Re: N900 consumes too much power

[Eero Tamminen]

Hi,

On 08/07/2012 04:20 AM, ext Cedric Cellier wrote:

Of course offline mode is not normal. I suggested it
for testing :) So that whatever software is the culprit
it can no longer drain the battery very fast. If you can't
stand several days in offline mode and without top reporting
a process that's heavy on the cpu


It doesn't need to be that heavy.

Even a process that constantly wakes up just at 1s interval,
(which shows only as 0-1% of CPU usage in top) will
reduce device idle use-time from one week[1] to a day.

This is when just CPU is used.  If also network is used,
the wakeups can be much rarer and they still ruin the use-time
even worse.  I'm not completely sure of these numbers, but
if your WLAN accesspoint power management is working, I think
even <5 minute interval wakeups are bad, with phone networks,
the network access intervals need to be much longer for device
to be able to save power.

These kind of wakeup frequencies you don't notice with top, you
need either strace the programs or use e.g. nethogs[3] utility.

Note also that the weaker the network signal strength is,
the more power is required.

[1] A week with a full, *new* battery [2],  no services installed etc.

[2] Battery capacity worsens with time, with use and higher temperatures
(check the manufacturing date when buying batteries!), see:
https://bugs.maemo.org/show_bug.cgi?id=7636#c16

[3] http://packages.debian.org/squeeze/nethogs



then you'll be certain your  hardware is faulty.



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

Re: Reducing N900 RAM usage

[Eero Tamminen]

Hi,

On 03/18/2012 04:07 AM, ext Jan Knutar wrote:

On Wednesday 07 March 2012, Paul Hartman wrote:

A freshly-flashed N900 is so much faster and more responsive compared
to my N900 with all my favorite apps installed, mail accounts, MfE,
chat accounts, themes, status bar widgets, and so on. I'm guessing
these things all have a negative impact...

Has anyone done a study into the impact of these various things and
RAM usage? For example how much RAM can I spare if I don't use Modest
at all, or don't use chat accounts, etc. allowing me to weigh the
  cost vs. benefit of using these features.


IIRC, Skype when I used it seemed to be about 5-10Meg impact on RSS.

In general, I'd imagine having any QT widgets would add 10-20M use. Not
so sure myself, I stopped using widgets entirely, not enough ram for
luxuries :)


You can use sp-endurance to measure and visualize usage of all system 
resources.



I would recommend getting the latest version from here:
https://maemo.gitorious.org/maemo-tools/sp-endurance

And the debian packaging from maemo-packaging branch and remove
the Harmattan specific "aegis" stuff from the debian/control file
before building the package.

The latest version has pretty nice graphing utilities.


Instructions on the use are here:

http://harmattan-dev.nokia.com/docs/library/html/guide/html/Developer_Library_Developing_for_Harmattan_Developer_tools_Performance_testing_tools_Using_sp-endurance-postproc.html
http://harmattan-dev.nokia.com/docs/library/html/guide/html/Developer_Library_Developing_for_Harmattan_Developer_tools_Performance_testing_tools_Using_sp-endurance.html

They're about the same as for N900:
http://wiki.maemo.org/Documentation/devtools/maemo5/sp-endurance
http://wiki.maemo.org/Documentation/devtools/maemo5/sp-endurance-postproc

Except for the new endurance_plot utility which generates the nice 
graphs with gnuplot.



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

Re: Troubles with chrooted Debian

[Eero Tamminen]

Hi,

On 01/15/2012 05:26 AM, ext Pavel Řezníček wrote:

But What I noticed during my experiments is that the loop mount support
is somehow bad. When Easy Debian was mounted from an image file,
as I started an I/O-intensive or CPU-intensive task such as copying the
contents of the image to the card,


Loop images aren't a good idea for several reasons.

* Using them can take a lot more (kernel) memory than
  just chrooting to a directory with debian stuff

  If device runs completely out of memory without it being
  cause by user-space processes that could be killed, or
  it requiring also SW watchdog protected things (like X server
  to be killed), device will be rebooted.



my device almost always ended up in a shutdown or reboot and huge
filesystem corruption inside the image file.


* That's fairly obvious. Journaling file system guarantee only
  consistency of the file system metadata, not file contents,
  and you're having two such things on top of each other.

  Therefore meta data of the file system within the loop image
  isn't protected and gets corrupted if you've been doing file
  system modifications inside the loop image just before reboot.

  For such modification not to be corrupted on reboot, they would
  need to have been first synched to the image itself by the FS
  used inside the image, then synched to the real media by the file
  system on where the loop image itself resides.



Yes, even if I ran /rsync/ with the highest nice level. This is
something for the power kernel developers to consider and think about.
(But keep in mind I work with the stock kernel.)



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

Re: N900 microSD card I/O errors and corruption

[Eero Tamminen]

Hi,

On 03/29/2011 06:32 PM, ext Paul Hartman wrote:

I've got three microSD cards. They work fine on my PCs, I've done
read/write tests and data is not corrupted. But, in my N900, two of
the three are not stable, leading to corruption.


Does it afterwards show as corrupted on the PC too?


Transcend 8GB class 6 - bad
Adata 16GB class 10 - bad
Sandisk 16GB class 2 - good

I suspect maybe the N900 isn't providing enough voltage to the SD card
and some cards are less tolerant of low-voltage situations than
others. Does anyone know if it's possible to tell what voltage it is
using or change the voltage of the SD card in N900?


You aren't by any chance changing the cards by taking the back cover
out without powering off your device first?

Opening the back cover does an emergency shutdown on disks in case
user rips battery out next (that's apparently a common way to get
"phone not reachable" message back to your boss/wife/dog when they
call you, at least in some parts of the world).

If there were writes being done to the card at that time, it may
corrupt.  Power off your device first if you want to be sure you
can switch the card safely.

Also, the back cover has a magnetic latch that's used for detecting
when it's opened. If you have something magnetic next to your phone,
it may cause phone to think that back cover is being opened. See:
https://bugs.maemo.org/show_bug.cgi?id=8235#c15


- Eero


Or if there's some other explanation... maybe I have bad luck, maybe
the cards are bad but only the N900 can expose it.

dmesg shows things like this:

[33713.501464] mmcblk1: error -110 sending read/write command,
response 0x900, card status 0xe00
[33713.501495] mmcblk1: error -110 transferring data, sector 27271168,
nr 8, card status 0xc00
[33713.570129] end_request: I/O error, dev mmcblk1, sector 27271169
[33713.570159] Buffer I/O error on device mmcblk1p4, logical block 0
[33713.570159] lost page write due to I/O error on mmcblk1p4
[33754.895355] mmcblk1: error -110 transferring data, sector 30941184,
nr 16, card status 0xc00
[33754.895690] end_request: I/O error, dev mmcblk1, sector 30941185
[33754.895721] Buffer I/O error on device mmcblk1p4, logical block 458752
[33754.895751] lost page write due to I/O error on mmcblk1p4
[33754.895812] end_request: I/O error, dev mmcblk1, sector 30941192
[33754.895843] Buffer I/O error on device mmcblk1p4, logical block 458753
[33754.895843] lost page write due to I/O error on mmcblk1p4
[33755.504272] mmcblk1: error -110 transferring data, sector 31203328,
nr 16, card status 0xc00
[33755.504638] end_request: I/O error, dev mmcblk1, sector 31203329
[33755.504669] Buffer I/O error on device mmcblk1p4, logical block 491520
[33755.504699] lost page write due to I/O error on mmcblk1p4
[33755.504760] end_request: I/O error, dev mmcblk1, sector 31203336
[33755.504760] Buffer I/O error on device mmcblk1p4, logical block 491521
[33755.504791] lost page write due to I/O error on mmcblk1p4
[33756.204315] mmcblk1: error -110 sending read/write command,
response 0x900, card status 0xe00
[33756.204345] mmcblk1: error -110 transferring data, sector 31465472,
nr 16, card status 0xc00
[33756.268493] end_request: I/O error, dev mmcblk1, sector 31465473
[33756.268524] Buffer I/O error on device mmcblk1p4, logical block 524288
[33756.268554] lost page write due to I/O error on mmcblk1p4
[33756.268585] end_request: I/O error, dev mmcblk1, sector 31465480
[33756.268615] Buffer I/O error on device mmcblk1p4, logical block 524289
[33756.268615] lost page write due to I/O error on mmcblk1p4
[33756.968139] mmcblk1: error -110 sending read/write command,
response 0x900, card status 0xe00
[33756.968200] mmcblk1: error -110 transferring data, sector 31727616,
nr 16, card status 0xc00
[33757.027191] end_request: I/O error, dev mmcblk1, sector 31727617
[33757.027221] Buffer I/O error on device mmcblk1p4, logical block 557056
[33757.027252] lost page write due to I/O error on mmcblk1p4
[33757.027313] end_request: I/O error, dev mmcblk1, sector 31727624
[33757.027313] Buffer I/O error on device mmcblk1p4, logical block 557057
[33757.027343] lost page write due to I/O error on mmcblk1p4
[33757.727172] mmcblk1: error -110 sending read/write command,
response 0x900, card status 0xe00
[33757.727203] mmcblk1: error -110 transferring data, sector 31989760,
nr 16, card status 0xc00
[33757.786773] end_request: I/O error, dev mmcblk1, sector 31989761
[33757.786804] Buffer I/O error on device mmcblk1p4, logical block 589824
[33757.786834] lost page write due to I/O error on mmcblk1p4
[33757.786865] end_request: I/O error, dev mmcblk1, sector 31989768
[33758.486755] mmcblk1: error -110 sending read/write command,
response 0x900, card status 0xe00
[33758.486816] mmcblk1: error -110 transferring data, sector 32251904,
nr 16, card status 0xc00
[33758.549682] end_request: I/O error, dev mmcblk1, sector 32251905
[33758.549774] end_request: I/O error, dev mmcblk1, sector 32251912
___

Re: Is mail app using any rootfs space?

[Eero Tamminen]

Hi,

On 03/01/2011 04:45 PM, ext Francisco Diaz Trepat - gmail wrote:

Should I delete mails from my phone to free up rootfs?


AFAIK mail data should be in the home partition, not rootfs.


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

Re: Pulseausio allocating +40Mo of RAM ?

[Eero Tamminen]

Hi,

On 07/17/2010 08:03 AM, ext ri...@happyleptic.org wrote:

Hello !

I periodically run memstat on the N900 to check for apps that
could leak memory (in order to find a culprit for some
unresponsiveness issue), and I'm surprised to discover that
every time pulse audio is the process that uses the most memory.
See for instance here :

memstat -w | grep pulseaudio

   75848k: PID   768 (/usr/bin/pulseaudio)
  72k( 68k): /usr/bin/pulseaudio 768

So the text of the binary itself is rather small (libs are counted separately),
but the heap is 75M while the device is idle and not outputing anything.


Are you sure it's heap and not e.g. (almost completely unused)
thread stacks?


- Eero


Is it normal ?
Out of curiosity, what's the purpose of all this ram when idle ?
does pulseaudio keep some huge precalc datas of some sort ?
If not, wouldn't be better to just discard this RAM when its no more needed,
in order to prevent swapping, which is quite slow ?

What do you think ?

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

Re: Can I compile a new version of BusyBox?

[Eero Tamminen]

Hi,

ext Francisco Diaz Trepat - gmail wrote:

I wish to know if installing the new busybox would be possible and if I do
it would it brick my phone or do something to scripts or is completely
unknown and would be a nice thing to test?


Upgrading Busybox has typically causes some issues that need to be
fixed, but you can find them out only by testing it (obviously using
the current busybox config as base).

You should also use the Maemo busybox packaging so that you get all
the relevant symlinks packages.


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

Re: My N900 has diffmo as an essential package; safe to remove?

[Eero Tamminen]

Hi,

ext Marius Vollmer wrote:

ext Paul Hartman  writes:


So I suppose I have two questions:
1) What causes a package to become "essential"?


The maintainer of that package has decided to make it essential and puts
a "Essential: yes" field into debian/control.

For Maemo, packages often become "essential" by accident when packaging
bits are copied over from Debian.


Yes, "diff" is essential in Debian.

It's not an essential in Maemo.



2) Is it actually safe to remove diffmo? I don't want an unbootable
device or anything bad like that.


I assume that diffmo is a replacement for busybox /bin/diff.  If so,
then you only need to make sure that you always have a working
/bin/diff.  Thus, if after removing diffmo you end up without /bin/diff,
repair that immediately, maybe by symlinking it to busybox, or by
installing diffutils-gnu.


Default Busybox isn't built with "diff" utility (diff version in
Fremantle Busybox is too incompatible with real diff), so symlink
to that would be wrong.

If no package is depending on "diffmo", I think it's safe just to
remove it.



(If this is not about /bin/diff, then I apologize for the confusion.)



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

Re: Navigation software

[Eero Tamminen]

Hi,

ext Mayuresh wrote:

On Mon, Oct 11, 2010 at 03:15:48PM +0300, Eero Tamminen wrote:

Above is the most frustrating bug with N900 that I ever noticed. Just
can't use a navigation app continuously for hours during a journey.

Triggering the driver bug typically seems to indicate bad behavior in
the application, it's using huge amount of memory (e.g. leaking it)
and doing updates to its window although the window isn't even visible.


I have noticed that this bug triggers when the display is off (though the
app is in the foreground). By visible you meant the display being on or
the app being in foreground?


Application gets notifications whenever its window visibility
changes (and it also loses keyboard focus). If the display is off,
the application window is not visible.

With the "xresponse" utility from the SDK tools repository you can check
whether application does window updates.  Or you can build the latest
xresponse version from gitorious:
http://maemo.gitorious.org/maemo-tools/xresponse

Then just connect to the device through SSH and run:
DISPLAY=:0 xresponse -u -w 0 -a '*'

To monitor window updates.



Former is also a performance issue (device swaps more) and latter is
a use-time issue.

Once the apps are fixed to behave more properly, bug 9150 is much less
likely to happen (I encountered it myself last in 2009,


Sorry I remembered that wrong, I saw them also in April 2010 when
testing Maep more.



but I don't use the navigation part of the map applications).


Well, that may be right, though wonder whether so many applications have
that problem.


Because nobody's filed bugs[1] about their behavior?

I wonder whether community QA even does this kind[1] of checks...
(If not, they should)



The bug has been confirmed on modrana, mappero, maep and may
be a few others described on the bug thread.


Of those Maep is the only one I've used more.

[1] It should've been fixed already before summer:
https://garage.maemo.org/tracker/index.php?func=detail&aid=5515&group_id=1155&atid=4332
https://garage.maemo.org/tracker/index.php?func=detail&aid=5520&group_id=1155&atid=4332


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

Re: Navigation software

[Eero Tamminen]

Hi,

ext Mayuresh wrote:

On Fri, Oct 08, 2010 at 08:20:10PM +0200, Johann Spies wrote:

The next morning my N900 was unresponsive and I needed to switch it
off.  After a restart Modrana would not show any map.


This may be related to bug https://bugs.maemo.org/show_bug.cgi?id=9150

(That has nothing to do with downloading. Just leave ANY application that
accesses GPS running and you are likely to run into this.)


SGX drivers have improvements in the next release.



