**** BEGIN LOGGING AT Thu Aug 01 02:59:58 2013 Aug 01 07:31:16 (Maemo 5) I just installed the latest CSSU-S update. How do I use the new camera ui? Aug 01 07:31:50 ebzzry: "the new camera ui" is not in cssu-stable Aug 01 07:32:57 freemangordon: Where is it? In CSSU-T? Aug 01 07:33:05 yep Aug 01 07:33:30 I have used the CSSU-S catalog. Can I install "over" the CSSU-T catalog? Aug 01 07:33:48 yep, but disable cssu-s catalog first Aug 01 07:34:00 freemangordon: Thanks. I'll do that. Aug 01 07:34:05 it won;t harm if you leave it enabled, but will slow down HAM Aug 01 07:34:21 slowdown? Aug 01 07:34:32 ok, even further :D Aug 01 07:34:37 since when did it ever go fast? Aug 01 07:34:42 lol Aug 01 07:36:12 i have disabled autochecker for HAM Aug 01 07:36:54 once in a while (like once in 3 months?) i run ham before going to sleep just to check for update Aug 01 07:37:07 :D Aug 01 07:37:57 otherwise, it's very annoying that when you're connected to the internet, ham checks for updates automatically Aug 01 07:39:33 I noticed that there is a CSSU Devel catalog? Is it real? Is it worth installing instead of Testing? Aug 01 07:53:39 hello everyone, will any of you help me in installing uboot with kp52 menu on it Aug 01 07:54:16 I have cssu thumb and already flashed my device nearly 14 times today Aug 01 07:58:34 doc, kerio or any other please help Aug 01 08:05:51 kupe: hi! Aug 01 08:06:05 install uboot Aug 01 08:06:10 install kernel-power-bootimg Aug 01 08:06:55 ln -s /etc/bootmenu.d/*power* /etc/default/bootmenu.item Aug 01 08:07:08 u-boot-update-bootmenu Aug 01 08:07:34 and don't boot with the stock kernel, it'll end in tears and in a reboot Aug 01 08:09:23 also, why am i being pinged for this stuff Aug 01 08:09:27 D: Aug 01 08:09:36 i'm just a user Aug 01 08:28:01 thank you kerio Aug 01 08:28:10 np Aug 01 08:28:14 have a nice day Aug 01 08:28:48 so lets me backup everything and then I will report you back Aug 01 08:30:10 ba Aug 01 08:30:16 backup what ? Aug 01 08:30:38 *all* the things! _ò/ Aug 01 08:30:56 what's wrong ? Aug 01 08:33:05 he doesn't trust me :c Aug 01 08:33:09 i feel unloved Aug 01 08:34:37 no Aug 01 08:35:04 I didn't mean that Aug 01 08:36:31 but its always good to have backup Aug 01 08:39:50 kerio can I use kernel-power 51vr1 bootimg if i have kpv51vr1 Aug 01 09:17:36 sure Aug 01 09:21:12 ebzzry: If you don't know what cssu-devel is, DO NOT ENABLE THE CATALOG Aug 01 10:23:58 freemangordon: Ok. Since we're there, what is it, and what does it do? Aug 01 10:24:30 ~cssu-devel Aug 01 10:24:42 ~ping Aug 01 10:24:42 ~pong Aug 01 10:24:48 freemangordon: Yes. How is it different from -testing? Aug 01 10:25:54 ebzzry: http://talk.maemo.org/showthread.php?t=84292 Aug 01 10:26:00 freemangordon: let me load that. Aug 01 10:26:13 ~cssu-devel is http://talk.maemo.org/showthread.php?t=84292 Aug 01 10:26:13 freemangordon: okay Aug 01 10:26:28 ~cssu-devel Aug 01 10:26:29 cssu-devel is probably http://talk.maemo.org/showthread.php?t=84292 Aug 01 10:26:49 freemangordon: ok. thanks for the warning. Aug 01 10:28:09 I'm at 90% of the cssu-testing download. I've been downloading it since ~3 hours ago. I don't know why it is taking that long. My unit's internet connection is good, except for HAM related stuff. Aug 01 10:28:44 hmm, weird, that shouldn't take that long Aug 01 10:29:18 I can run wget and get realistic expectations. But when I install anything from HAM (or apt-get), it is bloody slow. Aug 01 10:29:58 No, I haven't tweaked anything network related. Anyway, maybe I can do some search on TMO. :-) Aug 01 10:30:27 ebzzry: never heard of such an issue Aug 01 10:30:48 maybe your isp does some traffic shaping Aug 01 10:31:12 freemangordon: maybe. I think