**** BEGIN LOGGING AT Sun May 04 02:59:59 2014 May 04 09:46:08 hi all May 04 09:47:10 have a problem with my old n900, display is turning off/on randomly, any idea if that is hardware or software problem? May 04 09:47:47 googled a lot, found no one with such problem May 04 09:49:58 how can deactivate lock button on the right side of n900? May 04 10:00:10 maybe it's proximity sensor failure? May 04 10:00:56 did you reflash? May 04 10:02:27 please cover up upper left (portrait) corner of screen May 04 10:03:06 screen backlight is supposed to shut down at very high ambient light May 04 10:03:41 probe for ambient light is done every maybe 10s May 04 10:03:59 ALS sensor is upper left corner May 04 10:13:53 proximity shouldn't dim screen except when on phonecall May 04 10:14:28 but there are apps to do that nevertheless. That's why I asked if you did a reflash May 04 10:23:11 DocScrutinizer05, i did reflash two or three years ago May 04 10:23:24 to get latest official release May 04 10:23:52 didn't expriment with device last two years May 04 10:36:43 when screen goes "dark" is it only the backlight or also the complete content of LCD vanishing? May 04 10:36:58 complete content May 04 10:37:03 or maybe even only LCD goes black but backlight stays on? May 04 10:37:16 nope May 04 10:37:23 keyboard is turn off too May 04 10:37:26 the latter only visible in the dark May 04 10:37:39 DocScrutinizer05, noticed, that issue first appeared after some small square and black piece fall from the back of the device May 04 10:37:46 the former only visible in bright daylight May 04 10:38:00 o.O May 04 10:38:02 back cover was opened May 04 10:38:04 )) May 04 10:38:19 it was approx 3mmx3mm square May 04 10:38:32 a black square on screen? May 04 10:38:36 nope May 04 10:38:42 sharp edges? May 04 10:38:45 under the back cover May 04 10:38:51 plastic May 04 10:38:56 sorry? May 04 10:39:53 i don't know how to name it ) May 04 10:40:34 where battery located May 04 10:40:58 i opened cover to change sim May 04 10:41:07 wanted to close again May 04 10:41:10 lemme re-read what you posted May 04 10:41:36 oooh May 04 10:41:53 dang! May 04 10:42:02 small square piece from plastic, approx 3mmx3mm May 04 10:42:10 maybe little bigger May 04 10:42:28 suppose it was covering some sensor or so May 04 10:42:48 with a few silver dots on one side and a ying&yang alike line on the other? May 04 10:43:15 nope, it was pure plastic, no parts from the inside ) May 04 10:43:24 errr May 04 10:43:30 weird May 04 10:44:27 i don't know maybe it doesn't influence issue with display May 04 10:44:33 any such plastic for all I can imagine is completely unrelated to display light. Unless it's been the slider magnet May 04 10:44:47 just coincidence May 04 10:44:52 but that ould hardly fall out of backside of phone May 04 10:44:58 would* May 04 10:45:01 right May 04 10:45:12 is your slider properly detected May 04 10:45:15 ? May 04 10:45:35 yes, display turns on/unlocks when opened May 04 10:45:42 ok May 04 10:45:44 and off when closed May 04 10:45:45 most probably he broke one of holding clips from backcover May 04 10:45:55 aaah possibly May 04 10:46:00 yes May 04 10:46:04 think so May 04 10:46:21 there's one such thing at camera end of back cover May 04 10:46:44 another hypothesis is lock switch is broken May 04 10:46:50 on the right side May 04 10:47:01 hmm, can't see that May 04 10:47:14 and triggers on/off spontaneously May 04 10:47:23 well, *maybe* May 04 10:47:50 but that would be quite weird a failure pattern May 04 10:48:04 is it possible to turn it off on the software side? May 04 10:48:16 disable it May 04 10:48:30 err, in theory for sure May 04 10:48:52 hey, I know: do a `stop mce` May 04 10:49:08 and? May 04 10:49:38 and see what happens May 04 10:50:21 it reliably stops most of all ways to dim screen May 04 10:50:36 i see May 04 10:50:41 i'll try May 04 10:50:41 no more lockslider button, no more timeout, no ALS May 04 10:50:50 screen should stay on May 04 10:50:53 bad news are, that issue is not such easy to reproduce by intention May 04 10:50:56 you need root May 04 10:51:15 i.e. i cannot trigger it May 04 10:51:16 ooh May 04 10:51:22 that's bad May 04 10:51:29 not easy to test May 04 10:51:45 sometimes i think that it's gone May 04 10:51:47 tbh I suspect flex breakage May 04 10:52:07 but next day