Above is the most frustrating bug with N900 that I ever noticed. Just
can't use a navigation app continuously for hours during a journey.


Triggering the driver bug typically seems to indicate bad behavior in
the application, it's using huge amount of memory (e.g. leaking it)
and doing updates to its window although the window isn't even visible.
Former is also a performance issue (device swaps more) and latter is
a use-time issue.

Once the apps are fixed to behave more properly, bug 9150 is much less
likely to happen (I encountered it myself last in 2009, but I don't use
the navigation part of the map applications).


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

Re: Desktop icons disappeared after reboot

[Eero Tamminen]

Hi,

ext Christoph Eckert wrote:

I'm running PR 1.2 on my N900. I'm mainly using it as a mapping device.

Yesterday I replaced the battery. After the reboot, all desktop icons 
(shortcuts, widgets) have disappeard except for some bookmarks I saved on the 
desktop. Further, I now cannot add any desktop icons. When I'm in editing mode 
and klick on the desktop menu, nothing happens. Additional reboots didn't cure 
the problem.

...
> additional info: The battery drains, as the following application is
> constantly eating 96% of my CPU and 11.4% of my memory:
> /usr/bin/hildon-home

To me this sounds like hildon-home gets completely non-responsive after
it has (mostly) started up.  Or do the bookmarks on Home still work?


> Pretty interesting.

One possibility is a bad applet.  If you just kill -9 hildon-home, it
should skip all 3rd party applets on startup.


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

Re: N810 slow since rebooting

[Eero Tamminen]

Hi,

ext Lucas Maneos wrote:

On Sun, Aug 15, 2010 at 02:27:29AM -0700, Andrew Daviel wrote:

  PID USER STATUS   VSZ  PPID %CPU %MEM COMMAND
 1303 user DW  128M   354  0.0103.4 alarmd

I cannot remember what is normal, e.g. if it is usual for alarmd to
show such a large virtual size.


Definitely abnormal, that's the size of the tablet's entire physical
RAM!


VSZ is VmSize i.e. virtual size, not physical RAM.
It matters only if it's actually used.

Best tool available for N810 for looking into whole system memory
usage is propably sp-smaps-measure/sp-smaps-visualize.  Install
those packages from the SDK tools repo[1] and then do:

* On device:
sp_smaps_snapshot > smaps.cap
* On Sbox:
sp_smaps_analyze smaps.cap
* On Desktop:
firefox smaps.html

N810 version of sp_smaps_analyze might still be the Python version,
not C-one like the Fremantle one, and the produced HTML data is pretty
large, that's why I think it's better done on (e.g. x86) Sbox.

The produced HTML tables are better viewed on Desktop, device screen
is a bit too small for them.

[1] http://maemo.org/development/tools/diablo/#apt-example
http://maemo.org/development/tools/doc/diablo/sp-smaps-measure/
http://maemo.org/development/tools/doc/diablo/sp-smaps-visualize/


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

Re: reclaiming space on root filesystem

[Eero Tamminen]

Hi,

ext John wrote:

The biggest culprit seems to be this file in the posix-locales
package:

$ du /usr/lib/locale/locale-archive
29068   /usr/lib/locale/locale-archive


Root file system (UBIFS) is compressed, whereas that's uncompressed
size.  To know compressed size, you can try what it's size would
be lzop'ed. (UBIFS uses the LZO compression algorithm)


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

Re: Fwd: N900 is hot when charged & sucks battery life after updating to PR1.2

[Eero Tamminen]

Hi,

ext khalid khan wrote:

I have attached the details that how it looks when i write top in X Terminal.


Strangely you had multiple MEM & CPU lines, did you try
to paste the output multiple times from a running "top"
program?


This output would also indicate that you did some extra
activity (hald, maemo-xinput & pulseaudio are active)
while top was monitoring the device:

CPU: 32.0% usr 12.7% sys 0.0% nice 54.9% idle 0.0% io 0.
Load average: 0.87 0.57 0.39
  PID  PPID USER STAT   RSS %MEM %CPU COMMAND
 2853   969 user R10296  4.1 16.2 /usr/bin/osso-xter
  797   659 root S <  16324  6.6 14.6 /usr/bin/Xorg -log
 1091   969 user S10484  4.2  8.2 /usr/bin/hildon-de
  746 1 pulseS <   5016  2.0  2.3 /usr/bin/pulseaudi
  842   769 root S  912  0.3  0.7 hald-addon-input:
 1083   969 user S12560  5.1  0.5 /usr/bin/hildon-st
 2857  2855 user R  736  0.3  0.5 top
   10 2 root SW   0  0.0  0.5 [omap2_mcspi]
  716   659 root S <   2244  0.9  0.3 /sbin/mce --force-
  928 1 user S 2384  0.9  0.2 /usr/bin/maemo-xin
-

It's better if you do it like this:
1. start "top"
2. wait 10-15 secs for things to settle down
3. press "q" to quit "top"
4. tap the arrow icon to change into terminal's
   copy paste selection mode
5. copy top output

step 3 above would make sure that your own copy-paste
activity doesn't anymore change what top shows.


Anyway, at least the above "top" output doesn't show any process
using all the CPU constantly.  Battery life can still be ruined
by processes that use even a small amount of CPU, *if* they do it
constantly (also when user isn't interacting with the device)
though.

Another reason for battery drain can be networking related
issues (processes that talk to network too often, don't handle
network disconnects/re-connects properly etc).


> Can over charging cause battery drain in 7-11 hrs?

Device should be protecting itself from over charging.  Once
the battery is full, it stops charging and waits until battery
charge has dropped a certain amount until it start charging
again.

Btw. Charging always warms the device (as does using the device).
Is the warming somehow exceptional after the PR1.2 update?


- Eero

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

Re: Fwd: N900 is hot when charged & sucks battery life after updating to PR1.2

[Eero Tamminen]

Hi,

ext khalid khan wrote:

Thanks for your reply.I typed " top " in X Terminal there are so many
processes running some are using less 10 - 30 % & few are using 70 - 80%
 & very hard to recognize the path & don't know exactly what are these
& how to close it.


"top" being started and running in the terminal naturally consumes
some resources from the system, but you don't need to care about that.

Just wait 10 secs after starting "top" for things to stabilize a bit
and then paste the output from top here so that people can comment
on what looks to be wrong.


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

Re: PR 1.2 update over the air

[Eero Tamminen]

Hi,

ext Ove Nordstrom wrote:

2010/5/26 Xavier Bestel mailto:xavier.bes...@free.fr>>
On Wed, 2010-05-26 at 17:59 +0300, Marius Gedminas wrote:

As usual, I didn't have enough free space on the rootfs, but the old
trick of disabling all application catalogues to free up the space
normally taken by apt package list caches worked fine.

The update felt rather slow (maybe ~30 minutes)


One of the slow things on SSU is optifying of the rootfs content.
If one instead flashes the rootfs, optifying happens on bootup.



and a bit scary (I was
afraid to plug in the USB cable to recharge during the process, probably
without any good reason).

All my contact icons on the desktop had disappeared after the first
reboot into the new OS.  They reappeared after the second boot.

Many of my contacts are duplicated now (apparently this has something to
do with Facebook chat support; I remember seeing a blog post about it).

The Extras Devel repository didn't work (404 not found errors) after I
reenabled it, until I changed the Distribution field to "fremantle" (it
was blank before).

AFAICS everything else works fine.


Did the same. It went rather well, but curiously when the device decided
to reboot by itself, when showing the blue-on-white Nokia logo, there
was also a sentence like "Device malfunction, will stop in 10s" in green
letters.


MALF sounds serious, it shouldn't happen with properly working device HW
(unless you had an issue on updating the bootloader/kernel/CMT or there
was some extraordinary condition like too hot device temperature, badly
connnected battery etc).



I re-powered it on and it seems it's all OK now.
Battery life hasn't improved much, it seems.


If you get MALF again on boot, I would suggest reflashing the device.


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

Re: Upgrade to PR 1.2 gone wrong

[Eero Tamminen]

Hi,

ext John Sullivan wrote:

Alejandro López  writes:


operation). When the moment for the reboot came, the problems started:
my N900 would start booting never asking for the security code, and
when starting to show the desktop, it would reboot; and the loop
started again. It was impossible to power it off except by removing
the battery.


I had the same experience, with the reboot loop at the same point, after
doing apt-get dist-upgrade (which I did because the app manager was
telling me I needed to use the proprietary Windows Nokia software
suite).


Did you have any packages installed which stupidly declare exact version
dependencies to other packages?  For example several of the games in
Extras seem to be declaring exact version dependency to SGX driver.
Which of course causes a conflict on SSU which contains a new & improved
SGX driver.


IMHO packages with exact version dependencies to anything that doesn't
come from the same source package should just be rejected from Extras.

Could somebody looking after Extras fix this so that we don't get such
bad packages at least for PR1.2?



I ended up flashing it, and restoring my settings/apps from backup, and
it seems to be working fine now.



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

Re: how memory management mechanism of maemo, where can I find information?

[Eero Tamminen]

Hi,

ext Rafael Rocha wrote:

i'm actually just doing a job in college
well, if I wanted to know where to find information about operation of the 
architecture, including memory management

ie wanted to know how to read the maemo maximize memory management and gain 
relative to symbian too.


Google for "Linux memory management".


- Eero


On Mon, Apr 26, 2010 at 3:59 AM, Andre Klapper 
mailto:aklap...@openismus.com>> wrote:
Hi,

Am Sonntag, den 25.04.2010, 23:46 -0300 schrieb Rafael Rocha:

hi everybody, I am now just studying how memory management mechanism
of maemo, where can I find information?


Can you please be a bit more specific, or describe what you plan to do?
Makes it more likely to get an answer...

Thanks,
andre

--
Andre Klapper (maemo.org bugmaster)

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




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

Re: MADDE?

[Eero Tamminen]

Hi,

ext André Hänsel wrote:

I'm new to Maemo development.

So could someone clarify what is the difference/relationship between MADDE
(http://wiki.maemo.org/MADDE) and the Maemo SDK
(http://maemo.org/development/sdks/).


Shortly MADDE is more for application development and SDK for other
kind of Maemo development too (e.g. porting of already existing open
source projects to Maemo).

Madde is more user-friendly, but newer so it may still have issues.

Maemo SDK works only on Linux (or Linux virtual machine), MADDE (also)
on other platforms.


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

Re: Dealing with N900 responsiveness (or lack of thereof)

[Eero Tamminen]

Hi,

ext Dawid Lorenz wrote:

To list pulseaudio clients, use "pactl list".

Strangely:

r...@n900:~# pactl list
-sh: pactl: not found


It comes from pulseaudio-utils package.


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

Re: Dealing with N900 responsiveness (or lack of thereof)

[Eero Tamminen]

Hi,

ext Michel Dänzer wrote:

So if you ever see pulseaudio using non-0 CPU when there's no sound
playing


That will be hard to check without using SSH, unless one
disables the touchscreen sounds... :-)


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

Re: Dealing with N900 responsiveness (or lack of thereof)

[Eero Tamminen]

Hi,

ext Dawid Lorenz wrote:

I've had a look at top over ssh connection and seen
/usr/bin/pulseaudio --system --high-priority constantly
floating around 1-3% of CPU time. Needless to say, device
was lying on the desk next to me, doing nothing, not to
mention anything media-related. I did killall media-player
but that didn't help and pulseaudio was still working out.

However I've been listening to music earlier in the morning,
took couple of calls etx.


Does "killall tonegend" help?

Then it's probably this:
https://bugs.maemo.org/show_bug.cgi?id=6868



Then later in the day I've opened
a camera to take a picture, which threw weird "Another
application is using audio" message at me few times (while
no app that I was aware of was running in the background).


To list pulseaudio clients, use "pactl list".


ext Jan Knutar wrote:
> I guess not even closed NEP was possible as we've heard nothing
> of it since ;-)

It's a possibility still.  Let's see.

In the meanwhile, there seems to be a 3rd party app that provides
at least some (less accurate & fine-grained) information, see:
http://talk.maemo.org/showthread.php?t=43748


ext Jan Knutar wrote:
> On Tuesday 23 February 2010, Dawid Lorenz wrote:
>> I just get readings from tools like htop or conky.
>
> If htop just uses the MemFree field from /proc/meminfo, then it's
> pretty much supposed to hover at over 90% used all the time. Unused
> RAM is wasted RAM. What matters is if the system is able to free up
> RAM when it's needed without excessive I/O traffic slowing things
> down..

This has some useful tools & scripts for counting memory usage:
  http://wiki.maemo.org/Documentation/devtools/maemo5/sp-memusage

(With some updates hopefully coming to them in next SDK tools update.)


But for tracking _changes_ in _all_ device resources, I recommend this:
  http://wiki.maemo.org/Documentation/devtools/maemo5/sp-endurance


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

Re: Not enough memory

[Eero Tamminen]

Hi,

ext Craig Woodward wrote:
 Eero Tamminen  wrote: 

Compression can make normal ASCII data into 1/3 of its size


Yes, wonderful.  Is a default N900 using a compressed filesystem?
NO. Why are we talking about it?  Moot point.


Err?

UBIFS used for rootfs is most certainly using compression.

For example the 20MB icon cache file (which is generated so it's
not in any package) that was earlier the largest issue for SSUs
(due to Gtk keeping old versions open) and which we removed for
PR1.1, took "only" 3MB of the rootfs space per instance.

Btw. In case you're interested, here are the technical details
which explain why even the file system itself cannot give very good
estimate on how much "free space" it has:
http://www.linux-mtd.infradead.org/doc/ubifs.html#L_spaceacc




And the issue with that is that package doesn't know into which Fremantle 
release it's going to be installed.


Doesn't the package manager send some kind of identifying info? 


Sorry, you lost me.

Where you're expecting this proposed package space estimation
to happen and when?  When creating a package?  Before installing it?
After it's installed?

If it's before install, what would provide that information to device?


[...]
it's not saying it will take 17,367,483 bytes; it says "17.4M". 


Which I think is too accurate number when the things I mentioned
are taken into account. :-)


[..]

I think it's a bit too confusing to have on the Application Manager UI.


I disagree.  It think it's absolutely needed in the package manager,
especially when the app manager is telling users "insufficient
rootfs space" when trying to do PR updates.


/home is larger, but it can also run out of space.

Root running out of space is more important though because if it becomes
full (so that even root cannot write anything there), the system may not
anymore boot up.  /home partition becoming full means that applications
may not behave correctly (tracker cannot index new files, so media
player doesn't show them, email cannot be fetched etc).



The app manager is the default tool they're going to use to install
and remove packages.  Knowing how much space a package takes up on
the rootfs is critical in selecting which packages to remove to make
space for such an update.  Removing a non-optified 8M package is
going to free a lot more space on the rootfs than removing an optified
one that shows 60M of usage.  But you have to be able to tell if it's
optified or not, which right now there's no way of telling.

>
Maybe there could be a separate 3rd party application 


That would be great too, but having it in the app manager would make
it that much easier.  It's one small piece of metadata, a few bytes
per package.


Well, you're free to suggest such an UI change, but I have my doubts
about how well less-technical users understand the distinction between
rootfs and /home etc.  I assume you're suggesting that these values
would be shown in the UI always?



