maemo.org - Talk
Page 42 of 47 « First 324041 42 4344 Last »
Show 40 post(s) from this thread on one page

maemo.org - Talk (https://talk.maemo.org/index.php)
-   Maemo 5 / Fremantle (https://talk.maemo.org/forumdisplay.php?f=40)
-   -   [Announce] CSSU Testing thread (https://talk.maemo.org/showthread.php?t=80525)

Estel 2014-01-31 03:28
Re: [Announce] CSSU Testing thread
 
Out of curiosity, in which (else) directory you have rcS-late file?

sixwheeledbeast 2014-01-31 17:53
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Estel (Post 1409907)
Out of curiosity, in which (else) directory you have rcS-late file?
I believe none, rcS-late was "missing" due to being in the incorrect directory.

Android_808 2014-02-23 21:59
Re: [Announce] CSSU Testing thread
 
Has anyone else encountered the new battery widget not updating during charging. I started it with 5mAh left. Some time later, still charging and display reads 5mAh still. Disconnect from charger: 5mAh but reads 86% IIRC. Reboot: ~1140.

Brodyga 2014-02-24 14:46
Re: [Announce] CSSU Testing thread
 
Yes, Android_808, you are right.

marmistrz 2014-03-23 09:30
Re: [Announce] CSSU Testing thread
 
I might've found a bug: /var/lib/dpkg/info isn't optified. It may eat even 40M of precious rootfs space (16%). Can you check, whether there are any symbolic links on the way there on your device.

hxka 2014-03-23 09:40
Re: [Announce] CSSU Testing thread
 
Ubifs compresses files.
Since content of /var/lib/dpkg/info is text files, which are easily compressible, I don't think it's needed to be optified.
In my case, 30 MB of files in /var/lib/dpkg/info take only 5 MB on filesystem.

Estel 2014-03-24 01:57
Re: [Announce] CSSU Testing thread
 
It's worth to remember for future, though - if CSSU thumb even get close to a point, when we would be able to disable compression on rootfs, those 5 MB could get handy.

Not to mention, that with disabled compression, we might actually want to put some things back in rootfs for performance, so we might need even more free space on rootfs beforehand :D

Alecsandru 2014-03-24 15:44
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by marmistrz (Post 1418143)
I might've found a bug: /var/lib/dpkg/info isn't optified. It may eat even 40M of precious rootfs space (16%). Can you check, whether there are any symbolic links on the way there on your device.
what stops you to move-it in opt and symlink-it? many things run fine moved in opt

hardy_magnus 2014-05-08 18:11
Re: [Announce] CSSU Testing thread
 
wow i am not able to use hardware shift key on my nokia nopp. sorry without shift key working i cannot enter numbers . i can use vkb but hardware keyboard is one of the main features of this great device. the hardware keyboard layout in the text input is set to english, nederlands . hardware shift key works when the text input is other than english but then it inserts symbols not numbers. anyone facing this unwanted problem in the latest thumb or testing flavours . have i set hardware keyboard layout wrong. cant use question mark wtf.

hardy_magnus 2014-05-08 18:48
Re: [Announce] CSSU Testing thread
 
i just installed ukbdcreator and ukeyboard and shift key is now working fine. i had these two installed but both accidentally got removed a few days ago. but what if i remove these packages again???

ivgalvez 2014-05-13 14:21
Re: [Announce] CSSU Testing thread
 
Here suffering the "ringtone problem" for the first time since a few days ago. Many phone calls lost due to this problem that might affect more than a 60% of received calls.

I'm using only a default tone (no custom tones have ever been used since last reflashing) with KP52, CSSU Thumb 21.2011.38-1Tmaemo9.1+thumb1 and HD Codecs.

