How can I create multiple Dbus session in maemo?

[Kevin, K Park.]

Hi.

I'm stuck with mutiple dbus sessions in maemo...
Currently, maemo 3.1 is just possible to create a single dbus session.
But, I want to create multiple dbus sessions in maemo and scratchbox users
to use their own dbus sessions.

So, I watched the dbus-session-init.sh, right here..



if [ -e /targets/links/scratchbox.config ]; then
 SBOX='yes'
 PARAMS="--session --print-address=2"
else
  SBOX=''
  PARAMS="--session"
fi
USER=`whoami`
SOCKETPATH =/tmp/session_bus_socket.$USER
PROG=/usr/bin/dbus-daemon
SVC="D-BUS session bus daemon"

case "$1" in
start)
  if [ "x$SBOX" = "x" ]; then
echo " DBUS  "
source $LAUNCHWRAPPER_NICE_KILL start "$SVC" $PROG $PARAMS
echo "- socket open for DBUS -"
echo "export DBUS_SESSION_BUS_ADDRESS=unix:path=$SOCKETPATH" \
 > $SESSION_BUS_ADDRESS_FILE
  else
echo " DBUS launch -"
$LAUNCHWRAPPER_NICE_KILL start "$SVC" $PROG \
 $PARAMS 2>${SESSION_BUS_ADDRESS_FILE}.in
sleep 2
if [ -r ${SESSION_BUS_ADDRESS_FILE}.in ]; then
  TMP=`cat ${SESSION_BUS_ADDRESS_FILE}.in`
  echo "export DBUS_SESSION_BUS_ADDRESS=$TMP" >
$SESSION_BUS_ADDRESS_FILE
  rm -f ${SESSION_BUS_ADDRESS_FILE}.in
fi
  fi
  ;;
stop)
  # giving parameter also here so that dsmetool works...
  source $LAUNCHWRAPPER_NICE_KILL stop "$SVC" $PROG $PARAMS
  ;;
*)
  echo "Usage: $0 {start|stop}"
  exit 1
  ;;
esac



So, I think I might have to modify this script or af-definds.sh, but I'm
not sure that...
If you guys know that, please give me your hands... :-)


Thanks for reading,
BRs

Kevin.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: maemo.org update + call to bugzilla expertise

[Jussi Kukkonen]

[EMAIL PROTECTED] wrote:
> We are not happy with bugzilla closed. We simply don't want to expose
> users email addresses. If someone knows a good and quick fix/patch
> please share it. As soon as bugzilla is not showing email addresses we
> will reopen it.

Ok. A quick note on this: I don't know if you already run version 2.18
or newer, as the bugzilla doesn't advertize it... If the version you
have is older than that, you should definitely look at the obfuscation
feature introduced in that version.
 The obfuscation is very simple (replace "@" with "@"), but at
least in the past it has been effective. The patch in bug 120030* is
really short, so it's worth looking if it applies nicely to your version.

Second point: at least Red Hat bugzilla** has a template that makes
usernames appear as just text (without mailto link) when the user is
anonymous. I'm not sure if this is available in upstream bugzilla code base.

 -jussi


 *) https://bugzilla.mozilla.org/show_bug.cgi?id=120030
Don't bother reading most of the comments -- 98% are not relevant,
and the patch is simple.
**) https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=2
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: N800 experimental host mode patches available

[Tony Lindgren]

* Marcell Lengyel <[EMAIL PROTECTED]> [070509 00:48]:
> Frantisek Dufka wrote:
> >> I tried the usb host mode on my N800, but I did not get it working. Here's 
> >> what I did:
> >> - In scratchbox SDK_ARMEL I downloaded the 
> >> kernel-source-rx-34_2.6.18-osso40 sources,
> >> - Applied the patches-tusb-n800-2007-05-04.tar.gz patch set from muru.com 
> >> (it did apply cleanly)
> >> - make n800_defconfig
> >
> > Unless those patches modify default config you need to enable usb host 
> > mode drivers in kernel configuration. You also need drivers for 
> > peripherials - for usb harddrive you need scsi and usb storage drivers.
> 
> The patches do modify the default config. I double checked and usb storage, 
> scsi & scsi block devices are all compiled in to the kernel. Along with a 
> bunch of other usb stuff like hid drivers and all.
> Anyhow, even if the drivers for peripheral devices would be missing, I should 
> still see some messages in dmesg about a new usb device connected and also 
> the device data should appear in /sys/bus/usb if host mode would be working 
> fine.