I may run tcpdump on my gateway machine to see what's happening. Aug 01 10:31:52 freemangordon: Ya, some weird traffic-shaping rules, that (unexpectedly) bork connections to repository.maemo.org Aug 01 10:32:01 mhm Aug 01 10:32:21 freemangordon: I wish I can run iftop on the system, while the upgrade is running. :( Aug 01 10:32:44 connect via ssh :) Aug 01 10:32:47 freemangordon: Unfortunately, I can't SSH to it, for some odd reason. :) Aug 01 10:32:51 hmm Aug 01 10:32:54 weird Aug 01 10:32:57 SSH is ok when not upgrading. :) Aug 01 10:33:07 maybe HAM stops it Aug 01 10:33:10 freemangordon: any hints? :-) Aug 01 10:33:12 maybe. :) Aug 01 10:33:25 freemangordon: and, unpingable! :-) Aug 01 10:33:35 ooh, yes Aug 01 10:33:47 HAM puts the device on offline mode Aug 01 10:34:01 s/on/in/ Aug 01 10:34:01 freemangordon meant: HAM puts the device in offline mode Aug 01 10:34:12 freemangordon: I used to use KP but stopped using it because my system has been mostly unstable with it. A frequent annoyance is not being able to reboot properly, i.e., it shuts down instead of rebooting! Aug 01 10:34:27 freemangordon: Ah, ya. Offline mode! :-D Aug 01 10:34:33 ebzzry: which version of KP? Aug 01 10:34:47 that should be fixed in latest testing version Aug 01 10:36:02 freemangordon: the biggest show-stopper though it that at random times, it sets the frequency so low! Aug 01 10:37:24 freemangordon: and I frequently find myself running 'kernel-config ...' from time to time. Aug 01 10:37:34 The last time I used KP was about two years ago. Aug 01 10:38:39 freemangordon: I'll take a stab at it again. I'll finish the CSSU-Testing installation, create another rootfs and optfs backup using BackupMenu, then install the latest KP. Aug 01 10:41:05 ebzzry: ooh, lots of things changed since then :D Aug 01 10:41:20 ;) Aug 01 10:41:51 So, what's the name of the KP now? Aug 01 10:43:45 s/KP/KP package/ Aug 01 10:43:47 ebzzry meant: So, what's the name of the KP package now? Aug 01 10:43:56 kernel-power Aug 01 10:44:25 "enhanced kernel for power users" or something Aug 01 10:44:29 freemangordon: Thanks. How about its name in HAM? Aug 01 10:44:32 freemangordon: ok Aug 01 11:00:15 boom. there goes a backupmenu restore. Aug 01 11:00:31 somehow my cssu-testing installation borked up my system! >:-( Aug 01 11:01:01 thanks to backupmenu. *sigh* Aug 01 11:03:37 heh, that is a nice greet to cssu-testing :) not really for beginners. Aug 01 11:06:04 __sigh__ Aug 01 11:06:44 It would have been fine, had I known what happened. :) Aug 01 11:07:10 consolatory messages would have been great! :-) Aug 01 11:07:21 I got my n900, touch screen changed, cost about 21.5USD or 16.2 Euro, or 1300rupee. Aug 01 11:07:43 then I noticed, that the key board lights are not working. Aug 01 11:08:58 thought of ambient light sensor, but its clearly open, but the led got hidden under a black coating or whatever. Aug 01 11:09:23 I guess in some ways you might be able to backtrace to what may have caused the issue. Not sure about default HAM but HAM in red pill mode has access to interesting log section.. well in either case there should be a log of where HAM would be verbosee in its proceedings. Aug 01 11:10:31 okay LED is clearly under a coating, I cant even see it while charging, Aug 01 11:10:43 Should probably take it back and say that it was not properly fixed, that there was keyboard LED which normally lights up are all now hidden thanks to that black coating or whatever. Aug 01 11:11:09 no no no .. Aug 01 11:11:27 the top led is under a black coating, Aug 01 11:11:42 but about keyboard lights I dont know. Aug 01 11:12:06 I can type in it, but I guess, it doesnt detect keyboard is pulled out. Aug 01 