i find device discharged in case May 04 10:52:22 because of display was on whole day May 04 10:52:36 oooh, yes. when it's locked it for sure mustn't unlock May 04 10:52:58 unless some shitty app for proximity sensor got installed May 04 10:53:13 I suggest reflash May 04 10:53:45 would be time kill if it's a hardware problem May 04 10:54:12 i'll try first with deactivating mce May 04 10:58:16 DocScrutinizer05, power button is also not working with stopped mce )) May 04 10:58:37 seems reasonable May 04 10:59:36 if it would work, then my problem could be considered as solved May 04 11:00:19 how can disable automatic display turn off? in settings it's possible only set maximum timeout of 2 minutes May 04 11:03:32 lock switch is very sensitive May 04 11:03:45 i don't have to slide it maximum to lock/unlock May 04 11:04:13 just small shift triggers lock May 04 11:05:28 DocScrutinizer05, if i post dpkg -l somewhere can you take a look at it? maybe some obscure app indeed "plays" with display May 04 11:09:16 just in case: http://pastebin.com/XyEHRtW3 May 04 11:09:57 vdv: your lockslider switch is defect, disassemble and check! May 04 11:11:55 DocScrutinizer05, could be defect in such a way, that it's triggered even if i don't touch it? May 04 11:12:52 because sometimes display turns off/on if device is on table and i don't even touch it May 04 11:13:07 and also no calls, no usb connection May 04 11:28:07 check your syslog (install it when you didn't yet) - it logs all lockslider events on kernel level May 04 11:28:50 makes sense, thanks May 04 11:29:04 Apr 9 01:24:39 IroN900 kernel: [ 3.100524] kb_lock (GPIO 113) is now open May 04 11:29:38 Apr 9 01:26:06 IroN900 kernel: [ 93.730346] kb_lock (GPIO 113) is now open May 04 11:29:39 Apr 9 01:26:11 IroN900 kernel: [ 98.910156] kb_lock (GPIO 113) is now closed May 04 11:29:58 i see May 04 11:29:58 sorry May 04 11:30:01 thanks May 04 11:30:07 Apr 9 01:44:35 IroN900 kernel: [ 1203.050720] kb_lock (GPIO 113) is now closed May 04 11:30:09 Apr 9 01:44:35 IroN900 kernel: [ 1203.292877] kb_lock (GPIO 113) is now open May 04 11:31:23 Apr 9 01:51:59 IroN900 kernel: [ 1646.628936] kb_lock (GPIO 113) is now closed May 04 11:31:25 Apr 9 01:51:59 IroN900 systemui-tklock[1226]: Method call received from: :1.10, iface: com.nokia.system_ui.request, method: tklock_open May 04 11:31:26 Apr 9 01:51:59 IroN900 kernel: [ 1647.089752] kb_lock (GPIO 113) is now open May 04 11:34:08 http://privatepaste.com/1c4727f130 May 04 11:48:40 actually a slightly more comprehensive view of stop mce; start mce with a number of lockslider events: http://privatepaste.com/b38e022d6b May 04 11:49:12 blocking lockslider on kernel level is also possible, but not exactly trivial May 04 11:49:52 sth along echo foo >/sys/*/*/buttons/113/*/* May 04 11:54:57 less '+F' /var/log/syslog is fun May 04 12:05:08 DocScrutinizer05, does value written in /sys/*/*/buttons/113/*/* also persisted? May 04 12:05:26 i.e. will not disappear after restart May 04 12:06:54 DocScrutinizer05, "start mce with a number of lockslider events" <-- what do you mean? May 04 12:07:14 there's only log in paste May 04 12:07:30 no commands to start mce with a number of lockslider events May 04 12:08:09 http://maemo.cloud-7.de/share-service/20140504_003.jpg http://maemo.cloud-7.de/share-service/20140504_004.jpg http://maemo.cloud-7.de/share-service/20140504_005.jpg lockslider switch May 04 12:08:43 there's stop mce, a number of lockslider events, and start mce event in log May 04 12:09:22 ah, ok May 04 12:10:20 ((does value written in /sys/*/*/buttons/113/*/* also persisted?)) no May 04 12:10:35 is it possible that lockslider oscillate if i even don't touch device? May 04 12:11:12 yes, I see this every few days with a mechanical movement detector in a remote-control May 04 12:12:01 ok May 04 12:12:34 suddenly it activates button backlight even while sitting calmly on the shelf. For sure a walking in the room suffices to trigger it then, when it got that hypernervous state May 04 12:13:02 when the spring in the switch is broken, it will behave exactly similar May 04 12:13:56 in the photos you see a triangle notch which gets pushed down by slider motion from right to left May 04 12:14:28 err left to right, sorry May 04 12:15:52 it must have a noticable spring loaded force when you operate it by