X2Go Bug report logs - #402
Numlock issues

Full log

Message #15 received at 402@bugs.x2go.org (full text, mbox, reply):

Received: (at 402) by bugs.x2go.org; 30 Jan 2014 08:46:07 +0000
From jordi@amospalla.es  Thu Jan 30 09:46:06 2014
X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on
	ymir.das-netzwerkteam.de
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	version=3.3.2
Received: from ispconfig.net1.amospalla.es (amospalla.es [188.40.66.228])
	by ymir (Postfix) with ESMTP id 776FE5DA6C
	for <402@bugs.x2go.org>; Thu, 30 Jan 2014 09:46:06 +0100 (CET)
Received: from localhost (localhost [127.0.0.1])
	by ispconfig.net1.amospalla.es (Postfix) with ESMTP id 1F3B21AC12AD;
	Thu, 30 Jan 2014 09:46:06 +0100 (CET)
X-Virus-Scanned: Debian amavisd-new at ispconfig.amospalla.es
Received: from ispconfig.net1.amospalla.es ([127.0.0.1])
	by localhost (ispconfig.amospalla.es [127.0.0.1]) (amavisd-new, port 10024)
	with ESMTP id LCWakhDrPP5J; Thu, 30 Jan 2014 09:46:05 +0100 (CET)
Received: from amospalla.es (localhost [IPv6:::1])
	by ispconfig.net1.amospalla.es (Postfix) with ESMTP id D8E241AC0887;
	Thu, 30 Jan 2014 09:46:04 +0100 (CET)
Received: from 62.82.108.2
        (SquirrelMail authenticated user jordi@amospalla.es)
        by amospalla.es with HTTP;
        Thu, 30 Jan 2014 09:46:04 +0100
Message-ID: <bfeb081fdd3ec995c47b4ee7c9c96654.squirrel@amospalla.es>
In-Reply-To: <8be4d14fda106d24b56c8cf94ec9a918.squirrel@www.amospalla.es>
References: <8be4d14fda106d24b56c8cf94ec9a918.squirrel@www.amospalla.es>
Date: Thu, 30 Jan 2014 09:46:04 +0100
Subject: Re: [X2Go-Dev] Bug#402: Numlock issues
From: jordi@amospalla.es
To: jordi@amospalla.es,
 402@bugs.x2go.org,
 x2go-dev@lists.berlios.de
User-Agent: SquirrelMail/1.4.23 [SVN]
MIME-Version: 1.0
Content-Type: text/plain;charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Priority: 3 (Normal)
Importance: Normal

> Package: x2goserver
> Version: 4.0.1.12-0x2go2+git20140108.809+wheezy.main.1
>
> Hello, numlock does not behave correctly. It behaves like if it was
> randomly and continuously toggling it state.
>
> Happens when numlock key is pressed, and after some time, it seems to
> stabilize in some state (on or off), but it can take some time.
>
> Happens on Mate 1.6 desktop environment over Debian 6. No problem on same
> host with a KDE session.
>
> If numlock state is modified with numlockx binary, it works as expected,
> only the key triggers this.
>
> user@host:~$ while numlockx status; do sleep 0.1; done
> Numlock is off
> Numlock is off
> Numlock is on
> Numlock is on
> Numlock is off
> Numlock is on
> Numlock is off
> Numlock is on
> Numlock is off
> Numlock is on
> Numlock is off
> Numlock is off
> Numlock is on
> Numlock is on
> Numlock is off
> Numlock is on
> Numlock is on
> ...

Found that the Mate setting which makes numlock state change at a very
high rate can be disabled by:
  gsettings set org.mate.peripherals-keyboard remember-numlock-state false

Still, users notify me that numlock change randomly. I did monitor
numlock status with "numlockx status" command, and I can see it do
change but very intermitently.

I first monitored numlock status and actively changed it to on with a
script.

Lately, I added to the keyboard options, the equivalent to "setxkbmap -o
numpad:mac", which can be added to
/org/mate/desktop/peripherals/keyboard/kbd/options with
'compat\tnumpad:mac'

Here is a dirty script, which every user runs at session startup:

---------------%-----------------%------------------%------------------
#!/bin/bash
tab="$(printf "\t")"
cadena=( $(dconf read /org/mate/desktop/peripherals/keyboard/kbd/options |
sed -e 's/.*\[\|\]//g') )
for (( i=0;i<${#cadena[@]};i++ )); do
	# esborrar '[', ']' i ','
	cadena[$i]="$(printf "${cadena[$i]}" | sed -e 's/,//g')"
done
if ! echo ${cadena[@]} | grep -q 'compat\snumpad:mac'; then
	for (( i=0;i<${#cadena[@]};i++ )); do
		nova="$nova ${cadena[$i]},"
	done
	nova="$nova 'compat${tab}numpad:mac'"
	dconf write /org/mate/desktop/peripherals/keyboard/kbd/options "[ $nova ]"
fi
gsettings set org.mate.peripherals-keyboard numlock-state          'on'
---------------%-----------------%------------------%------------------

Send a report that this bug log contains spam.