11:12:11 You mean where there's that three LED colours that would show the state of the device? Aug 01 11:12:23 *(bad english) Aug 01 11:12:32 hmm Aug 01 11:12:36 Yes that one Aug 01 11:13:54 There's a GPIO sensor that would tell you if the keyboard "tray" is out or not. I'm not sure on your setup but I think you can find info from kernel via dmesg regarding if the keyboard slid open or was closed. Aug 01 11:14:00 okay, again. Charging LED is defenitely under a black coating, I can understand that, but why is the keyboard drawn out position not detected? Aug 01 11:14:38 ^^I couldnt fully understand what you said.. Aug 01 11:15:41 There's a sensor that governs whether the keyboard tray is out. Depending on your setup you might be able to see via "dmesg" (a command line based command) to see if kernel would see the keyboard was open or not. Aug 01 11:16:27 okay..I think I should understand this time :) can you link me to a pic or something of that sensor, and where is it located? Aug 01 11:17:24 I'm not an engineer, DocScrutinizer05 would know very well about stuff like this but "dmesg" command should easily tell you if the sensor did detect keyboard is either out or in. Aug 01 11:17:57 anybody what I am doing wrong here is uboot image Aug 01 11:17:59 postimg.org/image/r35tdlvlv/70e3b7a9/ Aug 01 11:18:28 still there ain't an option for kp 51r1 Aug 01 11:20:02 keyboard is definitely not detected, virtual keyboard opens up even with tray out, and portait mode, and it even doesnot unlock screen with drawing keyboard, Aug 01 11:20:39 And I think its Docs off duty time, Aug 01 11:21:15 will have to wait till night.... Aug 01 11:22:46 SAiF, in the meantime you could consult with N900 hardware schematics. There's info on the wmo about it. Aug 01 11:23:47 psycho please help Aug 01 11:24:18 kupe, I don't have uboot setup. I can't I'm afraid (even though I ought to move to uboot soon). Aug 01 11:24:42 * psycho_oreos looks at the link anyway. Aug 01 11:24:54 thanks Aug 01 11:25:55 From what I can see, I don't think there's any particular issue with your device. That might be the standard procedure as to what uboot does by default, is to prompt user choices. Aug 01 11:26:19 What happens when you try the first choice? Aug 01 11:27:55 SAiF_, did you get my last message? Aug 01 11:28:51 attach kernel takes me to omap 1 kernel Aug 01 11:29:30 I suppose it won't boot with the stock kernel? Aug 01 11:30:41 kupe: have you installed the bootimg? Aug 01 11:30:45 have you ran u-boot-update-bootmenu? Aug 01 11:30:48 psycho_oreos , got it, was afk Aug 01 11:30:54 looking Aug 01 11:31:15 here comes kerio yes i have ran u-boot-update-bootmenu Aug 01 11:31:55 what was its output? Aug 01 11:31:55 SAiF, from my N900 (with kp52) I slid the keyboard out and back in, "dmesg| tail" yielded these as results: [64952.292419] slide (GPIO 71) is now open for keyboard "tray" out and [64965.253295] slide (GPIO 71) is now closed for keyboard "tray" in. Aug 01 11:32:39 SAiF: check ~# dmesg Aug 01 11:32:46 okay.. Aug 01 11:32:47 SAiF: for >> slide (GPIO 71) is now open << Aug 01 11:33:17 It should be close to the bottom of the dmesg output, the more closer to the bottom of the output the more recent it is. Aug 01 11:33:37 indeed ;-) Aug 01 11:34:38 oops theres lot, pls wait Aug 01 11:34:42 there's some "black varnish" over the indicator LED always, that is the "sunglasses effect" and shouldn't stop indicator LED from shining thru it Aug 01 11:34:50 I thought you would start telling him where the sensor might be located :) I have a hunch its underneath the screen but that's just a wild guess. Aug 01 11:35:41 I could also thell him where on chip the BGA ball of the GPIO is, but would either