Even if it's collected during install (doing a df
pre/post and storing that per block device), it's something.


That would include also their dependencies, but those could be
shared between later installed packages.



Nothing but the app manager can do that though, since it's the one
doing the installs.  And really, nobody wants to hop from one app
to the other to find/delete rootfs hogs.


And what about the non-UI packages that aren't shown in the Application
manager UI (for a good reason) and which also take space?

That hypothetical 3rd party application could show also those and
nothings preventing it from implementing package removal too.


- Eero

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

Re: Dealing with N900 responsiveness (or lack of thereof)

[Eero Tamminen]

Hi,

ext Dawid Lorenz (maemobile) wrote:

- Original message -

When this happens, do you see "SGX" mentioned in "dmesg" output?


Don't know, but I'll try hard to not forget to check that next time round.


If it's as low as on desktop, you can get hitches while using the
device.  Something below 100 may be better though.


Is echoing value to /proc/.../swappiness a right way of setting this at 
runtime? Is there any /etc/sysctl.conf equivalent where I could store this 
value for boottime?


The device itself sets swappiness from /etc/init.d/rcS file,
but I would say that in general one needs to be pretty sure
of his/her changes when modifying essential bootup scripts. :-)



I have swappiness of 60 right now, we'll see how it goes.



As to memory usage, there are e.g. tools like sp-endurance from
tracking resource usage changes in the device:
http://wiki.maemo.org/Documentation/devtools/maemo5/sp-endurance
http://wiki.maemo.org/Documentation/devtools/maemo5/sp-endurance-postproc

(First you collect the data and then you post-process it after
collecting.  You can run the python scripts for latter even on
the device.)


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

Re: Dealing with N900 responsiveness (or lack of thereof)

[Eero Tamminen]

Hi,

ext Dawid Lorenz wrote:

Is this only my N900 @PR1.1 or others also experience overall device sluggishness after 
~2-3 days of uptime? There is semi-identified problem with hildon-home hogging CPU time 
for few seconds on each wakeup from standby [1], however that's "fixable" by 
killall hildon-home and re-adding widgets to desktop. Most likely case is memory leak in 
one of the widgets on desktop, however that hasn't been determined yet.

Apart from hildon-home, I've noticed my N900 is still misbehaving after few days since 
last reboot. Namely, it becomes very unresponsive at times, to the point where -- for 
example -- I can hear incoming call ringtone, but screen doesn't really catch-up, so I 
don't even have opportunity to see who is calling, not to mention taking that call. This 
is obviously quite annoying, however I am very far away from taking "reboot your 
device every second day" advice as a solution.


When this happens, do you see "SGX" mentioned in "dmesg" output?



What I am suspecting here is swap (over)usage. It does only seem to grow over 
time, rather than going down when I close unused apps, for example. Physical 
RAM, on the other hand, usually sticks around 180-200MB of use, regardless of 
apps opened in the background. I've seen browserd takes quite a lot of memory 
resources, so I usually do /etc/init.d/tablet-browser-daemon.init restart, 
which seems to release some memory, but not for very long tough.

Other thing I've noticed is system swappiness value, which is 100 by default. What 
I've learned [2] is that 100 value favours moving stuff to swap space quite 
frequently, which makes some sort of sense with experience I've got. I have 
rebooted my N900 today and set swappiness value to 60 by echo 60 > 
/proc/sys/vm/swappiness, however I am not sure whether that's right way of setting 
this (article refers to /etc/sysctl.conf file which simply doesn't exists in my 
rootfs, or perhaps is stored elsewhere). Anyway, I am going to observe how system 
overall performance evolves over next couple of days.


If it's as low as on desktop, you can get hitches while using the 
device.  Something below 100 may be better though.





Does anyone experience similar sort of problem in his/hers N900 and has some thoughts on 
the topic? Any input would be appreciated. Apart from "reboot" solution, of 
course.


[1] https://bugs.maemo.org/show_bug.cgi?id=8723
[2] http://www.heysky.net/digest/2008/07/linux-memory-management.html

--
Dawid 'evad' Lorenz * http://adl.pl

null://real men never use Gmail




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

Re: Not enough memory

[Eero Tamminen]

Hi,

ext Craig Woodward wrote:

As for the rest, the tools certainly can know where
the files are going.  The paths are stored in the package,
and a simple test-extract will tell you the paths and rough
size of a file.


I think dpkg tools check the size before packaging the data,
whereas latest maemo-optify thing seems to be doing things in
postinst i.e. after installation (not just extraction).



A simple pearl script could be setup to
tally the counts based on the default N900 mounting layout,
build a percentage, and show how much is root vs opt vs ...

No, you can't show what it will take for a compressed file
system,


Compression can make normal ASCII data into 1/3 of its size,
binary data like libs, maybe to 2/3 of their size, already
compressed files (like PNGs and MP3s) aren't affected.

Removal of docs can have also a very large effect, but most
packages don't have that much docs compared to rest of the
package contents size.

If this kind of testing is done elsewhere than on device,
corresponding package files can be just compressed with LZO
to see their compressed size (on device the files will take
a bit more, but this is a good approximation).  Here's
even a script to do it:
https://bugs.maemo.org/attachment.cgi?id=1510



or one that's been linked differently.  But even
a rough idea for a *standard* N900 setup would be nice.


That can change between releases (I think think it will
in next release, more stuff moves to eMMC to make SSU easier).

And the issue with that is that package doesn't know into which
Fremantle release it's going to be installed.



If a package moves things or unpacks things during it's
install, well... that's unavoidable, and again is _already_
not showing up in the usage estimate we're getting now.

I think you're just taking this a little too literally.
Nobody is asking for exact numbers for all scenarios.
People just want a little more detail on how the existing
estimate breaks down when it comes to rootfs vs storage.
Nobody is expecting it to work to the byte, but for most
packages it's going to give a pretty good idea of how much
space it will take, and make it much clearer which packages
have been optified vs which ones have not.


If extras www-pages are going to have something like this, I would
recommend it to round the numbers at least to MBs and empathize that
it's an estimate...


As to SSUs, its the package manager that should give this kind
of information, but I think it's a bit too confusing to have
on the Application Manager UI.

Maybe there could be a separate 3rd party application for the device
which goes through all packages, checks where they've installed their
files & how large they're and shows this information sorted by the size
taken from different partitions so that user knows which packages to
uninstall to get most space.

But that doesn't help if the disk issue isn't from files installed by
packages, but them being open, or there being cache files (for apt
etc)...


- Eero


Fixing of this mess is hopefully something that can be backported
from Harmattan to Fremantle after Harmattan is released...

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

Re: Not enough memory

[Eero Tamminen]

Hi,

Hamalainen Kimmo (Nokia-D/Helsinki) wrote:

On Fri, 2010-02-19 at 11:02 +0100, Tamminen Eero (Nokia-D/Helsinki)

ext Jason wrote:

On a more technical, get-it-done approach, my problem with OOM was
too much crap in /var/cache/apt/archive/ .  There are two ways to
handle this in a more user friendly manner.  Instead of the OOM
error message, offer to run 'apt-get clean', and/or symlink
/var/cache/apt out to /home/.var.cache.apt .  Which I just tried,
and seems to wfm.

$ sudo gainroot
# cd /var/cache/
# mv apt /home/.var.cache.apt
# ln -sf ../../home/.var.cache.apt apt

AFAIK:

* Application manager doesn't have caches on rootfs, but on
   the 2GB partition.


Apparently it still has them on rootfs, see the internal bug 144371.


That bug is about using apt directly, not application manager.
It's the same as this public one:
https://bugs.maemo.org/show_bug.cgi?id=5746



I
don't see it necessary to keep these caches on rootfs since the .deb
cache is not mandatory and the rest (.bin files) can be regenerated
anytime.



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

Re: Not enough memory

[Eero Tamminen]

Hi,

ext David Greaves wrote:

ext Jan Knutar wrote:

Another nice feature would be if the application manager, or a web
interface somewhere, could tell you beforehand how much space an app
consumes on /...


It cannot, for several reasons.

The size for the package included to the binary package metadata
is determined by the Debian tools based on the space how much these
files take on the build file system which may be block based,
unlike UBIFS.  And more importantly, these tools don't or even
cannot take into account:
1. Where the files go, root or /home i.e. how much space package
   will take from which file system.
2. If it goes to a compressed file system (like root), what is
   the compressed size.  The compression level used when mounting
   the partition affects this too (which can be changed by root
   user).
3. Which of the installed files will be removed by apt-hooks on
   the device (e.g. docs...).  User can install also his own hooks.
4. Will the package itself do something that affects the space, like:
   - moving files from one partition to another
   - adding / generating new files
   - removing files

Even larger issue is with updates.  When updating a file that is
in use, file system cannot get rid of the old version of the file
until all of its users close the old file (which happens e.g. at
reboot).   And if the release image was compressed with different
compression level than what is used at run-time (may be lower for
performance reasons), even copying the same file in place may increase
disk usage.  Application manager doesn't have any reasonable way to
find out about this (and the comments above apply also to this
space).



That too would be useful.


Not as much as if the app manager didn't takes >10Mb of space for repo indexes


The files are fairly repeatable ASCII.  Rootfs compresses them pretty
well, you should check the lzo compressed size.


- Eero

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

Re: Not enough memory

[Eero Tamminen]

Hi,

ext Jason wrote:

On a more technical, get-it-done approach, my problem with OOM was
too much crap in /var/cache/apt/archive/ .  There are two ways to
handle this in a more user friendly manner.  Instead of the OOM
error message, offer to run 'apt-get clean', and/or symlink
/var/cache/apt out to /home/.var.cache.apt .  Which I just tried,
and seems to wfm.

$ sudo gainroot
# cd /var/cache/
# mv apt /home/.var.cache.apt
# ln -sf ../../home/.var.cache.apt apt


AFAIK:

* Application manager doesn't have caches on rootfs, but on
  the 2GB partition.

* However, if you use apt _directly_, you need to tell it to
  use 2GB partition for its caching like application manager does.

Marius can maybe confirm this.


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

Re: Memory Error using mircob

[Eero Tamminen]

Hi,

ext Edward Johns wrote:

Whenever I try to edit or add a bookmark in microb I get an error saying:

"Operation temporarily disabled due to low memory"

All I have to do is open the browser and try to edit an existing
bookmark; it fails even when it's the first thing I do after a reboot.
Immediately after reboot top shows I have around 315K free, after
starting the browser I have around 167K free.


If you press "M" in top so that it sorts by memory usage,
what it shows as using most memory?



My disk usage looks like this:

/: 42M free
/home:1.7G free
/home/user/MyDocs:26.4G

I guess root is looking a bit full (84%) but I wouldn't of said that
was too serious (or a likely cause?)


I guess the message is about RAM usage as partitions have enough space.



Before I start uninstalling stuff does anybody have any bright ideas?
Does anybody have bookmarks working under a similar kind of resource usage?
Is it possible to get microB to give me any log files?



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

Re: N900: Finding power suckers ?

[Eero Tamminen]

Hi,

ext Xavier Bestel wrote:

Now, I don't think spending 8hrs sleeping should have emptied the
battery (correct me if I'm wrong), so I'd like to have a look at which
application kept it working. What's the best method ?


Htop & strace.

Especially the Time column in Htop is nice. If something would be using
so little enough CPU that its CPU usage shows as 0% in htop, the Time
column changes tell whether it wakes up or not.


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

Re: X forwarding on Maemo 5 (N900 is the ssh server)

[Eero Tamminen]

Hi,

ext micu wrote:
can anyone here tell me, how to establish X forwarding on the N900? Meaning: I 
see my N900/Maemo apps on my GNU/Linux machine forwarded to it.


I can, of course, connect to my N900 via ssh (root on the N900)—but no X 
forwarding.


Having the possbility to connect as a user via ssh would be a good idea for 
this‽


Any hints?


Ssh has options to give more information about the problems (-v -v -v).

I think the issue is xhost missing from the device.


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

Re: Unable to upgrade maemo 5 44-1 because of not enough disk space

[Eero Tamminen]

Hi,

ext Johan Helsingius wrote:

Reboot first.  Ubifs runs its garbage collector on boot and sometimes
manages to free up amazing amounts of space.


That seems to have done it. Many thanks!

Is there any way to run the garbage collector without booting?


"sync".

But I think the issue was applications keeping old versions
of replaced files still open (there's a bug in bugs.maemo.org
about this and the large Gtk icon cache files which was fixed
in 44-1).

After being terminated on shutdown, those old files cannot
be anymore kept open by apps and file system can free
the space taken by them.


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

Re: N900 memory usage

[Eero Tamminen]

Hi,

ext Kevin Kempter wrote:
> I've bought a 16G SD card for my N900. When I go to Settings --> Memory I see 
> this:
> 
> Nokia N900:
> 21.13 GB Available
> 
> Memory Card:
> 11.10 GB Available
> 
> Memory for installable applications
> 1.82 GB available
> 
> 
> Can anyone tell me what the  "Memory for installable applications" is used 
> for 
> and why do I have only 1.82 GB available? Is this a normal scenario?

Yes.  What you were thinking of installing that would fill that?

The 1.82 GB is actually the amount reserved for the /home partition
which includes user settings, mails, metadata indexer database,
thumbnails etc.  If that space becomes full, many things in the device
stop working normally (fetching mails, indexing files etc), so you
should make sure to leave there enough space.

Normally (pre-installed) packages are installed on the rootfs which has
much less space available, but 3rd party applications are recommended to
be packaged such that they install (most of) their files to the larger
/home partition (using the maemo optify helper) instead of to rootfs.

Rootfs is 256MB, is compressed, faster (has different flash memory
and file system) than the eMMC on which /home is mounted. But it's
almost full (<100MB space free for non-compressed data, more for
compressable data).


So as a summary, depending on how the applications are packaged,
you may actually have less space available.  Package manager
doesn't know how the installable packages are packaged and
the Settings->Memory doesn't list free space on rootfs, you
would need to use "df" in terminal to know better which is
of course less than ideal.  In Harmattan this is hopefully handled
much better (no need for the optify hack to have more space for
applications than is available on rootfs).


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

Re: N900 - Problems with builtin apps

[Eero Tamminen]

Hi,

ext Tim Ashman wrote:
> 6. Application List. - On the N810 and below I was able to take all of the 
> nokia apps that I would never use put then in the lowest menu option and 
> forget about them.  Not as good as being able to uninstall but It was ok.  On 
> the N900 I haven't figured out how to rearrange the app menu.  All of the 
> installed apps that I use regularly are on the second page at the bottom and 
> all of the items or apps that I will never use are front and center.  It sure 
> would be nice to be able to move these icons around.  I'm hoping I've just 
> missed something.

You can create shortcuts to the applications to the desktop.
The good point is that besides applications you can have there
also other stuff like URLs, applets...

I.e. when no windows are open, instead of doing:
1. tap to task launcher icon
2. tap "More..."
3. Pan down
4. Tap to start app

You just tap to the application icon on desktop.

(When you have windows open and you're not on desktop, it doesn't save
quite as many taps, but is still slightly more convenient.)


Just long press on empty Desktop area and select "Add shortcut"
from the Desktop menu.  URL shortcuts you can add to desktop from
the Browser (it asks where you want the bookmark to be added,
bookmarks list or desktop).


