

Case 1: keytable is deaktivated (chkconfig keytable off, reboot)

  keyboard layout at serial console: OK (QWERTZ)
  characterset at serial console: OK (even german umlaut)

  keyboard layout at local console: QWERTY instead QWERTZ 
  characterset at local console: seems to be us_US, german umlauts
                 are missing although locale is set to de_DE.UTF-8

  login with ssh: OK

Case 2: keytable is aktivated (chkconfig keytable on, reboot)

  keyboard layout at serial console: OK (QWERTZ)
  characterset at serial console: OK (even german umlaut)

  keyboard layout at local console: OK
  characterset at local console: OK
  login with ssh: OK

After keytable is executed the display at the serial console is  
screwed up: 

NFS statd starten: [  OK  ]
[  OK  ]r starten: [  OK  ]
Starting keytable:  Ä  OK  Ü
Ä  OK  Ü audit subsystemÄ  OK  Ü

I get TERM=linux for local and serial console in both cases. 

Especially for WS users with QWERTZ layout keyboards, working on their
local console might be very annoying without the correct settings. 
There are still some of the rare species out there, who still know how
to work with the keyboard. Not all are addicted to working with mice.

By the way: Shouldn`t the system take care that it is aware of the
type of console the user is using for login and switch to the
appropriate settings for the session? The users choice of the language
and keyboard settings should be respected and not superseded. OK:
Perhaps superseeding is needed for some special cases. For example for
serial consoles, but not for the local console. 

Oh: Which are the 99% of cases where keytable is superfluous? I don't
think that 99% of RHEL3 users are using us_US. 
And: If keytable causes bugs, the reason for those bugs should be
fixed. Discarding keytable is not the kind of bugfix I would prefer.

Comment 6 Bill Nottingham 2004-10-08 12:17:43 EDT
In the case where you are using only one type of console, keytable is
superfluous. The only case it may be needed is when you have secondary
consoles that aren't /dev/console.
Comment 7 Bill Nottingham 2004-10-08 12:18:17 EDT
And, as you stated, in that case it will break the serial console.

Note You need to log in before you can comment on or make changes to this bug.