help to solve his problem? Aug 01 11:36:21 GPIO 69 inactive Aug 01 11:36:33 SAiF: for >> slide (GPIO 71) is now open << Aug 01 11:37:21 I tried both ways, but it shows GPIO 69 is now inactive Aug 01 11:37:22 I dunno, was just thinking if it may help to see where the faulty part may lie and whether it was the technician's fault or just the affected part(s) are dying out. Aug 01 11:37:46 Yeah GPIO 69 != GPIO 71. They're connected to different stuff. Aug 01 11:37:59 oops again Aug 01 11:38:43 I suppose you could just do "dmesg| grep slide" Aug 01 11:38:50 okay Aug 01 11:40:40 nothing Aug 01 11:41:21 find /sys -iname '*engine*' Aug 01 11:41:38 kerio http://pastebin.com/ELeEGd3p Aug 01 11:41:49 please understand that I am a newbie, trying to keep up Aug 01 11:42:07 kupe: but booting with the keyboard out doesn't show those entries? Aug 01 11:42:27 yes Aug 01 11:42:54 i shoulb be root Aug 01 11:42:56 or Aug 01 11:42:57 ls -l /sys/class/i2c-adapter/i2c-2/2-0032/ Aug 01 11:43:24 kerio well what should i do now ?? Aug 01 11:43:41 SAiF: ls -l /sys/class/i2c-adapter/i2c-2/2-0032/ Aug 01 11:43:51 typing Aug 01 11:44:13 SAiF: use tabulator to auto-complete pathnames Aug 01 11:44:45 I am on a pc, and n900 in my hand, Aug 01 11:44:49 what is tabulator Aug 01 11:44:56 Alternatively if you have setup SSH, you can just SSH and paste those commands in :) Aug 01 11:45:00 okay dont mind Aug 01 11:45:06 I got the output Aug 01 11:45:23 /sy/cl/i2/i2/2... Aug 01 11:45:29 I will, soon, Aug 01 11:45:58 okay,, I am such a fool Aug 01 11:46:55 so what should i tell you from this output Aug 01 11:47:21 if there are 3 lines with engines, for example Aug 01 11:47:42 http://privatepaste.com/37e6b81243 Aug 01 11:48:07 engine1_mode ... engine3_mode Aug 01 11:48:21 yes Aug 01 11:48:47 3 engine lines Aug 01 11:48:54 SAiF: cd /sys/class/i2c-adapter/i2c-2/2-0032/ Aug 01 11:49:10 SAiF: cat selftest Aug 01 11:50:12 FFS I have a fail on LED3! Aug 01 11:50:58 Aug 01 11:51:04 got an Aug 01 11:51:07 OK Aug 01 11:51:08 might be related to my recent messing with triggers Aug 01 11:51:14 good Aug 01 11:51:28 hmm, I have fail on LED0 :( Aug 01 11:51:55 Can You please explain in short words what I just did? Aug 01 11:52:14 I guess testing, itself Aug 01 11:52:20 for errors.. Aug 01 11:52:21 SAiF: cd leg Aug 01 11:52:24 hmm, OK now Aug 01 11:52:46 hmm Aug 01 11:53:12 SAiF: ignore me :) Aug 01 11:54:00 SAiF: are you root? Aug 01 11:54:08 ^^ happy to see you, freemangordon, though I have failed to get 720p working Aug 01 11:54:12 yes root Aug 01 11:54:35 SAiF: are you in /sys/class/leds/lp5523:g now? Aug 01 11:54:40 yes Aug 01 11:55:33 echo 50 > brightness Aug 01 11:56:24 where is '>' Aug 01 11:57:03 got it Aug 01 11:57:36 okay done Aug 01 11:57:50 is indicator LED shining up green? Aug 01 11:58:24 no, its beneath Aug 01 11:59:04 beneath? Aug 01 11:59:31 yes, I could see it after turnig of room lights, that too slightly for 1/20th of a second Aug 01 11:59:59 please rephrase Aug 01 12:00:35 pls wait Aug 01 12:01:01 I saw green light shining for 1/20th of a second Aug 01 12:01:09 that's weird Aug 01 12:01:26 I could only barely see that through this black coating. Aug 01 12:02:16 I wonder if they installed some fake digitizer on your device Aug 01 12:02:31 anyway my main concern is not front led, Aug 01 12:03:03 I know it most probably may not be original Aug 01 12:03:09 mine is, since that isn't supposed to vanish after 1/20th of a second Aug 01 12:03:59 Is the led located on the touch screen digitizer? Aug 01 12:03:59 SAiF: cd .. Aug 01 12:04:12 no Aug 01 12:04:50 SAiF: cd *kb3 Aug 01 12:04:52 then what difference the digitizer makes in led times Aug 01 12:05:23 in kb3 now Aug 01 12:05:26 