Sounds like it should work based on what you described above. Do you
have a powered hub connected? Otherwise you can only use devices that
need minimal power or are self powered.

Another thing to check is to make sure you have some gadget driver
also loader, it's currently needed to initialize the OTG driver.
So you should have g_file_storage loaded when you do lsmod.

Tony
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: N800 experimental host mode patches available

[Tony Lindgren]

Hi,

* Frantisek Dufka <[EMAIL PROTECTED]> [070509 00:05]:
> Hello,
> 
> any tips how to make custom mini-B to female A cable so that N800 
> switches mode automatically? Which two wires I need to shortcut in the 
> cable? I  can see the pinout here 
> http://pinouts.ru/Slots/USB_pinout.shtml but don't see USB ID pin. Is it 
> the fifth pin on mini-b connector (which may have no wire in the cable)? 

Here's what I did to convert a mini-A to mini-B cable: I chopped off the
top part of the mini-A end connector aluminum shielding using side cutter
pliers and a hand file so the mini-A fits into the mini-B socket on N800.

The mini-A connector has the ID pin grounded while mini-B end has the
ID pin floating. Of course if you can figure out a way to ground the ID
pin on the mini-B end, that will do the trick too. ID pin is the fifth
pin on the mini connectors.

Regards,

Tony
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: maemo.org update + call to bugzilla expertise

[Neil MacLeod]

[EMAIL PROTECTED] wrote:
> Hi,
> 
> - Do not hesitate filing more reports even if they are minor things or 
> matter of taste. We are happy agreeing with you the next moves. We are 
> working on the fixes since the very minute the site was launch and we 
> have just agreed to put some more resources to help fixing these bugs.
> 

Before I file a bug, can anyone describe how the new "Wiki" can be searched 
without including hits from other parts of maemo.org?

The only visible search box appears to search the whole of maemo.org - for 
instance after navigating to Maemo->Wiki, a search for "Bluetooth DUN" returns 
links to maemo.org/pipermail etc. when I am only interested in finding Wiki 
pages. With the old site we could limit searches to only Wiki content (by title 
or text) so this seems a step backwards.

Perhaps we need a method to limit the scope of the Search (assuming one doesn't 
exist already which I have missed)? For example limit search to Wiki, 
Newsgroups, Documentation, Entire Site etc. Or have the search context 
sensitive - search only the Wiki pages when I navigate to the Wiki area, and 
search the entire site from the home page?

___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

RE: Bugzilla brokan?

[quim.gil]

>For example, whenever I receive an update on one of my bug 
>reports, the URL looks something like this:

Bug known, working on it:

https://maemo.org/bugzilla/show_bug.cgi?id=1365

Quim
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

RE: maemo.org update + call to bugzilla expertise

[quim.gil]

 
>I assume you don't mean bugzilla spammers

Sorry, I meant spambots gathering email addresses. In our default
bugzilla setup email addresses were exposed. Some pages accessible
publicly offered bug listings with dozens of addresses. Since we were
(and we are) busy with the new website we decided to cut completely the
access to anonymous users.

> Personally I believe this is a bad solution for an issue tracking
>system:

We are not happy with bugzilla closed. We simply don't want to expose
users email addresses. If someone knows a good and quick fix/patch
please share it. As soon as bugzilla is not showing email addresses we
will reopen it.


>The fact that new users may not know that their email address 
>will be publicly viewable is a problem

It's not about users knowing or not knowing. Bugzilla or any other web
tool shouldn't expose addresses at all, or at least not in a way that
spiders can process.


>No integration is actually a good thing at the moment, as many 
>garage accounts just do not work on the core website

Obviously we will fix bugs first and then we will go for new features.

> It would be nice to know if there is any progress on this 

Not according to https://maemo.org/bugzilla/show_bug.cgi?id=1152 .
Ferenc is basically fixing stuff full time. Even if this is an important
issue there are other things that are more broken. Sorry about that.

Quim
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Bugzilla brokan?

[Sean Luke]

Since the recent modifications to maemo, bugzilla appears to have  
broken.  I'd report the bugzilla bugs to bugzilla, but bugzilla's  
bugzilla bug reporting is broken too.  :-)  I love recursion.