> 7. Web Addons - What exactly is "single sign on for ovi websites" and why 
> can't I uninstall it from the browser.  Likewise what is the Atlas Plugin?

Atlas is extension used by Maps (Maps uses browser engine).


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

Re: Modest doesn't update one IMAP account anymore

[Eero Tamminen]

Hi,

ext Keywan Najafi Tonekaboni wrote:
> Am Donnerstag, den 12.11.2009, 13:48 +0200 schrieb Kimmo Hämäläinen:
> 
>> I had some problems with Modest before as well, but looking at the
>> syslog (/var/log/syslog) helped me. If you don't have that file, you
>> might want to "apt-get install klogd sysklogd" and check the log when
>> Modest tries to connect.
> 
> Thanks for the hint. Unfortunately I couldn't find anything useful,
> because modest didn't send anything to syslog. Do I need to reboot the
> device or something else? Or should modest just send messages their out
> of the box?
> 
> I have also a question regarding syslog. Do I need to clean up syslogs
> by my own or changing settings, so the root partition won't run full?

You need to clean the logs yourself, but they should be rotated on
reboot, if they're too large.

NOTE: As syslog is run as root, it can fill the rootfs completely
(some space is reserved for root only).  If that happens, the device
may not boot up and you need to reflash it.
-> I would run syslog only when directly debugging something


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

Re: N900 versus N810

[Eero Tamminen]

Hi,

ext Marius Gedminas wrote:
> On Mon, Oct 26, 2009 at 01:40:17PM -0400, R. A. Bilonick wrote:
>> I see that Nokia advertises the N900 as a mobile computer, not primarily
>> as a phone. So I was wondering how the N900 compares to N810 in terms of
>> processing speed, wireless connection speed, and so forth, in actual
>> use.
> 
> After using a N900 for two days, I didn't want to even touch my old
> N810.

I still use N810 for browsing lightweight www-sites (like the great
Linux Weekly News lwn.net one) because it has larger screen and for
some other things where larger screen matters less than performance
(my eyes are getting old...).  It's also still OK as an extra
internet radio / mobile news video player.

N810 having more keys (4 rows instead of 3) also compensates somewhat
for the otherwise noticeably better N900 keyboard.

Otherwise, there's no competition.


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

Re: N900 battery duration

[Eero Tamminen]

Hi,

ext Tuomas Kulve wrote:
> Tuomas Kulve wrote:
>> igor.sto...@nokia.com wrote:
>>> There is a bug in the cellmo SW related to absence of SIM: the current is 
>>> higher when the SIM is not in.
>>> So in real life use case (i assume people who will buy it will actually use 
>>> it as a phone) the use time will be longer.
>> I guess I need to do more testing then :)
>>
>> Although I'm not sure how to accomplish that as I need my SIM for the
>> daily use..
> 
> The previous test was without a SIM card. The device was very idle and
> run for 93.75 hours (with a guessed few hour error marginal).
> 
> After charging the battery I inserted the SIM card, rebooted and started
> a new test.

Why reboot?

I would expect most people just to charge without shutting down
the device.

(You may remember that e.g. N8x0 use less battery if it's put to
offline-mode instead of being shut down for the night.)


> I received two calls (hung up immediately), I received two
> SMS messages and sent one, and the notifier led was blinking for 45 mins
> at one point. Otherwise the device was idle and run for 93.62 hours. So
> pretty close to the test without a SIM card.
> 
> I have a small shell script running there that writes a log every 60
> seconds. After writing the log it runs the "sync" command and starts the
> loop again. I'm assuming it doesn't affect the total run time much in an
> otherwise idle device.


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

Re: N900 battery duration

[Eero Tamminen]

Hi,

ext Andrea Grandi wrote:
 Do you have RD-mode or RD-flags enabled?
>>> how can I know it? I just installed the utility to become root (sudo
>>> gainroot)
>> cal-tool --get-rd-mode --get-rd-flags
> 
> it replies: disabled
> 
> by the way: I've seen that connecting to Simyo UMTS/HSDPA (which is
> ALWAYS in roaming, since it's a virtual operator and it uses Orange
> network) it takes a lot of battery. If I keep connected to my home's
> wifi, battery last much much more.
> 
> I'll be able to test with Italian 3Hg carrier once I'll come back in
> Italy... but this will be for Christmas holiday :\

Phone side needs much longer idle network traffic periods than WLAN,
to be able to power down.  It's best if you can track the network
traffic to find out whether something causes it to be always up
(some process in the device[1], traffic in the phone network).


- Eero

[1] This could help, the Debian package builds fine for the device:
http://packages.debian.org/squeeze/nethogs
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: N900 battery duration

[Eero Tamminen]

Hi,

ext Andrea Grandi wrote:
>> Do you have RD-mode or RD-flags enabled?
> 
> how can I know it? I just installed the utility to become root (sudo gainroot)

cal-tool --get-rd-mode --get-rd-flags


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

Re: N900 battery duration

[Eero Tamminen]

Hi,

ext Andrea Grandi wrote:
> 2009/10/17 Felipe Contreras :
>> On Fri, Oct 16, 2009 at 10:18 PM, Andrea Grandi  wrote:
>>> 2009/10/16 Andrew Flegg :
 What widgets do you have on the desktop? RSS? Facebook? Foreca?
>>> I've Facebook, Calendar, Forecast and two shortcut to start Phone and 
>>> Contacts.
>> Try removing the third party ones: facebook and forecast.
> 
> This test was done with these situation: connected to home wifi,
> facebook ON, forecast ON, pressing sometimes the main menu to keep the
> light always ON ---> http://pastebin.ca/1625155
> 
> This other test is the same as previous but without facebook and
> forecast: http://pastebin.ca/1625163

Do you have RD-mode or RD-flags enabled?

Are you charging your device until the led shows green?


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

Re: 2Gb internal storage on N810

[Eero Tamminen]

Hi,

ext Frantisek Dufka wrote:
>> If you want swap, I would suggest a fast SD card with a separate
>> _partition_ for swap
> 
> AFAIK since version 2.4 of linux kernel, swapping to file no longer goes 
> via filesystem code at all and speed is similar/same to swapping to 
> block device directly. Kernel swapping code makes mapping for blocks 
> belonging to swap file on the beginning and then uses underlying block 
> device directly (both for speed and for deadlock prevention). If the 
> file is not heavily fragmented speed should be the same. Fragmentation 
> is visible in kernel log when swap file is enabled (number of extents).

I think the file system (especially one as trivial as FAT) overhead is
trivial compared the Flash writing speed of the card.


> As for corruption I'm not sure

At least there were much more corruption reports from people who
had enabled swap than from ones that hadn't.  (It's possible that
they are heavier users and therefore have more changes to mess
things up with accidental USB cable removal, taking card out while
it's still being used etc.  But there were no re-producible use-cases
for verifying these things.)


 > but I'd say it is not an issue too since file is already allocated
 > and filesystem code is not used when swapping.


Whenever you do a write to a Flash based media, the data is written
somewhere else and the old block is GCed (with the HW API provided
by the memory cards kernel just doesn't see it).  Also, I think
the Flash blocks could be larger than the ones used by kernel
internally for swap (4kB?) or the VFAT allocation units, or they might
not be aligned?

So... It's possible that swap can affect the contents of rest of
the VFAT, especially if the swap file was made on a fragmented VFAT.



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

Re: 2Gb internal storage on N810

[Eero Tamminen]

Hi,

ext Alexandru Cardaniuc wrote:
>> While some express concern/fear about memory errors on flash cards,
>> there seems to be little mention of such problems from n800 users who
>> have been using replaceable flash cards for several years. So while
>> there probably is some wall when these things may start to happen for
>> n810s, it's still a bit out in the future. For n800s users, there's
>> the simple fix of swapping a new card.
> 
> Easy fix for replaceable cards, but how about built-in 256MB flash
> memory?

As far as I remember correctly, on average it has enough write cycles
that you should be able to write to it *constantly* for tens of years.
I doubt the rest of device will last that long. :-)

Except for RSS, email etc refreshes which results are stored under
/home/ and operations for files on /home/MyDocs, there shouldn't be
much writes to it though.  If you have Flash enabled in Browser, but
no memory card, the video streaming is buffered on that too.

Real reason for avoiding frequent writes / disk churn on rootfs is that
this would  trigger JFFS2 garbage collecting much more frequently which
can *really* slow down your device:
https://bugs.maemo.org/show_bug.cgi?id=2615


The internal 2GB memory card in N810 has less write cycles than
the roots.  Because it's larger, it can spread the writes over more
blocks i.e. wearing shouldn't be a problem for that either.

Btw. Personally, I wouldn't enable swap on it (from Control panel)
though, at least constantly, for following reasons (ordered from more
important to less important):
1. with constant swap file writes, it's much easier to
accidentally cause the card FAT file system to get corrupted
2. there's a performance slowdown (although swap allows running
some use-cases that wouldn't be otherwise possible)
(3. Flash wearing)

If you want swap, I would suggest a fast SD card with a separate
_partition_ for swap and some hacking on the device startup scripts
to support this.


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

Re: Emacs

[Eero Tamminen]

Hi,

ext Peter Flynn wrote:
> Peter Flynn wrote:
>> Having fixed the ncurses-base dependency, I tried to install with
>>
>>> # mkdir /media/mmc1/emacs
>>> # dpkg --instdir=/media/mmc1/emacs -i emacs_22.1-1_armel.deb
>>> (Reading database ... 18457 files and directories currently installed.)
>>> Unpacking emacs (from emacs_22.1-1_armel.deb) ...
>>> dpkg: error processing emacs_22.1-1_armel.deb (--install):
>>>  error setting ownership of `./usr': Operation not permitted
>>> dpkg-deb: subprocess paste killed by signal (Broken pipe)
>>> Errors were encountered while processing:
>>>  emacs_22.1-1_armel.deb
>>> # ls -ld /media/mmc1/emacs/
>>> drwxrwxrwx  2 user root  32768 Jun 21 16:09 /media/mmc1/emacs/
>>> # 
>> Why should it have a problem with ownership when I'm installing as root?
> 
> Turns out that OS2008 allows the cards to be mounted as user.
> umounted mmc1 and remounted as root, and it appeared to start 
> installing, but now it's gagging on
> 
>> (Reading database ... 18457 files and directories currently installed.)
>> Unpacking emacs (from emacs_22.1-1_armel.deb) ...
>> dpkg: error processing emacs_22.1-1_armel.deb (--install):
>>  error creating hard link `./usr/bin/emacs-22.1': Operation not permitted

VFAT is lacking many features of real file system. VFAT is used
on MMC/SD cards by default for PC i.e. Windows/OSX/etc compatibility
reasons.

You need to use some other file system (it's a bit complicated,
is there a Wiki page on using other file systems on the MMC/SD
cards?).


> and it doesn't appear to have written any files (maybe it wipes all 
> traces if it aborts?)

Dpkg must remove aborted installations, otherwise you would soon
run out of disk space on rootfs even for root (and as a result device
would be in reboot loop if you'd reboot it at that state).


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

Re: ncurses

[Eero Tamminen]

Hi,

ext Peter Flynn wrote:
> I've been rearranging my SD cards and I wanted to install Emacs_22.1-1 
> (this is for OS-2008) to replace the older version.
> 
> It's looking for ncurses-base and libxaw7 but I can't find any that 
> install.

The package has or hasn't a dependency for ncurses-base?

(If you take & build Emacs package sources directly from
Debian (or Ubuntu), it won't have ncurses-base dependency
because in Debian that's an Essential unlike on Maemo.)


> I have the maemo-Extras repo installed, and I can see 
> ncurses-base in there, but it doesn't appear in the App Manager. If I 
> download ncurses-base-5.4-3.osso1.deb and try to install it manually, I 
> get the error Incompatible application package.
> 
> Is there a working .deb for OS2008 for these packages?

I think only thing needed is the Emacs package having an explicit
dependency to the ncurses-base package, you shouldn't need to
install it manually.  Ncurses-base dependency works at least for
the SDK tools (which are in a repository of their own).


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

Re: Where is telnet?

[Eero Tamminen]

Hi,

ext Matan Ziv-Av wrote:
>> Telnet isn't something that's either:
>> - Needed by the device itself
>> - An essential (needed in installing most of Debian packages
>>  without them declaring a separate dependency)
> 
> The same is true for chvt, netstat, uniq, and probably many others, yet 
> they were included.

chvt: needed by device bootup scripts
uniq: essential (also POSIX standard utility)
netstat: legacy. we've considered removing it (along with some other
  legacy non-essentials in Bysybox), but our own developers use
  it often[1] and removing it from public releases doesn't really
  save that much space

The reason why I talk about our own developers is that their situation
is very different from external application developers & (advanced)
users.

Our own developers may need to flash new releases many times
a day and need to debug situations where the new internal release
cannot even connect to network at bootup.  I.e. installing additional
packages from repositories isn't really an option.

However, the published releases don't have anymore these kind of issues
and releases happen only couple of times per year, so for end-users
installing tools from a repository after downloading & flashing
a new release is completely valid option and a right thing to do.


> The decision to not include telnet was a bad 
> decision. Admitting to mistakes and fixing them (even if only in 
> Fremantle) is better than attacking someone who expects to have telnet 
> on his internet tablet.

If my reply seemed like an attack, I apologize, it wasn't meant as such.


>> It's an (advanced) end-user tool and can be installed separately.
>>
>> Have you tested that Busybox telnet even works as well as real telnet?
>> With the past experience I have some doubts about the quality of misc
>> Busybox tools.  :-)
> 
> Yes, I have. And even if it does not, it surely does better than no 
> telnet at all.

The problem of adding new things to Busybox is that:
- they then conflict with the real alternatives
- they increase our essential/base set (busybox is essential so
   anything added to it becomes then an essential too)

That's not good from the package management point of view.
See e.g. bug:
https://bugs.maemo.org/show_bug.cgi?id=2896

This has been discussed quite extensively on this list and in
bugzilla.  I hope we have a solution for this in Fremantle.


> I wonder what kind of QA did Nokia run on busybox's top or nslookup (for 
> example) before deciding to include them?

Busybox top is (still[1]) crap and is problematic because it conflicts
with procps that provides much nicer top and also many nice other tools.
However, (some kind of) top being pre-installed on the device is
really a must for us & useful for all developers and many advanced
users.

nslookup removal has been considered but hasn't been done so far.


[1] Despite the multiple rewrites on the code I originally sent
 to Busybox authors. :-)


>>> Life with the tablet could be a bit simpler if Nokia chose to include more 
>>> stuff that is available in busybox (md5sum, cpio, hexdump, ...).

md5sum is an essential + there's a bug about it missing in
bugs.maemo.org.  As a result it's included into Fremantle busybox
(along with quite many other utilities belonging to Debian essential
packages Busybox claims to be providing).

cpio & hexdump aren't essential (in the upstream compatibility
/ Debian packaging sense), so they can be installed separately.
"cpio" is in Debian in "cpio" and "hexdump" in "bsdmainutils"
package.