sliding your nail from left to right over it. And it has a audible ping tone when it gets released and snaps up again May 04 12:20:48 i see May 04 12:28:03 you can probbaly trigger the problem by slightly knocking the device backside under the lockslider, while holding device with screen facing upward May 04 12:28:35 you should definitely install and check syslog May 04 12:42:19 DocScrutinizer05, thanks a lot, i'll definitely debug with syslog May 04 12:51:38 yw May 04 12:56:35 DocScrutinizer05, display turns on, but syslog says: ".. is now closed" May 04 12:56:55 sorry? May 04 12:57:37 i monitoring now with tail -f /var/log/syslog May 04 12:57:48 kb_lock (GPIO 113) is now closed -> the slider is engaged May 04 12:58:05 kb_lock (GPIO 113) is now open -> the slider switch got released again May 04 12:58:31 ah, then makes sense May 04 12:59:48 i pressed power button two times to turn display off, put device on shelf and observed line with "... is now closed" on my terminal May 04 13:00:08 looked at my n900 and display was on May 04 13:00:14 and it's still on ) May 04 13:00:42 to test lockslider it doesn't help to operate powerbutton May 04 13:00:45 now turned off, maybe because of 2 minutes setting May 04 13:01:11 i see May 04 13:02:01 engage lock slider button, watch for kb_lock (GPIO 113) is now closed. Release it and watch for kb_lock (GPIO 113) is now open May 04 13:02:13 both shall show up instantly in syslog May 04 13:02:36 see timestamps in http://privatepaste.com/1c4727f130 May 04 13:02:58 I engaged lockslider switch for 0.2 .. 1s May 04 13:03:56 thus you always see a May 04 13:03:58 May 4 13:56:35 IroN900 kernel: [108029.025146] kb_lock (GPIO 113) is now closed May 04 13:04:00 May 4 13:56:36 IroN900 systemui-tklock[1239]: Method call received from: :1.1408, iface: com.nokia.system_ui.request, method: tklock_open May 04 13:04:01 May 4 13:56:36 IroN900 kernel: [108029.329559] kb_lock (GPIO 113) is now open May 04 13:04:05 sequence May 04 13:05:31 sometimes I'm faster than the system, so the systemui-tklock message shows up after the kb_lock (GPIO 113) is now open May 04 13:05:42 think, it's now clear, that lockslider is triggered randomly without engaging or releasing it explicitly May 04 13:05:45 but it's triggered by the kb_lock (GPIO 113) is now closed May 04 13:05:54 toldya May 04 13:06:05 and that wakes up display May 04 13:06:09 sure May 04 13:06:20 your switch is broken May 04 13:06:25 now i'll try to disable it in kernel May 04 13:06:33 good luck! May 04 13:06:56 and if it works maybe put command somewhere in boot script? May 04 13:10:47 sure May 04 13:11:52 google searchterm /sys/bus/platform/drivers/gpio-switch/unbind May 04 13:12:36 there's no buttons node in /sys May 04 13:14:10 there's /sys/devices/platform/gpio-switch May 04 13:15:26 GPIO 0x48002110 is the CAM_B_EN signal of N900, which could change the camera to A or B. May 04 13:15:47 or it's some other "gpio" May 04 13:16:04 there's roundabout 250 GPIO on SoC May 04 13:16:23 we're talking about GPIO113 here May 04 13:16:43 )) May 04 13:16:50 understood ) May 04 13:17:12 ~gpio May 04 13:17:26 ~ping May 04 13:17:27 ~pong May 04 13:17:34 ~wtf gpio May 04 13:17:35 Gee... I don't know what gpio means... May 04 13:18:23 ~gpio is General Purpose Input/Output (a CPU pin for universal configurable use as either input or output) May 04 13:18:24 DocScrutinizer05: okay May 04 13:19:26 infobot: no, gpio is General Purpose Input/Output (a CPU pin for universal configurable use as either input or output). A CPU (particularly SoC) has hundreds of them, usually May 04 13:19:27 okay, DocScrutinizer05 May 04 13:20:42 infobot: no, gpio is General Purpose Input/Output (a CPU pin for universal configurable use as either input or output). A CPU (particularly SoC) has hundreds of them, usually. On linux, see /sys/devices/platform/gpio-switch May 04 13:20:43 DocScrutinizer05: okay May 04 13:21:34 there's slide node in gpio-switch May 04 13:21:42 and it has disable with value 0 May 04 13:21:50 maybe set it to 1? May 04 13:21:58 and see what happens? ) May 04 13:22:18 or kb_lock May 04 13:22:28 probably kb_lock May 04 13:23:06 exactly May 04 13:23:10 it's kb_lock May 04 13:23:53 lockslider doesn't work now May 04 13:24:07 hope it'll solve my original problem May 04 13:26:19 DocScrutinizer05, where should