My main suspects are HD codecs and latest CSSU update, both installed just before the problem appeared. Will try to revert to intermediate backups and test

Alecsandru 2014-05-13 15:08
Re: [Announce] CSSU Testing thread
 
i don't have it

petur 2014-05-13 15:15
Re: [Announce] CSSU Testing thread
 
I think I have it too, using only KP52 and CSSU Thumb.... just received a message and the phone only vibrated. Will keep an eye (ear) on it.

Alecsandru 2014-05-13 15:21
Re: [Announce] CSSU Testing thread
 
I'm using KP53 and CSSU Thumb

freemangordon 2014-05-13 16:25
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by ivgalvez (Post 1425181)
Here suffering the "ringtone problem" for the first time since a few days ago. Many phone calls lost due to this problem that might affect more than a 60% of received calls.

I'm using only a default tone (no custom tones have ever been used since last reflashing) with KP52, CSSU Thumb 21.2011.38-1Tmaemo9.1+thumb1 and HD Codecs.

My main suspects are HD codecs and latest CSSU update, both installed just before the problem appeared. Will try to revert to intermediate backups and test
May I have your syslog?

ivgalvez 2014-05-14 09:53
Re: [Announce] CSSU Testing thread
 
Thanks, but as usual with these kind of issues, I haven't been able to reproduce it since I enabled syslog.

Here's the log of a successfully received call (with hidden caller):

Code:
May 14 10:50:03 Nokia-N900 telepathy-ring[1619]: GLIB MESSAGE Modem-Call - incoming call from "+XXXXXXXXXXXX"
May 14 10:50:03 Nokia-N900 systemui-tklock[1555]: Method call received from: :1.10, iface: com.nokia.system_ui.request, method: tklock_close
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: speech start: srate=1, format=1, stream=1
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 1
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 2
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 3
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 4
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 5
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 6
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 7
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 8
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 9
May 14 10:50:04 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: Sending ul frame # 10
May 14 10:50:05 Nokia-N900 pulseaudio[1292]: client-ext.c: client-ext.c: Can't obtain command line
May 14 10:50:05 Nokia-N900 pulseaudio[1292]: module-alsa-sink-old.c: Buffer underrun ! frames: 1920
May 14 10:50:05 Nokia-N900 pulseaudio[1292]: ratelimit.c: 9 events suppressed
May 14 10:50:06 Nokia-N900 telepathy-ring[1619]: GLIB MESSAGE Modem-Call - mt-released incoming call from '+XXXXXXXXXXXX' com.nokia.csd.Call.Error.Network.UnspecifiedNormal: Unspecified Normal
May 14 10:50:06 Nokia-N900 pulseaudio[1292]: module-alsa-sink-old.c: Buffer underrun ! frames: 480
May 14 10:50:07 Nokia-N900 telepathy-ring[1619]: GLIB MESSAGE Modem-Call - terminated incoming call from '+XXXXXXXXXXXX' com.nokia.csd.Call.Error.Network.UnspecifiedNormal: Unspecified Normal
May 14 10:50:07 Nokia-N900 pulseaudio[1292]: voice-cmtspeech.c: speech stop: stream=0
May 14 10:50:07 Nokia-N900 pulseaudio[1292]: module-alsa-sink-old.c: Buffer underrun ! frames: 480
May 14 10:50:07 Nokia-N900 rtcom-call-ui[2025]: GLIB WARNING ** Gtk - gtk_widget_size_allocate(): attempt to allocate widget with width 768 and height -1
May 14 10:50:07 Nokia-N900 [1650]: GLIB WARNING ** mc-client - Unrecognized presence `online'
May 14 10:50:07 Nokia-N900 [1650]: GLIB CRITICAL ** libosso-abook - osso_abook_aggregator_contacts_added: assertion `signals[CONTACTS_ADDED] == hint->signal_id' failed
May 14 10:50:07 Nokia-N900 last message repeated 2 times
May 14 10:50:07 Nokia-N900 pulseaudio[1292]: module-alsa-sink-old.c: Buffer underrun ! frames: 1920
May 14 10:50:07 Nokia-N900 [1650]: GLIB CRITICAL ** libosso-abook - osso_abook_aggregator_contacts_added: assertion `signals[CONTACTS_ADDED] == hint->signal_id' failed
May 14 10:50:07 Nokia-N900 last message repeated 2 times
May 14 10:50:08 Nokia-N900 obexd[1800]: GLIB WARNING ** GLib - poll(2) failed due to: Bad address.
May 14 10:50:08 Nokia-N900 syncd[1885]: GLIB WARNING ** GLib - poll(2) failed due to: Bad address.
May 14 10:50:08 Nokia-N900 hildon-thumbnailerd[1706]: GLIB WARNING ** GLib - poll(2) failed due to: Bad address.
May 14 10:50:08 Nokia-N900 hildon-home[1659]: GLIB CRITICAL ** Gtk - gtk_table_resize: assertion `n_rows > 0 && n_rows < 65536' failed
May 14 10:50:08 Nokia-N900 tracker-indexer[2403]: GLIB WARNING ** GLib - poll(2) failed due to: Bad address.
May 14 10:50:09 Nokia-N900 systemui-tklock[1555]: Method call received from: :1.10, iface: com.nokia.system_ui.request, method: tklock_open