>> Some of them will be included into Fremantle; the ones that are in
>> Debian in packages that Busybox currently claims to provide and
>> conflicts with and which provide options compatible with GNU tools.
>> Other tools belong to somewhere else.
> 
> telnet belongs on an "Interner Tablet", and claiming otherwise is akin 
> to claiming that the WWW is the internet.

Device has application manager & apt-get.  Claiming that in
an "Internet" device everything needs to be pre-installed into
the device itself is so mid-90's.  :-)


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

Re: Where is telnet?

[Eero Tamminen]

Hi,

ext Matan Ziv-Av wrote:
>>> Unfortunately, I don't have it on my N810 and can't find a copy of it.  And
>>> the times I need it are when travelling and so I have not got around to
>>> porting it myself!

Have you tried whether "porting" it require anything else besides
downloading the sources from:
http://packages.debian.org/lenny/telnet
And doing "dpkg-buildpackage" in Sbox (after apt-getting
the build deps)?  It doesn't seem to have any funny deps.


>> Actually I would like to encourage you or someone else to port it if you 
>> need it.
>> We are very busy here and can't port every app you would like to have, so
>> if you really want it, please consider doing it by yourself.
> 
> This is very funny considering that telnet is available in busybox, so 
> _not_ having it installed by default is a conscious decision by Nokia. 

Telnet isn't something that's either:
- Needed by the device itself
- An essential (needed in installing most of Debian packages
   without them declaring a separate dependency)

It's an (advanced) end-user tool and can be installed separately.

Have you tested that Busybox telnet even works as well as real telnet?
With the past experience I have some doubts about the quality of misc
Busybox tools.  :-)


> Life with the tablet could be a bit simpler if Nokia chose to include 
> more stuff that is available in busybox (md5sum, cpio, hexdump, ...).

Some of them will be included into Fremantle; the ones that are in
Debian in packages that Busybox currently claims to provide and
conflicts with and which provide options compatible with GNU tools.
Other tools belong to somewhere else.


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

Re: xresponse / cnee question re: mimic of multi-key event

[Eero Tamminen]

Hi,

ext Burke, James wrote:
> Is there a way to use xresponse (or cnee) to mimic a multi-key event such as 
> Alt+F10 ?

I haven't tried cnee, but I think it should be able to record Alt+F10
(and then replay it).

With Xresponse it would go something like this:
xresponse -k "ISO_Level3_Shift,2000" &
xresponse -w 2 -k "F10"

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

Re: Browser Reload Tabs app?

[Eero Tamminen]

Hi,

ext Randall wrote:
> As my N800 locks up sometimes, 2-3 times a day (I've got nothing 
> running on it but the latest OS) is there a proggy that will 
> re-open all my open Browser windows when I reboot the browser - ala 
> Firefox?
 If you press the power button, do you get the power menu?
>>> Yes and I can lock the keys, and then unlock, but that only sometimes 
>>> kicks the unit back into gear.
>> So the device hasn't locked up, power key works and you can even
>> interact with the power menu?
>>
>> Is the use-case for the lockup following:
>> - You have a Browser page open with Flash video playing in it
>> - You switch away from Browser (to Desktop or another application)
>> - screen blanks while the Flash is still playing on the background,
>> - video finishes playing before you tap on the screen to wake it up
>> ?
>>
>> (If yes, this is NB#83173

(sorry, NB#85766)

>> which doesn't seem to be reported at 
>> bugs.maemo.org.  Could you then report your use-case there and put
>> me on CC?)
>>
> You are correct

To workaround the bug, avoid leaving Flash video playing on
the background.  (this way you also drain battery less)


It would seem to be a bug in Browser/Flashplayer, it has somehow
corrupted its ESD output (writes less to ESD socket than it should).

Because of this, ESD sound daemon starts busylooping, which will stop
all other clients that try to output any sounds (including Desktop).
This behavior may be caused by a "bugfix" we've inherited from
the upstream ESD version:
   https://bugzilla.redhat.com/show_bug.cgi?id=55877


To unfreeze the device, you need to do in (e.g. SSH) console:
killall esd

Even if ESD would work around[1] this issue, Browser would still be
unusable.  To get Browser working again you need to do:
killall browserd

(killing the daemons should get them restarted.)


> - BUT I can't get this posted on bugs - no matter what I do.

Sorry, what do you mean?


> You have a bug number - what's with that?

I found it from our internal bug tracker, but this issue doesn't
seem to be publicly reported.  That's why I asked you to report it
to bugs.maemo.org (the public/community bug tracker) so that somebody
could e.g. fix it for Diablo community updates (ESD isn't used in
Fremantle).

I could then attach a potential patch to fix the ESD issue to that bug.


 (I.e. is something just grabbing the input instead the device 
 freezing.)
>>> No.
 Do you have swap enabled?  If yes, have you checked that the memory 
 card FAT file system hasn't corrupted?
>>> Yes it is and no it's not (I like those answers).
>> Good. :-)


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

Re: Browser Reload Tabs app?

[Eero Tamminen]

Hi,

ext Randall wrote:
>>> As my N800 locks up sometimes, 2-3 times a day (I've got nothing 
>>> running on it but the latest OS) is there a proggy that will re-open 
>>> all my open Browser windows when I reboot the browser - ala Firefox?
>> If you press the power button, do you get the power menu?
 >
> Yes and I can lock the keys, and then unlock, but that only sometimes 
> kicks the unit back into gear.

So the device hasn't locked up, power key works and you can even
interact with the power menu?

Is the use-case for the lockup following:
- You have a Browser page open with Flash video playing in it
- You switch away from Browser (to Desktop or another application)
- screen blanks while the Flash is still playing on the background,
- video finishes playing before you tap on the screen to wake it up
?

(If yes, this is NB#83173 which doesn't seem to be reported at 
bugs.maemo.org.  Could you then report your use-case there and put
me on CC?)


>> (I.e. is something just grabbing the input instead the device freezing.)
> No.
>>
>> Do you have swap enabled?  If yes, have you checked that the memory 
>> card FAT file system hasn't corrupted?
> Yes it is and no it's not (I like those answers).

Good. :-)


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Alejandro López wrote:
> Eero Tamminen escribió:
>> Skype video requires significantly more power than for example Gtalk
>> video (which the device supports) and has quite strict latency
>> requirements (the call drops if Skype doesn't get enough CPU).
> 
> Great! Now I know the technical reasons, but I still don't know
> why this is not mentioned when it is said that Skype runs on
> the tablet.

I don't know about the others (or the device marketing), but
at least to me the Skype video use is very marginal even on Desktop
at home.  In the beginning the video call was a nice novelty, but
nowadays we enable it only sometimes when discussing with our young
nieces, for 99%[1] of the calls voice is enough.

As this thread is "Nokia device usage", what you would use
the Skype video for? :-)


- Eero

[1] of the _time_ used for all the calls video part might be more,
 the calls with our nieces are usually quite long as they show
 what they've done in pre-school etc...
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: Browser Reload Tabs app?

[Eero Tamminen]

Hi,

ext Randall wrote:
> As my N800 locks up sometimes, 2-3 times a day (I've got nothing running 
> on it but the latest OS) is there a proggy that will re-open all my open 
> Browser windows when I reboot the browser - ala Firefox?

If you press the power button, do you get the power menu?

(I.e. is something just grabbing the input instead the device freezing.)


Do you have swap enabled?  If yes, have you checked that the memory
card FAT file system hasn't corrupted?


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

Re: Installing packages onto /media/...

[Eero Tamminen]

Hi,

ext Peter Flynn wrote:
> I have a nicely-running N800, but the apps I use are taking up most of 
> the internal space. I want to add a couple of big ones (TeX is one) and 
> I have plenty of space on SD cards. Is there a command for package 
> installation that says, in effect, "install this all in /media/foo/bar 
> and symlink it back to where you would normally have installed it"? Or 
> some other way? I guess if I knew how to view the inside of a package I 
> could work out where all the big stuff is expecting to go, and create 
> symlinks in those places to the SD cards...or would that make the 
> installer spit blood when it finds that symlinks exist where it expected 
> to create new directories?

"Easiest" way is probably something like following as root
(NOT TESTED!):
   # cp -a /usr/share/ /media/mmc1/
   # cd /usr
   # mv share share.old
   # ln -s /media/mmc1/share share
   (# rm -r share.old)

(Depending on which memory card you want to use, you might change
mmc1 to mmc2.)

Memory cards have VFAT file system which doesn't support all
the required file attributes, so they're mounted so that you cannot
execute programs from them, but /usr/share shouldn't have any
executables.  The data files should take most of the space in
the packages, so moving /usr/share on to memory card probably
helps most.


***NOTE***: if/when your memory VFAT file system card corrupts,
your device will not anymore boot!

That's why it's instead recommended booting the whole system from
a memory card that has been formatted with a file system that doesn't
corrupt as easily as VFAT (needed for Windows/OSX compatibility)
and supports all the necessary file attributes.


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Mark wrote:
>> Do you have some (e.g. 3rd party app) running in the background
>> taking CPU?  Or something that frequently polls network
>> (Do you switch it to offline mode when you don't use it)?
>> Or doesn't allow the display to blank when you don't use the device?
> 
> I can't speak to the 3rd party app issue; I have so many apps
> installed that that is entirely possible, although according to
> LoadApplet there's nothing going on.

Program doesn't need to show up in "top" at all to decrease the device
use-time from a week to a day, for that it's enough for a single process
to wake up once a sec and do nothing else (i.e. <1% CPU use).
It prevents CPU from sleeping.


> I rarely switch to offline mode
> because much of the stuff I do with it requires network access, and
> even when I'm not doing anything constantly online there's stuff like
> OMWeather that needs to update every now and then. I *never* turn off
> bluetooth because it's too much of a pain digging through all those
> menus to get it back on again.

Ok, WLAN + Bluetooth and some apps using net frequently (say at one
minute interval or more frequently) could start explaining the use-time.


> I do have the display set to blank at 2
> minutes, and only stay on when charging. I also have the auto-lock
> enabled.


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Alejandro López wrote:
>> To assess whether this was a realistic expectation, were there other
>> mobile devices[1] which provided Skype *video*calls when you bought
>> the device?
>>
>> [1] mobile = ones that are same size or smaller, see my earlier
>> mail about power consumption & heat dissipation.
> 
> What I mean is that the N810:
> - was announced with Skype running on it,
> - at that time Skype had videoconf running on linux for some time,

On Desktop.  PC != mobile device.

> - the N810 run linux,
> - the device has a webcam.
> 
> With all this in mind, although it is not explicitly said (I accept!),
> how can you imagine that there is no videoconf? Everything is/was
> there to have it!

Skype video requires significantly more power than for example Gtalk
video (which the device supports) and has quite strict latency
requirements (the call drops if Skype doesn't get enough CPU).


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Alejandro López wrote:
>> Agreed, and fully understandable. Can we draw up a list of what -
>> exactly - the N8x0 fails to do out-of-the-box which it is advertised
>> it *can* do; and requires hacker-like skills to enable?
> 
> One thing that deceived me is that Skype was announced as and
> important application. Although it is true that Skype works,
> it's a rather old version (1.7) that doesn't support video
> conference. As the device includes a webcam, I was expecting
> Skype to be able to use it. In my particular case, videoconf
> is the only usage I have for Skype.

To assess whether this was a realistic expectation, were there other
mobile devices[1] which provided Skype *video*calls when you bought
the device?

[1] mobile = ones that are same size or smaller, see my earlier
 mail about power consumption & heat dissipation.


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Mark wrote:
> On Fri, Mar 6, 2009 at 9:16 AM, gary liquid  wrote:
>> the bike comes with a subscription to a community repair shop and new
>> elements are being created for it every day.
> 
> Provided you *want* all kinds of addons you will never need...
> 
>> it includes out of the box an expansive toolkit able to mend and repair
>> anything on your bike.
> 
> As long as you're mechanically inclined and/or an actual mechanic...
> 
>> A new water bottle was designed for the bike which actually allows you to
>> continuously ride for weeks at a time and have plenty of clean fluids
>> available using an innovative recycling system.
>>
> 
> What is that supposed to analogize? My battery doesn't last anywhere
> near the advertised time. I'm lucky to get 3 or 4 hours out of it, and
> lately not even that.

Do you have some (e.g. 3rd party app) running in the background
taking CPU?  Or something that frequently polls network
(Do you switch it to offline mode when you don't use it)?
Or doesn't allow the display to blank when you don't use the device?


- Eero

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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext OgnenD wrote:
>>> It is too slow when browsing the net (compared to, for example, my Asus
>>> EEE or my laptop).
>> Oh great, you are comparing an ultra low-power 320MHZ ARM CPU (RISC)
>> vs a 1Ghz x86 CISC.
> 
> It is not about computational power comparison, it is about functionality. If 
> I can spend the same amount of money and get something that does the same job 
> WAY better and is close in size and weight, why buy the underpowered thing?

 From the technical point of view, the size difference is huge.

More powerful CPU + integrated GPU and much larger amount of RAM will
consume a lot more power.  The larger form of Netbook can accommodate
a much larger battery and dissipate the heat generated by the HW and
battery.  You cannot fit such a battery to a smaller device and such
power usage would fry the smaller device.


- Eero

PS. Some historical perspective...

To some extent, the early contributions Nokia did to open source helped
OLPC to get started (at least they affected the OLPC technology
selections).   Whereas OLPC kicked the netbook marked into existence.
This then forced Microsoft to re-consider its licensing so you got
Windows option for Netbooks too.
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Gary wrote:
>> My N800 boots from a 16Gb SD card and I have plenty of space.
> 
> Have you increased your swap size yet?
> Settings -> Control Panel -> Memory -> Virtual

Swap (file on memory card) is OK for temporary use e.g. to be able
to use some www-site, but I wouldn't enable it permanently because:
- It makes it much more likely that you accidentally corrupt your memory
   card FAT file system (and you may not notice it being corrupted &
   needing repair/reformat)
- Swap file on a corrupted file system makes device *really* unstable
- In some cases it may cause a performance issue too:
   https://bugs.maemo.org/show_bug.cgi?id=2615


 > I use mine for the Internet radio streaming, VPN client, ssh client,
 > and the occasional game.

Regarding the need for swap... Some of the (proprietary?) codecs used by
the media server seem to have tiny memory leaks.  If you play/stream
media files for a long time, even tiny leak can grow huge.  So, if you
run "top" and press "m" to sort things by memory usage and
osso-media-server shows high on the list, restart it with:
killall osso-media-server


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

Re: Nokia device usage

[Eero Tamminen]

Hi,

ext Ognen Duzlevski wrote:
> lead disappears and you have to set it up again (here is the problem: I 
> am driving and the unit shuts off the screen to conserve power

In fullscreen mode Maemo mapper prevents screen blanking.


> - when I  need the screen, I touch it but then I loose the lead

And if you set the dim and blank timeouts from the Control panel to
same value, the device doesn't dim the screen (which IMHO is just
annoying), but just directly blanks it after the timeout.

When the screen is blanked, tapping it will just unblank it. The tap
that wakes the screen doesn't go to the application.


> and it
> can also zoom in on another portion of the map - naturally I don't have the 
> freedom to play with the menus and re-set everything up since I am 
> driving!). Maps are also an issue, if I decide to download too many zoom 
> levels, the maps are too large and Maemo Mapper dies processing them. 