For example, whenever I receive an update on one of my bug reports,  
the URL looks something like this:

https://bugs.maemo.org/show_bug.cgi?id=1256

Going to all new such report URLs now gives a 404: Not Found.  It  
think it should have been:

https://maemo.org/bugzilla/show_bug.cgi?id=1256

For much of this morning and yesterday, bugzilla has been completely  
down, at least from all the spots in the DC area I have access to  
(several different college campuses).

Sean
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: Can we launch Maemo UI on several Xephyrs from different clients connected to a server?

[Eero Tamminen]

Hi,

ext Kevin, K Park. wrote:
>> Please don't use hostname for local connections and use "-nolisten tcp"
>> in the Xephyr invocation (you don't usually want to open extra holes to
>> network and using unix socket for X communication for instead of inet
>> one is faster & what happens on the target device).
> 
>  you mean, when I lanuch start-xephyr.sh, add that flag(-nolisten tcp) into
> this script?

For Xephyr, yes.  This disables the tcp/ip socket so that the X server
offers only unix socket.


>  please, more detail description about what you said...
> 
>>  7) Lastly, dino launched Maemo UI, like this
>> >
>> >   [sbox-SDK_X86: ~] > af-sb-init.sh start
>> >
>> >..as a result, it terminated...with that log.
>> >
>> >  /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid
>> identifier
>> >  /usr/bin/af-sb-init.sh: line 1: export:
>> `/tmp/session_bus_socket:':
>>
>> I think D-BUS socket is fixed one in Maemo.  You might be able to make
>> it to use display or user name in the startup scripts.
> 
>   startup scripts is af-sb-init.sh, right?

I meant the startup scripts in /etc/osso-af-init/ dir.
AFAIK af-sb-init.sh just calls then like init would.


- Eero
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: maemo.org update + call to bugzilla expertise

[Jussi Kukkonen]

[EMAIL PROTECTED] wrote:
> - About maemo Bugzilla, the tool itself. We have upgraded the software
> and integrated the design to the rest of the site. We have basically
> closed the access to non-registered users to avoid spammers even if we
> agree it's not the best option.

I assume you don't mean bugzilla spammers -- as preventing anonymous
searching and viewing doesn't really help there -- but e-mail spammers
harvesting addresses from bugzilla?
 Personally I believe this is a bad solution for an issue tracking
system: Although the fear of spam may prevent some people from
submitting bugs, making the bugzilla practically invisible will probably
do more damage (bugs won't show up in search results for example).

The fact that new users may not know that their email address will be
publicly viewable is a problem, but that can be fixed by adding a note
to https://maemo.org/bugzilla/createaccount.cgi. As for older
accounts... well, that milk's already spilled.

ref: https://maemo.org/bugzilla/show_bug.cgi?id=1344

> We will integrate soon the users so same login will apply to the
> core website, Garage and bugzilla.

No integration is actually a good thing at the moment, as many garage
accounts just do not work on the core website: now I can at least file
bugs when things don't work.
 It would be nice to know if there is any progress on this (I'd like to
work on the wiki).

ref: https://maemo.org/bugzilla/show_bug.cgi?id=1152

 - jussi

___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Can we use local network inside scratchbox?

[Kevin, K Park.]

Hi,

I want to use private IP inside scratchbox like VMware..
So, scratchbox users can be assigned these virtual IP as static assigning..
Finally users in scratchbox can trasfer packets to each other...

For this, I created virtual IPs from host PC...However, I couldn't modify
any configuration files in scratchbox at all. How to solve this problem?
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: Can we launch Maemo UI on several Xephyrs from different clients connected to a server?

[Kevin, K Park.]

On 5/9/07, Eero Tamminen <[EMAIL PROTECTED]> wrote:


Hi,

ext Kwansu Park wrote:
> 4) Then I logged in scratchbox(Apophis (Current stable version)) as each
> user.
>
> 5) For the first time, kevin set its DISPLAY, like this
>[sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:2
>
>and then another user, dino did the same work except to number of
> DISPLAY
>
>[sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:3

Please don't use hostname for local connections and use "-nolisten tcp"
in the Xephyr invocation (you don't usually want to open extra holes to
network and using unix socket for X communication for instead of inet
one is faster & what happens on the target device).



 you mean, when I lanuch start-xephyr.sh, add that flag(-nolisten tcp) into
this script?
 please, more detail description about what you said...