Android_808 2014-05-14 12:59
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by ivgalvez (Post 1425266)
Thanks, but as usual with these kind of issues, I haven't been able to reproduce it since I enabled syslog.
Did you enable syslog via a reboot? Only time I've experienced this is with high swap usage.

ivgalvez 2014-05-14 13:41
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Android_808 (Post 1425283)
Did you enable syslog via a reboot? Only time I've experienced this is with high swap usage.
I had previously rebooted already, so if it's caused by heavy swapping it might not happen in a while.

However, the phone has been running for more than 2 months without rebooting and haven't shown this issue, it only happened with latest CSSU Thumb update and/or HD codecs which, in any case, forced me to reboot.

Android_808 2014-05-14 16:22
Re: [Announce] CSSU Testing thread
 
Symptom for me is usually the phone actually starts vibrating before doing anything else. Screen tries to rotate to portrait but takes ages to do so and is unresponsive when it actually does. In some cases it can go to voicemail before I even get a chance to answer.

I still use adblock, despite the known memory leaks, which I believe to be the cause for me. I can't stand the sight of all the ads and it helps reduce data usage. IIRC I'm using a slightly newer version for FF3.6 (1.3.10 vs 1.1.1 on garage site) that I modified to accept microb being based on the beta but it still leaks like mad. I did manage to get the latest version installed a couple of months ago in Microbe but couldn't get the interface for adding subscriptions to work.

sixwheeledbeast 2014-05-14 17:08
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Android_808 (Post 1425314)
Symptom for me is usually the phone actually starts vibrating before doing anything else. Screen tries to rotate to portrait but takes ages to do so and is unresponsive when it actually does. In some cases it can go to voicemail before I even get a chance to answer.

I still use adblock, despite the known memory leaks, which I believe to be the cause for me. I can't stand the sight of all the ads and it helps reduce data usage.
Do you use Swappolube or Swap on uSD, this setup correctly may help.

Also have you tried using a /etc/hosts type adblocking instead of the plugin?
http://wiki.maemo.org/Ad_blocking

Estel 2014-05-14 18:20
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Android_808 (Post 1425314)
Screen tries to rotate to portrait but takes ages to do so
Slightly different note, but I HATE this "rotate to portrait during call" thing. It made me reject call instead of answering thousand times (just when you hit answer, it rotates, and you're hitting "reject" instead). It breaks x11vnc due to geometry change. It lags. It sucks. It's pointless.

If any changes made by CSSU could allow to make some lock, that would force it to remain landscape *no matter what*, it would be really, really great things to have.