I've found out that it's better to do this before the trip (and restart
Maemo Mapper after it, otherwise it can take a lot of RAM and make
the whole device slow).


> also difficult for me. Thus, I gave up. Next, Skype: I have tried to 
> have a phone call or two over wireless, but with mixed luck, sometimes 
> too slow, sometimes alright.

Skype has worked fine for me.  Have you some ancient release on your
device (there were sound quality issues in 2007 and earlier releases)?


> So, all in all, what do people use their Nokias for

- As a portable map with Maemo Mapper
   - OpenStreet and the Finnish topographic maps are great
- For Skype
   - when abroad as I don't need Skype otherwise
- Browsing public transportation time-tables etc over phone
- Looking morning TV-news at breakfast
   - from a mobile internet video streaming service
- Viewing photos from our digi-camera when on vacation
   - and in general as a memory card reader
- Some music and games (e.g. graphical adventures with ScummVm)

Basically stuff for what phone is too small for and laptop is too
unwieldy.  It's nice when traveling (I don't need to drag a laptop
with me).


 > and are they happy with the overall usability of the apps and their 
documentation? Or am I
> just being lazy and giving up easily? It seems to me, I would have been 
> better off waiting for an Android based phone, for example.


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

Re: Screen rotation support

[Eero Tamminen]

Hi,

ext Tuomas Räsänen wrote:
> Few questions about screen rotation in Maemo http://wiki.maemo.org/Rotation:
> 
> - Why does the kernel need to know anything about rotation? Isn't it
> just the X's business?

Unless you don't care about rotation being dog-slow, you want it to
be done by the HW (display controller).  HW is controlled by kernel
drivers, not X[1].


- Eero

[1] On x86 / Linux X has traditionally been banging the HW directly
 which is "fine" for reliability & security...
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: damaged power connecter

[Eero Tamminen]

Hi,

ext Mike Sherman wrote:
> The other day I tried to push my N810 out of the car mount forgetting
> that I had plugged the charger in.  After that stress on the power
> jack the N810 won' t charge unless you apply a downward force on the
> power cable.  The jack in the N810 seems to be damaged.

First I'd suspect the charger itself, it's power cable is quite thin
and can break.  Have you tried another charger?


> Any suggestions?  Has anyone gone inside one of these?  I've repaired
> and replaced parts in Palms, but the N seems several levels beyond.


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

Re: media-player library out of date

[Eero Tamminen]

Hi,

ext Keywan Najafi Tonekaboni wrote:
> Am Freitag, den 06.02.2009, 16:33 +0200 schrieb Eero Tamminen:
> 
>> The issue is probably this bug:
>>   https://bugs.maemo.org/show_bug.cgi?id=3970
>>
>> (package for a fixed version of crawler is attached to the bug.)
> 
> yes, this helps. Do you need any information? Can I test this package
> somehow?

To install it, see the comments 13-16 in the bug.
You might be able to install it also with application manager
(click to attachment link in device browser etc).


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

Re: media-player library out of date

[Eero Tamminen]

Hi,

ext Keywan Najafi Tonekaboni wrote:
> Am Freitag, den 06.02.2009, 11:30 +0200 schrieb Eero Tamminen:
> 
>> Which release do you have and
> 
> OS2008 / diablo with the newest updates
> 
>> is the crawler process
>> (do "ps | grep crawl" in the terminal) still running?
> 
> no. How can I start it?

The issue is probably this bug:
https://bugs.maemo.org/show_bug.cgi?id=3970

(package for a fixed version of crawler is attached to the bug.)


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

Re: media-player library out of date

[Eero Tamminen]

Hi,

ext Keywan Najafi Tonekaboni wrote:
> after nobody could help me here I tried to "reset" the library, but
> eventhough I deleted .osso/mediaplayer-* it still use the wrong data.
> Any hints? Did I asked the wrong way?

Which release do you have and is the crawler process
(do "ps | grep crawl" in the terminal) still running?

- Eero

> 
> Regards,
> 
> Keywan
> 
> Am Dienstag, den 03.02.2009, 12:54 +0100 schrieb Keywan Najafi Tonekaboni:
>> Hi,
>>
>> my media-player library contains some albums/tracks, which aren't on the
>> device anymore and didn't contains music I recently copied to the N810.
>> A restart didn't help.
>>
>> How can I update the library manually?
>>
>> Regards,
>>
>> Keywan
>>
> --
> Keywan Najafi Tonekaboni
> http://www.prometoys.net
> 
> peo...@world:/# apt-get --purge remove capitalism
> After unpacking world will be freed.
> You are about to do something potentially beneficial
> To continue type in the phrase 'Yes, do as We say!'
> 
> ___
> maemo-users mailing list
> maemo-users@maemo.org
> https://lists.maemo.org/mailman/listinfo/maemo-users
> 

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

Re: New battery?

[Eero Tamminen]

Hi,

ext Laura Conrad wrote:
>> "Peter" == Peter Bart  writes:
> 
> Peter> I'm not sure whether it's battery related or not, but
> Peter> I have a few tips that may help. Do you have virtual memory
> Peter> enabled?
> >>
> >> No.  I just tried enabling it, and it said "unable to create virtual
> >> memory".
> 
> Peter>  Then I would guess that you do not have a memory card 
> inserted in the
> Peter> external slot.
> 
> Yes, I do.  That's what I read the books off of all the time.

Maybe your memory card FAT file system is corrupted[1] and because
of this the device has made it read-only?

[1] Microsoft FAT file system corrupts easily.  All you need for that
 is unplugging the device USB cable without using "safely remove",
 taking the card or battery out, or device rebooting while there
 are unfinished writes to the memory card...

I would recommend checking the card file system on PC to see whether
the it has issues, that could cause all kinds of trouble for the device.
You can use PC to repair the file system, but simplest might be just
reformatting the card and copying the data again to it.


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

Re: Claws Mail won't start

[Eero Tamminen]

Hi,

ext Peter Bart wrote:
>> I'm running (or at least trying to run) the latest updated version of
>> Claws mail on an N810 with the latest version of Diablo (with all the
>> latest updats). But it refuses to start. This started several weeks
>> ago. I thought eventually updates would fix it but they haven't.
>> Should I try uninstalling and re-installing?
> 
> If you have the mailboxes on one of the memory cards check to see
> if the card is read only. I had that issue and discovered a RS-MMC
> had spontaneously protected itself.

This usually means that the card file system has corrupted.

Memory cards use the common MicroSoft FAT file system.
It's easy to corrupt by:
- unplugging the USB cable without using "safely remove"
- taking the card out while it's being written to
- device freeze/reboot (no reproducible cases known currently)
   when some data is only partly written to the card

Linux kernel remounts disk (in this case memory card) as read-only
when it notices that it's corrupted.  This is done to protect it
from further corruption.

You can resolve this by repairing the file system.  Device
file manager has a utility for this, but there might not be
enough memory to run that if your card is large and full,
so I'd recommend doing it from the PC.


I reformated the card,

Reformatting works also, but loses all of the data on the card.


> verified that was the problem and replaced the card.

There's a good probability that the card itself was still fine.


- Eero

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

Re: Claws Mail won't start

[Eero Tamminen]

Hi,

ext Rick Bilonick wrote:
> I'm running (or at least trying to run) the latest updated version of
> Claws mail on an N810 with the latest version of Diablo (with all the
> latest updats). But it refuses to start. This started several weeks ago.
> I thought eventually updates would fix it but they haven't. Should I try
> uninstalling and re-installing?

You should check why it doesn't run.  If you run it from terminal, what 
it says?

(I've never used Claws myself.)


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

Re: Flashing and flashing the tablet

[Eero Tamminen]

Hi,

ext COURTAUD Didier wrote:
> I am tired on flasing my tablet every week.
> 
> Since the last release of Diablo, the system seems to be very unstable.
> 
> One week ago, my tablet stopped working being unable to reboot ( the status 
> line at the bottom stops at 3/4 and the lamp at the top
> left lights up ).
> 
> I reflashed my tablet with the last available image, redo the two updates of 
> Diablo

If you flash the latest release (last one released today[1]), you don't
need to do SSU updates.  In worst case using SSU will waste few MB of
RAM (as SSU updated binaries & libraries aren't prelinked).

[1] http://maemo.org/news/announcements/maemo_sdk_4-1-2_and_new_ssu_release/


 > and reinstall all my apps and data.

Which packages?

Packages can do anything, so the extra ones are the first things to 
suspect.  They might not for example like if the file system is full.

The device pre-installed SW is tested to boot fine in situations
where user cannot anymore write to the file system.  If you, or some
3rd party package has filled the file system so that even root cannot
write anything to the device (there's some space reserved that only
root can write and the pre-installed SW is tested not to fill that[2]),
device cannot anymore boot.

[2] For example package management aborts installation if there's
 not enough space and cleans the partly installed package away.
 However, if package is badly done, it can install stuff that's
 not in package database (i.e. cannot be cleaned away by dpkg),
 or if it doesn't properly check for errors ("set -e" in install
 scripts), things can break.  Package installation is done as
 root.


> Eveything worked fine until today where the answers of the operaring system 
> seems "strange".
> 
> I have asked for a reboot ( typing reboot as root in an xterm ) and now once 
> again, the tablet does not want to reboot !

It may be safer to use (e.g. long press of) the power button as
that's the method used in our testing.

(I'm not sure Busybox "reboot" does everything properly, for example
whether apps are notified with sigterm about reboot.)


> Is there a way to trap the log of what is wrong at boot time in oreder to 
> correct it unstaead of flashing once again ?

Unless you're (fairly advanced) power user, no.

If you have already SSH daemon installed and it's started early enough
in the boot, you may be able to log into the device with SSH.  Otherwise
you would need to make yourself a serial cable (as they aren't readily
available) and in serial console stop the device boot to initfs. You
can then login to the initfs shell, mount rootfs (and chroot to it
so that you have necessary tools missing from initfs) and inspect
why the device doesn't boot properly.


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

Re: Unzip utility for endusers

[Eero Tamminen]

Hi,

ext Faheem Pervez wrote:
> The existing ones are busybox ones which frankly suck. For example,
> busybox's tar will not accept the j argument which is used to uncompress and
> untar *.tar.bz2 archives.

It seems that Busybox could support bzip2.  File a bug and it could
be enabled for Fremantle.   Currently the tar related options there
seem to be:
CONFIG_TAR=y
CONFIG_FEATURE_TAR_CREATE=y
CONFIG_FEATURE_TAR_GZIP=y
# CONFIG_FEATURE_TAR_BZIP2 is not set
# CONFIG_FEATURE_TAR_LZMA is not set
# CONFIG_FEATURE_TAR_COMPRESS is not set
# CONFIG_FEATURE_TAR_AUTODETECT is not set
# CONFIG_FEATURE_TAR_FROM is not set
# CONFIG_FEATURE_TAR_OLDGNU_COMPATIBILITY is not set
CONFIG_FEATURE_TAR_OLDSUN_COMPATIBILITY=y
CONFIG_FEATURE_TAR_GNU_EXTENSIONS=y
# CONFIG_FEATURE_TAR_LONG_OPTIONS is not set
# CONFIG_FEATURE_TAR_UNAME_GNAME is not set
CONFIG_FEATURE_TAR_TOUCH=y

We might not add new tools to Busybox, but I don't see why we wouldn't
enable features in existing tools that make them more compatible with
their desktop variants (if they really do them more compatible).

If you notice any other similar features missing that could be enabled
to Busybox, please list them in the same bug (or create another one) and
put me on CC.


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

Re: COMPLETE newbie with some questions.

[Eero Tamminen]

Hi,

ext Andre Klapper wrote:
> Am Dienstag, den 02.12.2008, 07:18 -0500 schrieb Screamin Ike:
>> One of my big pulls towards this particular device (which is the N810)
>> was that it runs linux, so I'm operating under the assumption that it
>> has a terminal emulator... 
> 
> It definitely has by default (osso-x-term).
> 
>> I'm also under the assumption
>> that said terminal would have full unicode support (finch through
>> screen, anyone?).
> 
> Not in general for the device, see
> https://bugs.maemo.org/show_bug.cgi?id=1263 .
> The terminal is based on an older Busybox version (1.6), don't know for
> sure - also see https://bugs.maemo.org/show_bug.cgi?id=2918 .

Unicode has different aspects:

* Use of unicode character set
   + supported, everything in UI uses UTF-8.
 - Pre-installed Busybox shell doesn't have proper UTF-8 support,
   but one can install e.g. Bash from the SDK repository for
   interactive use (should work at least in Diablo).

* Support for the unicode fonts
   + supported:
 - fonts are "unicode fonts".
 - pango has support for using multiple fonts in case single
   font doesn't cover all required glyphs.

* Support for the Unicode layout
   + supported, pango can do complicated glyph layout (bi-directional,
 glyph composition etc) and this is plugin based so additional
 layout engines can be added later on by 3rd parties
 - Not sure whether this works in Browser, as Gecko does its
   own text layouting for speed&control reasons.
 - Gtk supports right-to-left layouts, but Hildon UI doesn't.
   Terminal might also have issues with right-to-left languages
   I assume.

* Fonts covering glyphs for the required unicode characters
   + This depends on what glyphs are required, it's not feasibly
 to provide all define 32-bit unicode glyphs or even the 16-bit
 subset (64 thousand different Glyphs) as the fonts would be huge
 i.e. take too much RAM and Flash.  There are a few free fonts which
 cover larger set of Glyphs than the Nokia fonts and fontconfig
 should handle their installation fine[1].

* String localization:
   + Device applications are translated only a to certain set
 of languages, but if one wants to use some other language,
 apps show so called "unique IDs" instead of english, which
 makes the pre-installed applications fairly unusable in other
 languages. I'm not sure what's the Nokia stand on 3rd party
 translations.

I.e. pre-installed software has some understandable gaps in Unicode
coverage, but except for translations, it "should" be pretty easy for
the community to cover them if needed[1].

(All above is AFAIK.  Test to be sure. :-))


- Eero

[1] I'm not commenting on the quality of the free unicode fonts with
larger Glyph coverage, but if really needed, I guess one could just buy
commercial TrueType fonts and put them into /home/user/.fonts/ &
(re-)run fc-cache.
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: Free42 Still not Installing on N800/OS2008

[Eero Tamminen]

Hi,

ext Tommy Persson wrote:
>>> If I added the chinook extra repository manually I could install
>>> it. So the question is why clicking on the install button did not add
>>> the repository.
>> Because you are using the Diablo version, and not the Chinook one.
>> The logic is use Diablo repositories with Diablo.
>>
> 
> Aha. But the error message from the "one click to install" button was
> then a bit confusing. 

Maybe you could add a bug about that to bugs.maemo.org with
some proposal for a better error text?

(Could be fixed for Fremantle, for Diablo things like this are too late.)


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

Re: Headset plugged in disables ringtones

[Eero Tamminen]

Hi,

ext Henrik Frisk wrote:
>  On the other hand, if you are listening to music while the
> phone is ringing, you will want to turn it off before answering the
> phone which somewhat eliminates the problem...

Device has so called "voip-mode" implemented by the media-server
which AFAIK should take care of that automatically.