ok seems this 20th second might have some reason Aug 01 12:06:20 20th second in the sense, it shines at the instant when I press enter with that command Aug 01 12:06:26 while :; do echo 100 >brightness; done Aug 01 12:06:58 (stop by ctrl+C ) Aug 01 12:07:37 shall light up led behind I, O, K Aug 01 12:07:57 No, it doesnt Aug 01 12:08:23 yes Aug 01 12:08:47 hm? Aug 01 12:08:55 again in a similar manner Aug 01 12:09:15 when I press enter, it lights up and fades out within a second Aug 01 12:09:28 that's extremely strange Aug 01 12:09:39 but shows your problem is sw side Aug 01 12:10:06 thanks then I could do a reflash Aug 01 12:10:11 and actually some weird stuff around your digitizer having no transparent spot where indicator LED sits Aug 01 12:10:43 I wonder whether your Ambient Light Sensor has a transparent spot Aug 01 12:10:58 that indicator LED, I can Adjust, because I had to wait 1.5months to get even this digitiser Aug 01 12:11:10 .. haha there is no transparent spot Aug 01 12:11:20 thats what I was trying to say Aug 01 12:11:25 hahah Aug 01 12:11:38 please do: cat /sys/class/i2c-adapter/i2c-2/2-0029/lux Aug 01 12:11:46 this keyboard thing I can reflash Aug 01 12:13:11 then cover upper right corner of screen frame next to NOKIA label with your thumb and do again (cursor up fetches last line) Aug 01 12:13:32 then expose device to bright light and do again Aug 01 12:13:50 report the three numeric values you got Aug 01 12:14:31 pls wt Aug 01 12:14:52 SAiF: then do: cat /sys/devices/platform/gpio-switch/proximity/state Aug 01 12:15:02 cover same corner and redo Aug 01 12:15:29 it should show "open", and then when covered show "close" Aug 01 12:15:40 "closed" even Aug 01 12:15:52 Hmmmm... Aug 01 12:15:56 no such directory Aug 01 12:16:11 pls wait trying again. sorry to keep you waiting Aug 01 12:16:18 np Aug 01 12:16:23 use TAB Aug 01 12:16:32 it never does spelling errors Aug 01 12:18:21 cat /sy/deiplgpprst Aug 01 12:18:53 same with autocompletion in capitals: Aug 01 12:18:58 ho Aug 01 12:19:18 I got the first one 30 Aug 01 12:19:30 Just curous to know if that little magnet exists in it's house under screen on top of third column from the right side of keyboard. Aug 01 12:19:38 *curious Aug 01 12:20:07 cat /syS/deVices/plaTFOWM/gpIO-SWITCH/prOXIMITY/stATE Aug 01 12:21:00 amin007110: yes, that's next to check, though we already kinda did that, via dmesg Aug 01 12:21:18 aha, ok Aug 01 12:21:44 was reading you above comment nokia was written on the left side? Aug 01 12:21:48 and that was rather negitive result, though I don't trust in that result Aug 01 12:22:29 yes, the metal frame has a NOKIA tab on left upper corner o.O Aug 01 12:23:13 yes now understand, the sensor, you mean. closed shows 0 and open shows 30 32 Aug 01 12:23:19 in room light Aug 01 12:23:34 that's almost OK Aug 01 12:23:36 285 near monitor Aug 01 12:24:08 OK Aug 01 12:24:21 wat about proximity sensor? Aug 01 12:24:37 power failed, I may go offline Aug 01 12:24:46 let me check Aug 01 12:26:01 got stuck when I opened the phone function, Aug 01 12:26:12 havent opened it for two months Aug 01 12:26:16 hmm? Aug 01 12:26:35 not responding.. Aug 01 12:26:43 SAiF: then do: cat /sys/devices/platform/gpio-switch/proximity/state Aug 01 12:27:00 restarting Aug 01 12:27:37 your device is thoroughly fsckd up, please do a full reflash Aug 01 12:28:11 actually, I reflashed a week before giving to technician, Aug 01 12:28:26 and what did technician flash to it? Aug 01 12:28:49 nope, he just changed the digitizer Aug 01 12:28:54 and which malware did he install? Aug 01 12:29:04 but A cause may be this. Aug 01 12:30:03 sorry, busy. reflash, then test again and come back Aug 01 12:30:04 I had