/Estel

Android_808 2014-05-14 18:29
Re: [Announce] CSSU Testing thread
 
Bit OT, but I have both and your flopswap to manage it. I need to look into other options at some stage. More recent versions (>2) of adblock have better memory management than the pre 2.0 builds. Maybe I'll try to get a version of Privacy Badger working with Microbe at some stage.

Edit: Privacy badger doesn't install in current Microbe version. Maybe when I get a Qt5 version going.

sixwheeledbeast 2014-05-14 21:50
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Estel (Post 1425326)
Slightly different note, but I HATE this "rotate to portrait during call" thing. It made me reject call instead of answering thousand times (just when you hit answer, it rotates, and you're hitting "reject" instead). It breaks x11vnc due to geometry change. It lags. It sucks. It's pointless.
There's always been "Turning Control" option in call-ui, Just set it to Landscape.

nokiabot 2014-05-15 03:29
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by sixwheeledbeast (Post 1425355)
There's always been "Turning Control" option in call-ui, Just set it to Landscape.
really :what:

sixwheeledbeast 2014-05-16 10:55
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by nokiabot (Post 1425373)
really :what:
Yep.

@Estel your PM's are full and your not available on IRC....

Open Phone Application
Select top menu
Select Turning Control
Select Landscape
Make sure Launch by turning is unchecked
Save

Phone app now locked to Landscape

Estel 2014-05-16 11:24
Re: [Announce] CSSU Testing thread
 
Thanks for info about my PM box, it should be fixed now.

Anyway, either I'm supid or e have different phone-ui's ;) I don't see a trace of "turning control" in that thing, anywhere. Could yoiu specikfy what method of "open phone application" are you reffering to?

@Others
Sorry for OT.

handaxe 2014-05-16 14:00
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by Estel (Post 1425608)
Could yoiu specikfy what method of "open phone application" are you reffering to?
The Phone handset icon - brings up app called "Phone" with "Dialling pad" and "select contact" buttons with a list of recent calls below. Drop down menu from "Phone" has "Turning control" button amongst 3 others "Call timers", "Clear recent calls" and "Accounts".

biketool 2014-05-30 05:55
Re: [Announce] CSSU Testing thread
 
Even when you lock to portrait the phone still needs to do a transition animation.
Is there a way to completely kill all turning animation or at least animation related to the phone app and just have phone app pop in portrait? That quarter second or less turning animation can get pretty bogged when you already have a few apps open and running. I often miss calls if I don't have a bluetooth headset in where I can just hit a hardware answer button.
Turning animation is pretty in funtime apps but it steals resources for sparkle at the exact time that you really need all resources available to just take the call.

Android_808 2014-05-31 15:48
Re: [Announce] CSSU Testing thread
 
Maybe edit your transitions.ini to shorten or remove animation?
We have blacklist/whiltelist for apps to rotate. Maybe we also need a noanimate list or get each app to register that it doesn't want animation for certain invocations via dbus. As phone app is closed source options are limited.

nokiabot 2014-05-31 15:58
Re: [Announce] CSSU Testing thread
 
Is opensource phone ui possible ?

Estel 2014-07-11 22:33
Re: [Announce] CSSU Testing thread
 
Lately, I've reproducible problem with CSSU'ish camera-ui - everytime I take a bunch of photos in short succession (short - in time span of 3 minutes or so), only one of them, seemingly randomly chosen, is saved. so, for example, after taking 8 or 9 photos, I end up with only a single one saved in DCIM folder on MyDocs.

Whats even more strange, after every photo, I get image-viewer pop-up showing it. mmcblk0p1's filesystem is Ok, no related error messages in terminal or syslog or dmesg.
---

Anyway, considering the epic amount of serious bugs/regressions with CSSU's camera-ui and no chances for anyone fixing them in foreseeable future, I opt for moving it "back" to cssu-devel, at least until someone able to maintain it and fix bugs is found.