 7) Lastly, dino launched Maemo UI, like this
>
>   [sbox-SDK_X86: ~] > af-sb-init.sh start
>
>..as a result, it terminated...with that log.
>
>  /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid
identifier
>  /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':

I think D-BUS socket is fixed one in Maemo.  You might be able to make
it to use display or user name in the startup scripts.



  startup scripts is af-sb-init.sh, right?
  so, user name in this script?
  or display, how do I deal with display for solving this problem?



  - Eero


___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: Can we launch Maemo UI on several Xephyrs from different clients connected to a server?

[Eero Tamminen]

Hi,

ext Kwansu Park wrote:
> 4) Then I logged in scratchbox(Apophis (Current stable version)) as each
> user.
> 
> 5) For the first time, kevin set its DISPLAY, like this
>[sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:2
> 
>and then another user, dino did the same work except to number of
> DISPLAY
> 
>[sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:3

Please don't use hostname for local connections and use "-nolisten tcp"
in the Xephyr invocation (you don't usually want to open extra holes to
network and using unix socket for X communication for instead of inet
one is faster & what happens on the target device).


>  7) Lastly, dino launched Maemo UI, like this
> 
>   [sbox-SDK_X86: ~] > af-sb-init.sh start
> 
>..as a result, it terminated...with that log.
> 
>  /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid identifier
>  /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':

I think D-BUS socket is fixed one in Maemo.  You might be able to make
it to use display or user name in the startup scripts.


- Eero
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: A Pitch for making SAMBA Server a core part of the N800 OS

[Acadia Secure Networks]

Danny,

I have had great success using SAMBA in my lab to "interwork" between a 
RHEL3.0 Linux server running on a Thinkpad570 and my main desktop system 
running Windows XP (as well as other OS's including Vista an Windows 
Server 20003  by means of VMware and Virtual PC).


In those cases where I have had a need to edit files that originated on 
the Linux server (typically .conf files) I simply used Notepad which 
seems to work just fine for the purpose. Also I make use of 
products/apps such as Ethereal where the file contents are structured 
the same whether Ethereal is running on Linux or on Windows. Thus I can 
pull an Ethereal file off of a Linux system and view it on my Desktop 
system using Ethereal on Windows XP.



In fact it would be nice to have a version of Ethereal installable on 
the N800 in order to perform packet capture for later playback on a 
different system.


Depending upon which version and which features the Nokia team chooses 
to implement with respect to Samba, and, assuming they will use Samba 
3.0,  it might even be possible to join the N800 as a member serve in a 
Windows Active Directory domain. See, for example, the recipe at the 
following url:


   http://www.wlug.org.nz/ActiveDirectorySamba


I should add that SAMBA represents one of the great ongoing 
reverse-engineering projects in the Linux world for what has already 
been accomplished up until now and the Samba team is still at it. His 
the url to some recent commentary from a SAMBA team member on how the 
process works:


  
http://searchenterpriselinux.techtarget.com/originalContent/0,289142,sid39_gci1243859,00.html



Have you ever tried to take a steak away from a rottweiler? Like
they say though, if you want to learn French you have to sit in a
French café and learn it. All we do [at samba] is stare at packets
and network traffic. Active Directory is mostly standardized, so
we look at the traffic and see what works. A big thing about Samba
and bugs is if Windows does it and we don't it's considered a bug.
No Windows client is going to use a port that it wasn't designed
to use. It's as simple as that.



Best Regards,



John Holmblad



Acadia Secure Networks





Danny Milosavljevic wrote:

Hi,

On Wed, 28 Mar 2007 22:22:01 -0400, Acadia Secure Networks wrote:
  
I think that making SAMBA a core (i.e. supported by Nokia) part of the 
N800 os would greatly facility development, testing, and experimentation 
with the N800, at least for those, like me, who work mostly using 
Microsoft Windows desktop systems.


The ability  to share out the N800 file system to a Windows computer and 
thereby have the ability to  inspect/modify files using, e.g. notepad, 
would make me more productive. 



You wouldn't see half of the file flags ("executable" comes to mind) (at least not like a 
normal human would expect to see them - as a check box with the text "Executable").
The files you edit wouldn't work (mswindows uses different line endings).

Now of course you can use special windows apps (special editor, special file 
manager), but then it's not easy anymore.

And my idea of an internet tablet somehow doesn't include it being a file 
server ;-) Of course, you might see that differently.

  
I know I can install SAMBA on my own but 
it seems to me that this is, what I would call, a "bread and butter" 
capability of a Linux OS that needs to co-exist in the real world of 
SMB/CIFS as it is today and, which, will not go away anytime soon.