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

Re: Abiword

[Eero Tamminen]

Hi,

ext Peter Flynn wrote:
> Dr. Nicholas Shaw wrote:
>> Hmmm... Interesting.  Abiword was in my catalog and is installed on my 810
>> (running Diablo).  The current version is 2.6.4.  I've had no problems.
> 
> Turns out it's still in extras-devel, which is (natch) not added by 
> default to a new installation of Diablo. Working fine now.
> 
> I think extras-devel should probably be included in an OS distro, disabled.

No, packages should get from extras-devel to to extras when they're
found to work fine.


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

Re: Headset plugged in disables ringtones

[Eero Tamminen]

Hi,

ext Luca Olivetti wrote:
> El Thu, 27 Nov 2008 17:48:35 +0200
> Eero Tamminen <[EMAIL PROTECTED]> escribió:
> 
>> So, in which all situations it in your opinion should switch?
> 
> Always route the alarms to the loudspeakers and route "normal"
> audio to the headphones if connected or to the loudspeakers if not?
> Or make the alarm output configurable ("always loudspeakers", "always
> headphones", "auto switch")

Well, it goes to one place at the time.  So, you're proposing that
music would switch to loudspeakers when alarm comes and after alarm
it would switch back to headphones?  That could sound wierd...


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

Re: Headset plugged in disables ringtones

[Eero Tamminen]

Hi,

ext Henrik Frisk wrote:
>>> This suggests it would be possible to reroute the audio of the ring
>>> tone to the speaker while the headset is plugged in.
>> But not to both at the same time (i.e. music to headphones,
>> alarm to speaker).  Would you remember to switch sound to
>> speakers when you take the headphones off your head?  I think
>> it's just easier to unplug the headphones from the device.
>>
> I would answer your question with another question: "Would you
> remember to unplug the headset after each call?" and, at least for me,
> the answer is "No." But obviously the solution should not be another
> manual switch, but an automatic.

So, in which all situations it in your opinion should switch?


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

Re: Headset plugged in disables ringtones

[Eero Tamminen]

Hi,

ext Henrik Frisk wrote:
> Just after sending the above off I found this in the
> http://lists.maemo.org/pipermail//maemo-users/2008-July/022127.html:
> 
> "You're mistaken :) The fm radio ui has buttons to toggle between speaker
> and headphone output. As for the bt headset case, dunno."
> 
> This suggests it would be possible to reroute the audio of the ring
> tone to the speaker while the headset is plugged in.

But not to both at the same time (i.e. music to headphones,
alarm to speaker).  Would you remember to switch sound to
speakers when you take the headphones off your head?  I think
it's just easier to unplug the headphones from the device.


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

Re: App-Manager failure

[Eero Tamminen]

Hi,

ext Peter Flynn wrote:
 http://wiki.maemo.org/Updating_the_tablet_firmware
>>> I'm about to flash to Diablo. Everything is backed up, but in case I
>>> need it, whereabouts (what file) are my wlan connection data stored? I
>>> know most of the dozen or so access points I use regularly but there are
>>> a few that I'd like to be able to retrieve.
>> You won't need it. If you've backed up "Settings", access points are
>> included[1].
> 
> Backup never worked; it always hung. I just backed up by tgz'ing /etc 
> and /home/user to a spare SD card. But I couldn't identify where the 
> data was...

Probably this (fixed) bug:
https://bugs.maemo.org/show_bug.cgi?id=3591

(see also the duplicate: https://bugs.maemo.org/show_bug.cgi?id=3425)


- Eero

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

Re: App-Manager failure

[Eero Tamminen]

Hi,

ext Mark Haury wrote:
> Has either of them tried to flash the tablet?

I think only power users do that.  So called normal users would
probably go to a Nokia service point...?


> Had an update fail?

I think the SSU related bugs in the pre-installed packages can
be triggered only(?) by changes people have done from the command
line, like removing the pre-defined folders, doing symlinks,
removing/replacing packages not visible in application manager.

Because of this Nokia testing hadn't catched them.  We of course are
interested about power user problems and try to fix them, but the
testing effort naturally concentrates to what normal people can and
do with the devices.


> Frankly, 
> they've just been lucky so far. Sooner or later there's going to be a problem 
> they can't fix, and it's going to be up to you to deal with it. Either that, 
> or 
> they're never going to do an update or upgrade, and could just as well be 
> using 
> a PalmOS device, because they're not using any of the features or apps unique 
> to 
> the tablets.

Quality of the 3rd party software rests mostly with the community
and that will hopefully get better with time. :-)


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

Re: App-Manager failure

[Eero Tamminen]

Hi,

ext Mark Haury wrote:
> Ryan Abel wrote:
>>> Yes, upgrade to Diablo, it has a much more useful Application Manager. :)
> 
> ...and a bunch of other stuff is broken that previously worked...
> 
>>> If that's not an option, running apt-get update as root in xterm will 
>>> give you more verbose errors 
> 
> ...which is fine for advanced users, but not an option for the average 
> consumer 
> end-user. Just getting root access is not trivial.

Didn't the Application manager log provide enough information?

(And if not, what was missing from it and could you file a bug about
that?)


> If you're going to sell a device that requires significant Linux knowledge to 
> use, that should be clearly stated in any sales information.

You're mailing on the "maemo-users" list, which in its
"sales information":
https://lists.maemo.org/mailman//listinfo/maemo-users

states that it's:
"A list for users of maemo Development Platform."

(It's a bit confusing name.)


Maemo development platform information:
http://maemo.org/development/

states:
--
Developing Maemo applications and platform components is easy if you are
used to programming Linux desktop applications or other components, and
fairly simple if you already know about Linux and GTK+.
--

I'm not sure whether that can be taken as "requires significant Linux
knowledge", but the developer users should be able to use apt at least.


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

Re: the hell of Maemo repos

[Eero Tamminen]

Hi,

ext Mark Haury wrote:
> As for reflashing, Chinook is still better than Diablo as far as
> stability and certain apps working properly. I've had my tablet long
> enough that it came with Bora, and there were compelling reasons to
> upgrade to Chinook. The final Chinook upgrade was a nightmare because
> the backup/restore didn't work and I had to reinstall everything from
> scratch. The only compelling reason to upgrade to Diablo thus far is
> the "flashless" updates, except people have been reporting similar
> breakage with those...

Personally I think the main advantage of Diablo are the performance
improvements in Browser.  It depends a bit on which pages you visit,
but for some pages (having transparent overlays, especially when
zoomed), the browser is significantly faster.


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

Re: N810 dead?

[Eero Tamminen]

Hi,

> On Tue, Nov 18, 2008 at 2:36 PM, Mike Sherman <[EMAIL PROTECTED]>wrote:
>> This morning I turned it on and the boot screen comes on and then the
>> brightness drops (as usual) and then it goes black.

Device is turned off, display isn't just very, very dark?

Are you sure the device charged properly?
(maybe the device charger cable is broken?)


>> The corner light
>> comes on during the boot (I think it supposed to) in the white with
>> blue and green tinges.  But that's all it does.

So there's no progress bar at the bottom of the screen (which comes
shortly after rootfs is mounted)?  If not, this would most likely
be a problem either with mounting the rootfs or battery handling.



ext Aniello Del Sorbo wrote:
> Not sure, but maybe you ran out of space on the device when saving
> the 7hrs long trace ?

To cause bootup problems there should be also some (3rd party) root
process that fills the disk, user processes cannot fill the device
full enough to prevent boot.

I would also expect it to cause the device to enter reboot loop
if there's no space even for root to write anything.


- Eero

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

Re: All applets dying

[Eero Tamminen]

Hi,

ext Eero Tamminen wrote:
> I had recently updated my other device to latest Diablo and happened
> just to install Statusbar load-applet (haven't used that earlier) to
> it and after using it for a while, Desktop crashed.  I haven't gotten
> any Desktop crashes without load-applet.
> 
> The backtrace looks like this:
> Signal 11 (SIGSEGV)
> 0 0x4097013c  IA__gdk_pixbuf_fill (pixbuf=0x786f4274, pixel=0) at 
> gdk-pixbuf.c:544
> 1 0x42fa0350  ?? () from unknown
> 
> And when looking at the load-applet code in Garage, it's at least using
> that function, so it might explain some crashes:
> https://garage.maemo.org/plugins/scmsvn/viewcvs.php/trunk/load-applet.c?root=load-applet&view=markup

Ahem, forgot to look into the SMAPS data included into "rich-cores".

Yes, this crash is definitely in load-applet code as 0x42fa0350
address is where load-applet is loaded.

...
42f9c000-42fa4000 r-xp  1f:04 21551 
/usr/lib/hildon-status-bar/libload.so
Size:32 kB
Rss: 32 kB
Shared_Clean: 0 kB
Shared_Dirty: 0 kB
Private_Clean:   32 kB
Private_Dirty:0 kB
...


So, if you had load-applet running in Desktop before the crash,
you could install the Gtk libgtk2.0-0-dbg package, reproduce your
crash and check with Gdb[1] whether it shows something similar.

If yes, please complain to load-applet developers:
https://garage.maemo.org/projects/load-applet/


- Eero

[1] For more instructions, see:
http://maemo.org/maemo_release_documentation/maemo4.1.x/node15.html#SECTION001523230
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: All applets dying

[Eero Tamminen]

Hi,

ext Tommy Persson wrote:
>> It could be any applet causing the problem.  The only way to find it is to 
>> remove some applets and see if it recurs.  
> 
> Could a status bar application cause the problem?
> 
>> Alternatively you can post your list of applets and see if anyone
>> else has had a bad experience with any of them.
> 
> I will test to disable applets until I find the guilty one.
> 
> The ones I usally use are: Clock, Disk Free Size, Display IP, GPE
> Summary, Internet Search and IpHome.

What about statusbar applets?


I had recently updated my other device to latest Diablo and happened
just to install Statusbar load-applet (haven't used that earlier) to
it and after using it for a while, Desktop crashed.  I haven't gotten
any Desktop crashes without load-applet.

The backtrace looks like this:
Signal 11 (SIGSEGV)
0   0x4097013c  IA__gdk_pixbuf_fill (pixbuf=0x786f4274, pixel=0) at 
gdk-pixbuf.c:544
1   0x42fa0350  ?? () from unknown

And when looking at the load-applet code in Garage, it's at least using
that function, so it might explain some crashes:
https://garage.maemo.org/plugins/scmsvn/viewcvs.php/trunk/load-applet.c?root=load-applet&view=markup


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

Re: All applets dying

[Eero Tamminen]

Hi,

ext Graham Cobb wrote:
> It would be useful to have a debug mode in the desktop process where it would 
> produce a log of interesting events (and all Glib logging messages)

One can install syslog to the device and restart hildon-desktop.

(Just remember to uninstall or at least disable syslog when it isn't
anymore needed, the extra JFFS2 disk churn caused by it contributes
to things like: https://bugs.maemo.org/show_bug.cgi?id=2615)


 > and log which library contained the faulting instruction when any
 > signals occur.  Maybe even take a core dump.

Core dumps are enabled by creating "core-dumps" directory to the memory
card.

(If core dumps are sent to Nokia, also "sp-rich-core" package should
be installed from the tools repository[1] as it collects some extra
system information needed in outside device debugging, such as what
exact release is in question, memory usage statistics etc.)


- Eero

[1] http://maemo.org/development/tools/
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: slow N800

[Eero Tamminen]

Hi,

ext Jason Edgecombe wrote:
> I watched the fennec demo and tried out the program, but my N800 was
> about 3-5 times slower than the demo.

Did you visit the same www-sites?

I haven't tried Fennec myself, but I've understood it to be a bit
of a memory hog and if you're not using sites benefiting from its
faster JavaScript VM, well, I guess you'll just have the memory
hogging...


Does the demo have flashplayer disabled?  Disabling that can have
a large effect on the browsing speed.


> My N800 is running diablo with all of the updates. What should I do
> to speed up my tablet?   I thought the N810 has the same memory and
> CPU  as the N810,

Correct.


> so how can I make my tablet as fast as the N810 in  the fennec demo video?

To check whether the memory usage is issue, try "top" in the terminal
and check how much memory is available (free & buffered & cached).

Besides using the same www-sites and disabling the Flashplayer, you
might try rebooting the device to get more memory available if that
seemed an issue.

(If you've watched a lot of videos in the mediaplayer since last boot, 
the tiny leaks in different video codecs could have amounted to a bit
more in the media server over time, this you can fix just with
"killall osso-media-server" in the terminal.)


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

Re: Fennec vs MicroB

[Eero Tamminen]

Hi,

ext Ryan Abel wrote:
> On Oct 21, 2008, at 9:09 AM, John Holmblad wrote:
>> If The ~6.5x performance gain of Fennec over Microb as asserted in the
>> Aris Technica article holds as this product moves to production then I
>> will vote with my "install apps" button soon thereafter.
> 
> It's only a JavaScript improvement, not necessarily a general  
> performance improvement. Fennec is only faster because it is based on  
> a newer version of Firefox. MicroB will get faster, too (it already  
> has over the past two SSU updates), as it comes inline with Firefox  
> trunk.

These performance improvements were due to (mostly ARM specific)
optimizations to the Xomap and MicroB's older Gecko engine graphics
handling, not changes from upstream.


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

Re: Write access for MMC

[Eero Tamminen]

Hi,

ext Shivkumar Chandrasekaran wrote:
 I have a 770 with OS2007 Hacker's edition with a 1GB removable card.
 Everything worked fine until yesterday when I suddenly lost write-access
 to the card. So I can no longer do backups or write or remove files from
 the File Manager. I tried connecting it to my Linux machine using the
 USB cable, but the Linux machine also insists that the file system on
 the USB drive is read-only (so I could not delete files for example from
 linux terminal, even with super-user permission).

 Can anyone give me some advice so that I can re-acquire write access?
>> the file system card can be corrupted and needs an fsck.
> 
> Any idea how I can do that? When I chose "Memory card -> mmc2 ->
> Format ..." from File Manager it tells me that the memory card is in
> use. I tried booting up without the memory card and then inserted it
> later to no effect.

Metalayer-crawler media indexer is probably going through it.
If waiting a while doesn't help, you can disable the crawler
using /etc/init.d/ script (as root).


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

Re: WiFi borked after update

[Eero Tamminen]

Hi,

ext Marius Vollmer wrote:
>> True. Perhaps we (as a community) should try pushing this message
>> more: red-pill mode is intended as a rescue environment in the event
>> of b0rkage; not for every day use (even by power users)?
> 
> Yes.  What about making red-pill mode non-persistent: on the next start
> of the AM, it would be back in blue-pill mode.

Sounds good to me.

It needing virtual keyboard (that could be borked) on N80
is another concern though...


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

Re: WiFi borked after update

[Eero Tamminen]

Hi,

Marius Vollmer wrote:
> Eero Tamminen <[EMAIL PROTECTED]> writes:
>>> Unlike apt-get, the AM tries to ignore broken packages so that, for
>>> example, a broken maemomapper installation does not prevent a OS
>>> upgrade. 
>> Um, wouldn't a "broken" package database be a good reason
>> to refuse to do on OS upgrade?
> 
> Hrmmhrmmh.  Yeah  But it is quite hard to unbreak packages
> with only the Application manager, and if we do it right, there
> shouldn't be any harm coming from unrelated, unconfigured or otherwise
> broken packages.
> 
> Sometimes I am thinking that maybe dpkg should ignore most exit codes of
> maintainer scripts, and the scripts would need to explicitly exit with
> 112 to let the installation fail or something... most maintainer script
> failures in maemo are bogus.