Believe me, I'm requesting it with heavy heart, considering how I love idea of FOSS camera replacement and the features camera-ui introduced. It's just completely unreliable to use ATM, without any bugfix for a year or godsknowhowlong.

Considering the fact that FOSS camera-ui is requiment for 720p video recording, there is another possibility - moving camera-ui replacement to extras-* as a non-conflicting package, so people (like me) could use normal camera-ui for everyday needs, and replacement solely for photo taking.

And yes, I know that every can ditch replacement camera-ui on own device (managing dependencies, mp-fremantle-community-pr pestering about them, etc), but it's not the point - I think that keeping stuff in cssu-testing makes sense if we have maintainer that is able and willing to fix discovered bugs, not stockpiling on them over years without clear purpose, right? I'm all for introducing it back if someone starts to work on it again, by any means.

/Estel

marmistrz 2014-07-12 12:29
Re: [Announce] CSSU Testing thread
 
I agree with Estel's suggestion. camera-ui2 is buggy as hell, and no one seems to maintain it. I've experienced a very infuriating bug [1]. Imagine yourself on an abroad trip with each <3;20> photo not being saved. A camera-ui hang instead. Btw. no feedback from so-called maintainers (in the bug report I asked how to get some sensible log). And there've been similar bug reports in the past.

[1] https://bugs.maemo.org/show_bug.cgi?id=12731

panjgoori 2014-07-12 18:24
Re: [Announce] CSSU Testing thread
 
i also agree with marmistrz and Estel.

freemangordon 2014-07-19 09:09
Re: [Announce] CSSU Testing thread
 
estel: IIRC you promised to open bugs against camera-ui a couple of months ago. It could be that I didn't search correctly, but I just can't find those bug reports on BMO. Your post above is a rant, not a bug report, sorry, not to say I am unable to reproduce what fails on your device. AFAIK you have more than one n900, could you please take one of those, flash it to stock rootfs/eMMC, install ONLY CSSU-T (or CSSU-thumb) and try to reproduce the issue.

camera-ui is not unmaintained, I just don't see bug reports against it which contain enough information for me (or some other guy) to start chasing the bug.

I know your (and marmistrz') device is heavily tweaked and it might be the reason why camera-ui fails for you while it works perfectly reliable for me (and not only me, my GF uses N900 too, so if there is some really big issue I would have heard about it already, lots of times ;) ).

Estel 2014-07-20 20:22
Re: [Announce] CSSU Testing thread
 
Agreed that bugs were not posted on BMO (although, last time I checked, camera-ui2 thread was full of reproducible bugs... I thought that they're "common knowledge", understanding, that chasing them in such long thread may be a little time-wasting for devs), and glad to hear that CSSU's camera-ui isn't unmaintained.

I'll browse through camera-ui2's thread and try to reproduce it (and compare) in both devices, my daily-use one and the CSSU-Thumb only one. Then, it's time to gather them all in BMO report.

Don't expect it to happen tomorrow, though ;) Help from any people experiencing problems (in form of refreshing my memory about issues you've encountered, posted in fresh post in camera-ui2 thread) highly appreciated.

/Estel

// Edit

Glad to see that my "rant" resulted in at least:
https://bugs.maemo.org/show_bug.cgi?id=12731

...getting responses and actual debugging work, after like what, 3 months of being totally ignored? Not that I don't agree with request for posting everything important on BMO instead of random TMO threads, but even the sole bug report seems to indicate that my "feeling" about camera-ui being maintainer-less wasn't totally unjustified, to say at least.

Anyway, if you're going to be the one squashing bugs in it, I'm more than happy and motivated for some bug-spotting, so all well that ends well!

gianko 2014-07-30 22:15
Re: [Announce] CSSU Testing thread
 
an issue with the sharing-service-flickr package has been reported here probably due to changes in flickr authentication

is this something to be fixed in cssu-testing? (and hopefully stable as soon as we have a manteiner?)

geektech 2014-08-16 00:10
Re: [Announce] CSSU Testing thread
 
After a while I returned to my beloved N900 , it showed a bunch of updates and I start doing manually, I started from CSSU updates they install then I update pigdori I think, my data was not working, I reboot it and it stuck on the boot loop.