Having SMB/CIFS _client_ functionality probably would be useful.

(Time to check whether the source of the file manager is available yet :-) - on 
second thought, I have ssh access to all my machines, what's the point)
(Hmm, how about adding SFTP - or just a nice FUSE mount gui - to the file 
manager? I wonder if I can)

cheers,
  Danny

___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

  
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Can we launch Maemo UI on several Xephyrs from different clients connected to a server?

[Kwansu Park]

Hi,
I'm a new comer in this field with maemo.

Actually, I want to launch Maemo UIs onto several Xephyrs.
So I found a thread related to this problem on this mailing list. At that
time, author was Sampath. Maybe the subject of that thread is the same as
mine. :) In the previous thread, there was just one solution about launching
several Xephyrs using a wrapper script( e.g. start-xephyr.sh). There was not
an answer for Maemo UI problems

Finally, when I try to launch Maeomo UI(af-sb-init.sh start) on each xephyr,
it failed except the first one.
The detail situation of is follows..

1) I made 2 start-xephyr.sh, it works for each user.
For a user "kevin"
#!/bin/sh -e
prefix=/scratchbox/users/kevin/targets/SDK_X86/usr
export LD_LIBRARY_PATH=${prefix}/lib; export LD_LIBRARY_PATH
exec ${prefix}/bin/Xephyr :$1 -host-cursor -screen 800x480x16 -dpi 96
-ac

 For a user "dino"
#!/bin/sh -e
prefix=/scratchbox/users/dino/targets/SDK_X86/usr
export LD_LIBRARY_PATH=${prefix}/lib; export LD_LIBRARY_PATH
exec ${prefix}/bin/Xephyr :$1 -host-cursor -screen 800x480x16 -dpi 96
-ac


2) Then, I executed these scripts on the gdm xserver(Ubuntu 6.1 Edgy
distributed)
   root$start-xephyr-kevin.sh 2
   root$start-xephyr-dino.sh 3

as a result, on the screen, 2 Xephyr launched.

3) On an other PC, executed terminal program(SecureCRT) and then I logged
in server as other users, kevin and dino.

4) Then I logged in scratchbox(Apophis (Current stable version)) as each
user.

5) For the first time, kevin set its DISPLAY, like this
   [sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:2

   and then another user, dino did the same work except to number of
DISPLAY

   [sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:3


6) Then, kevin launched Maemo UI firstly

  [sbox-SDK_X86: ~] > af-sb-init.sh start

   ..as a result, it worked well, I could find that maemo UI
launched on the xephyr displayed on 2

 7) Lastly, dino launched Maemo UI, like this

  [sbox-SDK_X86: ~] > af-sb-init.sh start

   ..as a result, it terminated...with that log.

 /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid identifier
 /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':
not a valid identifier
 Sample files present.
 Starting DBUS system bus
 Starting D-BUS session bus daemon
 /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid identifier
 /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':
not a valid identifier
 /etc/init.d/maemo-launcher: line 1: export: `bus:': not a valid
identifier
 /etc/init.d/maemo-launcher: line 1: export:
`/tmp/session_bus_socket:': not a valid identifier
Starting Maemo Launcher:
maemo-launcher/home/dino/.osso/current-gtk-key-theme:1: Unable to find
include file: " keybindings.rc"
maemo-launcher: died binding to invoker socket
start failed.
Starting Sapwood image server
sapwood-server[12755]: GLIB INFO default - server started
Starting Matchbox window manager
matchbox: WARNING: failed to load /usr/share/matchbox/mbnoapp.xpm .
Disabling icons.
matchbox: failed to load keyboard config
Starting clipboard-manager
Starting Keyboard
Starting MAEMO AF Desktop
[sbox-SDK_X86: ~] >

8) Following End.


Well, I think it might be problem at DBUS sessions and maemo UI
initscript...
I expected that it'd work well if that problems are fixed...
Somebody who knows that issue, please, let me know how to fix it...

Thanks,
BRs.

Kevin, K Park.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Can we launch Maemo UI on several Xephyrs from different clients connected to a server?

[Kwansu Park]

Hi,
I'm a new comer in this field with maemo.