i put "echo 1 > /sys/devices/platform/gpio-switch/kb_lock/disable" now? not in rcS, right? ) May 04 13:26:45 wow! May 04 13:26:52 echo 1>disable works? May 04 13:27:00 yes May 04 13:27:05 coolo! May 04 13:27:18 wait May 04 13:27:18 is that sarcasm? )) May 04 13:27:23 no May 04 13:27:27 ok ) May 04 13:31:20 actually yes, I'd add it to /etc/init.d/rcS, way to the buttom, among the other echo commands you find there May 04 13:31:56 prolly directly before "exit 0" line May 04 13:33:21 I got some funny garbage heritage there: May 04 13:33:47 echo 5 > /proc/sys/vm/page-cluster May 04 13:33:48 echo 500 > /proc/sys/vm/dirty_expire_centisecs May 04 13:33:50 echo 4 >/sys/block/mmcblk0/queue/nr_requests || echo "ohnoes, hope it boots1" May 04 13:33:51 echo 4 >/sys/block/mmcblk1/queue/nr_requests || echo "ohnoes, hope it boots2" May 04 13:33:53 echo 65536 > /proc/sys/net/core/rmem_default May 04 13:34:35 the echo "ohnoes, hope it boots1" should give you some idea about the risks involved in editing this file ;-) May 04 13:35:31 )) May 04 13:56:24 vdv: you still alive? May 04 13:56:51 yep May 04 13:57:02 vdv: wouldn't be the first time we never again heard of somebody trying to edit initscripts ;-) May 04 13:57:40 you already edited and booted? May 04 13:57:51 you should get BM installed before you do May 04 13:57:56 ~bm May 04 13:57:56 somebody said backupmenu was http://talk.maemo.org/showthread.php?t=63975 May 04 13:57:56 )) i'm not a guru of course, but also not a newbie ) May 04 13:58:38 also recoveryOS should come in handy, just in case May 04 13:58:49 ~recoveros May 04 13:58:54 ~recoveryos May 04 13:59:00 duh May 04 13:59:04 ~rescueos May 04 13:59:04 extra, extra, read all about it, rescueos is http://n900.quitesimple.org/rescueOS/ May 04 13:59:50 for sure BM is simpler and more convenient to use May 04 14:00:32 rescueOS is prolly smarter, using BM shell is even more smart May 04 14:02:03 never heard of both, seems rescueOS is like livecd May 04 14:02:53 rescueOS must not be flashed on device in advance right? May 04 14:03:29 failed boot -> boot rescueOS -> fix -> boot normally , right? May 04 14:05:14 and BootMenu is like lilo or grub? May 04 14:16:58 vdv: actually rescueOS can be put on-device May 04 14:17:02 via uboot May 04 14:17:22 my n900 has it onboard May 04 21:44:26 how can I use WPA2 from CLI (so I don't interrupt GPRS)? May 04 21:46:36 Luke-Jr: you want to have both wifi and gprs up? May 04 21:47:17 freemangordon: yes May 04 21:47:43 hmm, I am not exactly person to help you, but what about iwconfig? May 04 21:47:48 just an idea May 04 21:48:40 yeah, that's what I'm hoping for..but iwconfig is the furthest thing from easy to use May 04 21:48:57 and I can only barely get on IRC right now; webpages are hopeless May 04 21:49:27 Luke-Jr: hmm, maybe you could download qtmobilehotspot source and read through the magic id does May 04 21:49:39 I suspect there are dbus calls to bting the interfaces May 04 21:49:47 *bring May 04 21:49:53 up that is May 04 21:50:14 wtf is with my typing today ?!? May 04 21:52:13 Luke-Jr: what about http://wiki.maemo.org/N900_dbus#com.nokia.icd2 ? May 04 21:52:37 esp #define ICD_DBUS_API_CONNECT_REQ "connect_req" May 04 21:55:43 I wish I could open links May 04 21:56:40 ooh May 04 21:57:27 sorry, can't help further, I have the same problem with iwconfig as yours :) May 04 22:58:22 hmm, seems route is the first thing to think about, eh? May 04 23:01:39 but honestly why not just use mobilehotspot? May 04 23:02:26 you still can fite a matching set of route configs after starting it May 04 23:02:35 fire* May 04 23:03:17 prolly pointless monologues since he left, eh? May 04 23:35:30 =q/quit May 05 00:47:09 wow. maemo lives XD i love it! hello all May 05 00:48:57 hello May 05 00:49:25 help me pls :D im new in n900 maemo and want to install "silverlight", moonlight, mono, i dont know May 05 00:49:30 it is possible? May 05 00:51:32 I remember mono being available, but it's probably an outdated version now May 05 00:52:18 and can you tell me the correct package name? cause apt-cache search mono replies with 218 packages :) May 05 00:55:36 I suppose http://maemo.org/packages/view/mono-complete/ May 05 00:55:55 seems to be an outdated version as I mentioned early May 05 02:00:01 thank you! **** ENDING LOGGING AT Mon May 05 03:00:00 2014