I've been going through the ITT forum and although it's only(?) a power
user issue, checking that SDK repo isn't enabled could save some grief
in case user has forgotten to disable that.  Or maybe the update should
first disable all repositories that aren't needed for the update to make
sure that they don't interfere with the upgrade process?


Btw. As I didn't notice here a link to this, here's an excellent summary
of the findings so far and some advise for power users who can use apt
directly:
http://www.internettablettalk.com/forums/showpost.php?p=229093&postcount=304


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

Re: WiFi borked after update

[Eero Tamminen]

Hi,

ext Marius Vollmer wrote:
> "ext Tommy Persson" <[EMAIL PROTECTED]> writes:
> 
>> I suspected I did not have enough space to install but I suppose the
>> reason for the failure could have been something else.
> 
> There is one stupid bug in the Application manager that I suspect might
> be to blame for some of the failures.
> 
> If you have broken packages on your device, that might trick the AM into
> believing that an OS update will succeed while in reality it will fail
> with missing dependencies, conflicts or similar.
> 
> Here are the details, in case you are interested:
> 
> Unlike apt-get, the AM tries to ignore broken packages so that, for
> example, a broken maemomapper installation does not prevent a OS
> upgrade. 

Um, wouldn't a "broken" package database be a good reason
to refuse to do on OS upgrade?

If things are already broken, wouldn't OS upgrade just break it more...?


When preparing the OS update (or any other package operation),
> the AM 'simulates' what would happen to the dependencies, and if any
> package is broken by the simulation, the OS update is not done.
> 
> The stupid thing is this: in order to ignore packages that are already
> broken before the simulation, the AM remembers the number of broken
> packages before the simulation, and if there are _more_ packages broken
> after the simulation, the operation is rejected.
> 
> Now, it can happen that the simulated operation fixes some of the
> previously broken packages, and breaks some others.  Thus, the number of
> broken packages can be the same or even lower eventho the operation
> should be rejected.
> 
> I'll fix that.


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

Re: Insufficient Space to upgrade

[Eero Tamminen]

Hi,

ext Scott wrote:
Along these same lines, is there a file browser that will see the  
entire "disk" including the hidden folders so I can see where the  
files are piling up and which temp ones can be deleted? Or is there a  
utility for the n810 similar to TreeSize (windoze program) that will  
show you all the folders and how much space each folder (recursive)  
takes up?


I use something like the attached scripts from command line on
the device:
du-fs-contents.sh -- shows sizes of directory contents, dir at the time
package-sizes.sh -- list sizes of all packages according to pkg database

NOTE: the package (database) "Installed-Size" information isn't
trustworthy, the package can take a lot less disk space:
* on install certain documentation directories are removed from
  the packages (with docpurge) and amount of that can be quite large
* The file system is also LZO compressed, so files may take
  significantly less space than they appear. For example:
# cd /home/user/MyDocs/.documents
# lzop osso_software_copyright.pdf
# ls -s osso_software_copyright.pdf*
1359 osso_software_copyright.pdf
334  osso_software_copyright.pdf.lzo

I.e. the PDF file actually takes only few hundred KB from the file
system instead of over MB.  Some other files like jpeg, png, mp3, wma
etc ones are in an already compressed format.


Debian localepurge might also be useful, but I don't know whether
it can be easily used on our tablets:
http://packages.debian.org/etch/localepurge


- Eero


du-fs-contents.sh
Description: application/shellscript


package-sizes.sh
Description: application/shellscript
___
maemo-users mailing list
maemo-users@maemo.org
https://lists.maemo.org/mailman/listinfo/maemo-users

Re: Another annoyance: bluetooth icon disappears

[Eero Tamminen]

Hi,

ext Andrew Flegg wrote:
> On Fri, Sep 26, 2008 at 4:27 PM, Ryan Abel <[EMAIL PROTECTED]> wrote:
>> There is a project to combine the Bluetooth and WiFi applets into a
>> single applet (an "Advanced Connectivity" of sorts), but it hasn't
>> really gotten off the ground lacking anybody with real knowledge of
>> libconic and the connectivity interfaces, and, unfortunately, because
>> both applets are closed there are no examples to work from.

I think this is then the real issue here.  Is there a bug about
both of these issue (the interfaces lacking documentation and
applet sources) which people could vote on?


> Don't you see, this is Nokia's "differentiation" at work. By having a
> UI which doesn't meet the user's requirements, they can ensure that
> competitors can't take their code and produce something just as
> limited! (*cough* #303 *cough* ;-))
 >
> In all seriousness, I just leave Bluetooth on all the time now, and
> allow paired devices to initiate connections that way. I would imagine
> that Bluetooth hasn't had quite the power usage focus as WLAN  but it
> should have had *some*.

Correct.  I use the control panel myself, but I don't need Bluetooth
that often.


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

Re: Another annoyance: bluetooth icon disappears

[Eero Tamminen]

Hi,

ext Mark wrote:
> Wonderful... yet another "wontfix" response to usability issues...

If I understood this correctly, the enhancement requested that
for a set of users who use separate bluetooth keyboard, the
pre-installed device software usability would be improved by
changing the UI as proposed in the bug.

This was replied with:
---
However, we have a strong requirement to reduce the amount of permanent
Status bar indicators. Status bar is crowded already and we should
prefer temporary Status indicators (like alarm and usb indicators).
The primary function of Status bar indicators is to indicate status -
controlling something from there is secondary. Of course you may
disagree and can continue that discussion in some other forum ;)

We are going to change this design in future releases. Maybe not as
a permanent indicator, but let's see.
---

I think better time to do UI enhancement requests is when Nokia does
software alpha / beta releases for the forthcoming product / release,
not when it's already released and in bugfix phase.  Especially if you
want it for the same HW.

However, I don't understand what's the problem with this functionality
being offered by a 3rd party component if the component is easily
available from the extras repository?


> One problem is that putting the device in offline mode disables
> bluetooth, but taking it out of offline mode usually doesn't re-enable
> bluetooth (at least 75% of the time on my N800). Another is that any
> situation where the other device is the main physical interaction
> (headset, keyboard, etc.), it has no ability to turn on the bluetooth.
> Sure, if some app on the tablet itself asks for bluetooth it can get
> it, but not external devices. There *does* need to be a quick way to
> access the bluetooth settings.
> 
> The only workable mode right now if one uses any bluetooth devices at
> all is to always leave bluetooth turned on. It's much too big a PITA
> to go through the many levels of the Control Panel to turn on
> bluetooth every time you need it, even if it's infrequently.
> 
> Mark
> 
> On Thu, Sep 25, 2008 at 3:30 PM, Andre Klapper <[EMAIL PROTECTED]> wrote:
>> Am Donnerstag, den 25.09.2008, 14:48 -0300 schrieb Fernando Cassia:
>>> What is the reason for the Bluetooth icon to disappear when bluetooth
>>> is disabled?
>> Basically a design decision, see the discussion in
>> https://bugs.maemo.org/show_bug.cgi?id=971


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

Re: N810 : Locking the virtual keyboard

[Eero Tamminen]

Hi,

ext COURTAUD Didier wrote:
> Is there a way to lock the virtual keyboard to uppercase and numbers ?
> 
> When I put it in "upper case mode" to type a lettre in Capital, it falls back 
> immediatly to lower case
> and it is very difficult to enter a text entirely in upper case !

You need long-tap on the "CAPS" VKBD key for it to stick down.


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

Re: Problems with RSS feed reader in Diablo

[Eero Tamminen]

Hi,

ext Michael Conklin wrote:
> I will try to report it. Being a complete linux novice this may be beyond my

Instead of making a new bug, you could attach the data also to:
https://bugs.maemo.org/show_bug.cgi?id=3335


> current capabilities (attaching the contents of a device folder to the bug).

You could do something like following in the X terminal
(can be found under Utilities in the applications menu):
tar -zcf /media/mmc1/rss.tgz /home/user/.osso_rss_feed_reader/

(this assumes you have the memory card in mmc1 slot)

And then connect the device USB cable to PC and attach the rss.tgz
archive from the card to the bug.


> I will also try reflashing and reinstalling my apps to see if I can
> duplicate the problem and pin down the cause. I won't have time to do
> the reflash until the weekend however.

The above listed bug mentions a workaround for the RSS issue.  Could you
first try that and see whether it works for you and add comment about
that to the bug?


- Eero

> Mike
> 
> On Tue, Sep 23, 2008 at 2:36 AM, Eero Tamminen <[EMAIL PROTECTED]>wrote:
> 
>> Hi,
>>
>> ext Michael Conklin wrote:
>>
>>> Yes and no - I can browse to the sites, and everything appears fine, plus
>>> they were 4 completely unrelated sites that all failed at the same time.
>>> Rebooting does not help. At about the time that the RSS feed stopped
>>> working
>>> I had uninstalled and reinstalled Boingo because I was having problems
>>> with
>>> that.  Another user suggested running maemo-summoner to see what error
>>> messages might show up and output from that indicates a GTK widget fails
>>> but
>>> I don't know if that is relevent or not.
>>>
>> I don't recall a bug like this.  Could you report it to
>> bugs.maemo.org and attach to the bug the contents of your
>> device "/home/user/.osso_rss_feed_reader/" directory?
>>
>> If you mail the bug number to the list, others can then check
>> whether those RSS feed reader settings & cache trigger the same
>> issue for them.
>>
>>
>> You could also install "strace" according to instructions here:
>>http://maemo.org/development/tools/
>>
>> Check with that what RSS feed reader does when/after it gets stuck:
>>strace -p $(pidof osso_rss_feed_reader|cut -d' ' -f1)
>>
>> And attach that information to the bug too.
>>
>>
>>  On Mon, Sep 22, 2008 at 10:11 AM, Eero Tamminen <[EMAIL PROTECTED]
>>>> wrote:
>>>> I recently reflashed my N800 with Diablo on Saturday Sept 6th and my RSS
>>>>> posts and subscriptions continued to show. As of Sept 17th at 9:21 pm my
>>>>> tablet was able to update its RSS feeds for the last time. Since then
>>>>> all
>>>>> feeds are dead. When I try to update them I get an exclaimation point in
>>>>> the
>>>>> little RSS icon next to the feed and nothing changes. I thought I would
>>>>> try
>>>>> unsubscribing and then resubscribing to the feeds. I successfully
>>>>> unsubscribed, but when I attempt to subscribe again the application just
>>>>> freezes.
>>>>>
>>>>> Has anyone else experienced anything like this or do you have any
>>>>> suggestions.
>>>>>
>>>>> You can connect with the device browser to the site providing the feed
>>>> etc?  (i.e. networking and the sites themselves work?)
>>>>
>>>> If you reboot the device, does that affect anything?
>>>>
>> - Eero
>>
> 

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

Re: Problems with RSS feed reader in Diablo

[Eero Tamminen]

Hi,

ext Michael Conklin wrote:
> Yes and no - I can browse to the sites, and everything appears fine, plus
> they were 4 completely unrelated sites that all failed at the same time.
> Rebooting does not help. At about the time that the RSS feed stopped working
> I had uninstalled and reinstalled Boingo because I was having problems with
> that.  Another user suggested running maemo-summoner to see what error
> messages might show up and output from that indicates a GTK widget fails but
> I don't know if that is relevent or not.

I don't recall a bug like this.  Could you report it to
bugs.maemo.org and attach to the bug the contents of your
device "/home/user/.osso_rss_feed_reader/" directory?

If you mail the bug number to the list, others can then check
whether those RSS feed reader settings & cache trigger the same
issue for them.


You could also install "strace" according to instructions here:
http://maemo.org/development/tools/

Check with that what RSS feed reader does when/after it gets stuck:
strace -p $(pidof osso_rss_feed_reader|cut -d' ' -f1)

And attach that information to the bug too.


> On Mon, Sep 22, 2008 at 10:11 AM, Eero Tamminen <[EMAIL PROTECTED]>wrote:
>>> I recently reflashed my N800 with Diablo on Saturday Sept 6th and my RSS
>>> posts and subscriptions continued to show. As of Sept 17th at 9:21 pm my
>>> tablet was able to update its RSS feeds for the last time. Since then all
>>> feeds are dead. When I try to update them I get an exclaimation point in
>>> the
>>> little RSS icon next to the feed and nothing changes. I thought I would
>>> try
>>> unsubscribing and then resubscribing to the feeds. I successfully
>>> unsubscribed, but when I attempt to subscribe again the application just
>>> freezes.
>>>
>>> Has anyone else experienced anything like this or do you have any
>>> suggestions.
>>>
>> You can connect with the device browser to the site providing the feed
>> etc?  (i.e. networking and the sites themselves work?)
>>
>> If you reboot the device, does that affect anything?


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

Re: Problems with RSS feed reader in Diablo

[Eero Tamminen]

Hi,

ext Michael Conklin wrote:
> I recently reflashed my N800 with Diablo on Saturday Sept 6th and my RSS
> posts and subscriptions continued to show. As of Sept 17th at 9:21 pm my
> tablet was able to update its RSS feeds for the last time. Since then all
> feeds are dead. When I try to update them I get an exclaimation point in the
> little RSS icon next to the feed and nothing changes. I thought I would try
> unsubscribing and then resubscribing to the feeds. I successfully
> unsubscribed, but when I attempt to subscribe again the application just
> freezes.
> 
> Has anyone else experienced anything like this or do you have any
> suggestions.

You can connect with the device browser to the site providing the feed
etc?  (i.e. networking and the sites themselves work?)

If you reboot the device, does that affect anything?


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

Re: Diable/Modest - No delete from server solution

[Eero Tamminen]

Hi,

ext Tim Ashman wrote:
> 5. Modest STILL loads even though I have deleted all accounts, and turned off 
> automatic check for mail.  I checked the modest bug tracker and this is a 
> reported bug.  I'm so disgusted with Nokia for one for switching the email 
> client and for whatever is causing modest to not only reload about every 5 
> minutes but stay in memory even though I have no accounts setup and no auto 
> retrieve setup.  Major bummer #2

I didn't find this from bugs.maemo.org, could you give the bug number?

(Bug 3610 seems related, but not quite the same.)


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

Re: Remarkable community projects

[Eero Tamminen]

Hi,

ext Eric Warnke wrote:
> Barring any objections I will start a new page designed to be updated  
> over time to highlight new and upcoming projects based on a number of  
> simple criteria.  Based on the existing criteria.
> 
> *Alive: Maintained regularly and available in the last stable release.
...
> *maemo.org active: available in extras, complete information page in  
> Downloads.

Aren't the above two overlapping (the "available in the last stable
release" part)?

Maybe it would make sense to have "Alive" from the developer perspective
(the code proceeds in the repository and has releases) and "maemo.org
active" as the user perspective (packages available, can be installed
without user needing to build the sources himself)?

I mean, these two activities can be done by different person,
the second one is about packaging and uploading and this can be
done also by an "outsider".


- Eero

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