Actually, I want to launch Maemo UIs onto several Xephyrs.
So I found a thread related to this problem on this mailing list. At that
time, author was Sampath. Maybe the subject of that thread is the same as
mine. :) In the previous thread, there was just one solution about launching
several Xephyrs using a wrapper script(e.g. start-xephyr.sh). There was not
an answer for Maemo UI problems

Finally, when I try to launch Maeomo UI(af-sb-init.sh start) on each xephyr,
it failed except the first one.
The detail situation of is follows..

1) I made 2 start-xephyr.sh, it works for each user.
For a user "kevin"
#!/bin/sh -e
prefix=/scratchbox/users/kevin/targets/SDK_X86/usr
export LD_LIBRARY_PATH=${prefix}/lib; export LD_LIBRARY_PATH
exec ${prefix}/bin/Xephyr :$1 -host-cursor -screen 800x480x16 -dpi 96
-ac

 For a user "dino"
#!/bin/sh -e
prefix=/scratchbox/users/dino/targets/SDK_X86/usr
export LD_LIBRARY_PATH=${prefix}/lib; export LD_LIBRARY_PATH
exec ${prefix}/bin/Xephyr :$1 -host-cursor -screen 800x480x16 -dpi 96
-ac


2) Then, I executed these scripts on the gdm xserver(Ubuntu 6.1 Edgy
distributed)
   root$start-xephyr-kevin.sh 2
   root$start-xephyr-dino.sh 3

as a result, on the screen, 2 Xephyr launched.

3) On an other PC, executed terminal program(SecureCRT) and then I logged
in server as other users, kevin and dino.

4) Then I logged in scratchbox(Apophis (Current stable version)) as each
user.

5) For the first time, kevin set its DISPLAY, like this
   [sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:2

   and then another user, dino did the same work except to number of
DISPLAY

   [sbox-SDK_X86: ~] > export DISPLAY=127.0.0.1:3


6) Then, kevin launched Maemo UI firstly

  [sbox-SDK_X86: ~] > af-sb-init.sh start

   ..as a result, it worked well, I could find that maemo UI
launched on the xephyr displayed on 2

 7) Lastly, dino launched Maemo UI, like this

  [sbox-SDK_X86: ~] > af-sb-init.sh start

   ..as a result, it terminated...with that log.

 /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid identifier
 /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':
not a valid identifier
 Sample files present.
 Starting DBUS system bus
 Starting D-BUS session bus daemon
 /usr/bin/af-sb-init.sh: line 1: export: `bus:': not a valid identifier
 /usr/bin/af-sb-init.sh: line 1: export: `/tmp/session_bus_socket:':
not a valid identifier
 /etc/init.d/maemo-launcher: line 1: export: `bus:': not a valid
identifier
 /etc/init.d/maemo-launcher: line 1: export:
`/tmp/session_bus_socket:': not a valid identifier
Starting Maemo Launcher:
maemo-launcher/home/dino/.osso/current-gtk-key-theme:1: Unable to find
include file: "keybindings.rc"
maemo-launcher: died binding to invoker socket
start failed.
Starting Sapwood image server
sapwood-server[12755]: GLIB INFO default - server started
Starting Matchbox window manager
matchbox: WARNING: failed to load /usr/share/matchbox/mbnoapp.xpm .
Disabling icons.
matchbox: failed to load keyboard config
Starting clipboard-manager
Starting Keyboard
Starting MAEMO AF Desktop
[sbox-SDK_X86: ~] >

8) Following End.


Well, I think it might be problem at DBUS sessions and maemo UI
initscript...
I expected that it'd work well if that problems are fixed...
Somebody who knows that issue, please, let me know how to fix it...

Thanks,
BRs.

Kevin, K Park.
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Gregale SDK on new website?

[Graham Cobb]

I need to install the gregale SDK (as I still want to build for 770 users 
using IT2006).  But I can't find the rootstrap after the website 
reorganisation.

Before the change I bookmarked the most important page: 
http://maemo.org/platform/docs/tutorials/Maemo_tutorial.html#Installing-Maemo-rootstrap.
  
Unfortunately that page does not exist any more.

I noticed that the Gregale announcement is on the maemo.org home page but the 
links it contains are broken.  So, next, I tried searching for 2.2 in the 
search box -- but that seems to just do a Google search which still contains 
the old links.

After looking around I eventually found 
http://maemo.org/development/sdks/maemo-2-2--gregale-.html and 
http://maemo.org/development/documentation/tutorials/Maemo_2.2_Tutorial.html.