Any solution before flash it? I did not backed it up. fxxc me...

nokiabot 2014-08-16 07:33
Re: [Announce] CSSU Testing thread
 
when next cssu stable coming?

freemangordon 2014-08-16 09:39
Re: [Announce] CSSU Testing thread
 
Quote:
Originally Posted by nokiabot (Post 1435970)
when next cssu stable coming?
should be soon, quote from IRC:

Quote:
merlin1991 12,29,29
*looks for his n900 with stable on it*
gonna testrun the new release before throwing it onto anyone

marmistrz 2014-08-28 07:23
Re: [Announce] CSSU Testing thread
 
I noticed a bug (either in CSSU's Qt or in the components).

How to reproduce:
1. Install useragenttool and qt-components-1.4
2. Launch it
3. Click the + in the toolbar
4. Write anything in any of the fields
5. Select something of it
6. Try to move one of the drops at the ends (to change the selection, as in Harmattan)

Log & backtrace:

Code:
(gdb) r
Starting program: /opt/UserAgentTool/bin/UserAgentTool
[Thread debugging using libthread_db enabled]
[New Thread 0x42a6d480 (LWP 4514)]
MLocalThemeDaemonClient::MLocalThemeDaemonClient(const QString&, QObject*) Theme:  blanco  (hardcoded)
LocalThemeDaemonClient: Looking for assets in ("/usr/share/themes/blanco/meegotouch", "/usr/share/themes/blanco/meegotouch")
X Error: 151 151
  Extension:    148 (Uknown extension)
  Minor opcode: 2 (Unknown request)
  Resource id:  0x66202020
MDeclarativeStatusBar "The name com.meego.core.MStatusBar was not provided by any .service files"
QGLFramebufferObject: Unable to resolve framebuffer object extensions - make sure there is a current context when creating the framebuffer object.

Program received signal SIGSEGV, Segmentation fault.
QGLContext::format (this=0x0) at qgl.cpp:3358
3358        qgl.cpp: No such file or directory.
        in qgl.cpp
(gdb) bt
#0  QGLContext::format (this=0x0) at qgl.cpp:3358
#1  0x4069cfce in QGLFBOGLPaintDevice::setFBO (this=0xdd9994, f=0x9d1b00,
    attachment=QGLFramebufferObject::CombinedDepthStencil)
    at qglframebufferobject.cpp:321
#2  0x4069d0ac in QGLFramebufferObjectPrivate::init (this=0xdd9958,
    q=0x9d1b00, sz=<optimized out>,
    attachment=QGLFramebufferObject::CombinedDepthStencil,
    texture_target=3553, internal_format=6408, samples=0)
    at qglframebufferobject.cpp:416
#3  0x4069db80 in QGLFramebufferObject::QGLFramebufferObject (this=0x9d1b00,
    size=..., format=...) at qglframebufferobject.cpp:769
#4  0x42e65380 in ShaderEffectBuffer::ShaderEffectBuffer(QSize const&, QGLFramebufferObjectFormat const&) ()
  from /opt/maemo/usr/lib/qt4/imports/com/nokia/meego/libmeegoplugin.so
#5  0x42e646fc in ShaderEffectSource::updateBackbuffer() ()
  from /opt/maemo/usr/lib/qt4/imports/com/nokia/meego/libmeegoplugin.so
#6  0x00000014 in ?? ()
#7  0x00000014 in ?? ()
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
Is it something related to Qt or should I show it to the components' devs?


All times are GMT. The time now is 06:49.
Page 42 of 47 « First 324041 42 4344 Last »
Show 40 post(s) from this thread on one page

vBulletin® Version 3.8.8