However, neither page has valid links to the SDK rootstraps!  The 
maemo-2-2-gregale.html page has no link to the rootstraps at all (despite 
having a "quick downloads" section).  The tutorial page claims to have a link 
but it is "site:downloads/download-maemo", which does not work, of course.

So: 

1) How do I find the 2.2 SDK downloads?

2) More generally, how do I find anything in the new structure as the search 
only gives the old links?

3) Are there any plans to fix the broken links?  At least in the Nokia section 
of the website?

Graham
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: Never Try to Outsmart the OS

[Mikhail Sobolev]

Hi John

On Tue, May 08, 2007 at 08:19:42PM -0400, Acadia Secure Networks wrote:
> I have been unable to get the x11vnc and free42 apps to show up in the 
> simple launcher menu. Unfortunately when I attempt to select either of 
> these two apps the checkbox does not check. These are the only 2 apps on 
> the simple launcher apps list, in fact, that do not check when the 
> checkbox is selected.
Could you please advise where you took the applications from?  I'm
_slowly_ working on SimplLauncher now, but I'd like to make a release
soon and this kind of issues I'd like to address as much as possible.

--
Misha


signature.asc
Description: Digital signature
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: N800 experimental host mode patches available

[Marcell Lengyel]

Frantisek Dufka wrote:
>> I tried the usb host mode on my N800, but I did not get it working. Here's 
>> what I did:
>> - In scratchbox SDK_ARMEL I downloaded the kernel-source-rx-34_2.6.18-osso40 
>> sources,
>> - Applied the patches-tusb-n800-2007-05-04.tar.gz patch set from muru.com 
>> (it did apply cleanly)
>> - make n800_defconfig
>
> Unless those patches modify default config you need to enable usb host 
> mode drivers in kernel configuration. You also need drivers for 
> peripherials - for usb harddrive you need scsi and usb storage drivers.

The patches do modify the default config. I double checked and usb storage, 
scsi & scsi block devices are all compiled in to the kernel. Along with a bunch 
of other usb stuff like hid drivers and all.
Anyhow, even if the drivers for peripheral devices would be missing, I should 
still see some messages in dmesg about a new usb device connected and also the 
device data should appear in /sys/bus/usb if host mode would be working fine.

BR,

Marcell





 

Get your own web address.  
Have a HUGE year through Yahoo! Small Business.
http://smallbusiness.yahoo.com/domains/?p=BESTDEAL
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: Launch maemo browser in fullscreen mode

[Eero Tamminen]

Hi,

ext Erin Stadler wrote:
> The problem with the solution below is that you can still minimize the app
> and then you are stuck with a white screen with nothing left to do but
> restart it or re-flash it. Is there a way to let the desktop boot up but
> stop other applications from loading other than the browser, so essentially
> you keep the person from being to browse to anything other then the program
> you give them?

You need to:
- disable Desktop from starting and start only Browser
- and use the same policy for Browser as for Desktop so
  that it's restarted when user closes it (see desktop
  startup script)
- change Matchbox theme (xml file, documented in Matchbox site)
  so that window titlebar doesn't have minimize button


- Eero



> 2007/3/23, Simon Moore   >:
>> It does work, I've tried it.  Not specific to the
>> browser of course, can send any key to any
>> screen, but in this case you send F6 when the
>> browser is the front window to make it full screen.
>>
>> Anyone got any tips on changing X windows (i.e.
>> removing window manager) so just the browser
>> starts up with no other Apps?  It would make the
>> above work so much better as a kiosk. (nokia 770)
>  
> A quick and dirty way would be to replace maemo-af-desktop with osso
> browser in maemo-af-desktop startup script
> (/etc/osso-af-init/maemo-af-desktop.sh).
>  
> A more proper way in my opinion would be to write a real init script
> for it, and just remove the unneeded programs from the af initscripts.

___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers

Re: N800 experimental host mode patches available

[Frantisek Dufka]

Hello,

any tips how to make custom mini-B to female A cable so that N800 
switches mode automatically? Which two wires I need to shortcut in the 
cable? I  can see the pinout here 
http://pinouts.ru/Slots/USB_pinout.shtml but don't see USB ID pin. Is it 
the fifth pin on mini-b connector (which may have no wire in the cable)? 
Thanks.

Frantisek
___
maemo-developers mailing list
maemo-developers@maemo.org
https://maemo.org/mailman/listinfo/maemo-developers