power kernel installed in it and he never switched off phone, he just remove the battery. and each time you boot in it shows kernel loading failed Aug 01 12:30:07 cya Aug 01 12:30:57 Doc.. will be back soon. Aug 01 12:31:02 thanks Aug 01 12:32:53 it shows open and closed correctly Aug 01 12:33:22 WOOOOOOW, cat /sys/class/i2c-adapter/i2c-2/2-0032/selftest does a NIIIIICE lightshow Aug 01 12:33:28 never knew Aug 01 12:33:36 LOL Aug 01 12:34:21 :( Aug 01 12:34:22 you of course can SEE how it checks each LED Aug 01 12:35:45 SAiF: why :( ? Aug 01 12:36:02 i dont see light show :) Aug 01 12:36:14 lol Aug 01 12:36:36 amin007110 said something abt a magnet? Aug 01 12:37:15 do I have to open the back to check whether that magnet is still present Aug 01 12:37:37 cd /sys/bus/platform/devices/slide Aug 01 12:37:42 cat state Aug 01 12:37:55 "open" / "close" Aug 01 12:38:20 >> "open" / "closed" Aug 01 12:38:35 depending on your kbd slider Aug 01 12:38:52 always closed Aug 01 12:39:02 watch cat state Aug 01 12:39:15 both Aug 01 12:39:16 (stop with ctrl+C) Aug 01 12:39:55 Aug 01 12:40:04 yep Aug 01 12:40:11 thats an issue, I dont have to use ctrl+c, it automaticly comes to # Aug 01 12:40:15 when it never shows "open" then your magnet missing. Get a magnet and place it next to upper row of opened kbd, slide along from Q to P Aug 01 12:40:51 wait WHAT? Aug 01 12:40:58 watch echo state Aug 01 12:41:18 does NOT keep displaying the recent value? Aug 01 12:41:25 yes Aug 01 12:41:25 until you abort it Aug 01 12:41:33 that's a severe problem Aug 01 12:41:52 but fortunately SW right? Aug 01 12:41:55 anyway, please do a full reflash Aug 01 12:42:24 won't cure your probably missong magnet, that got lost at technician when he replaced digitizer Aug 01 12:42:42 yes after checking out that magnet from Q to P Aug 01 12:42:45 (the magnet is "inside" screen slider) Aug 01 12:43:03 O.o, Aug 01 12:43:38 What are the usual "successful" ingredients to accomplish a complete CSSU-Testing installation? Aug 01 12:43:43 I dont want to think that much I will do a full reflash by night and be back soon, bye Aug 01 12:44:08 WTF? Aug 01 12:44:20 DocScrutinizer05, cat command represent just one result, it doesn't keep printing states. I think. Aug 01 12:44:33 @ ebzzry Aug 01 12:45:04 amin007110: check out ~# watch cat /sys/bus/platform/devices/slide/state Aug 01 12:45:40 now yes Aug 01 12:46:15 Aug 01 12:46:20 watch -bn1 cat state Aug 01 12:46:21 Aug 01 12:46:43 WTF??? @kupe Aug 01 12:47:23 freenode issues? Aug 01 12:57:10 [general notice] we achieved another step towards "maemo.org finally rached useful condition": promotion should work, see http://talk.maemo.org/showpost.php?p=1364210&postcount=88 Aug 01 13:01:17 please ask your government to make week 31 "maemo fixing celebration holiday" Aug 01 13:02:58 There's already a religious minority responsible for celebration holidays here Aug 01 13:03:22 isn't it like this everywhere on this globe? Aug 01 13:03:39 and isn't maemo community a religious minority? Aug 01 13:04:40 DocScrutinizer05: the whole week? :) Aug 01 13:04:47 yessir Aug 01 13:05:09 and we will see more to come till this week ends Aug 01 13:05:12 :-) Aug 01 13:05:13 my calendar doesn't show work weeks :( Aug 01 13:05:46 toldya this bug I spotted was in the heart of maemo.org infra Aug 01 13:06:10 and rendered it broken since end of last year Aug 01 13:06:28 now that it's fixed we finally will sort out the few unrelated things easily Aug 01 13:08:01 So what are those unrelated things just curiously? Aug 01 13:08:13 http://talk.maemo.org/showpost.php?p=1364210&postcount=88 Aug 01 13:08:23 Ahh yes the autobuilder :) Aug 01 13:08:36 dir owner/group/permissions Aug 01 13:08:39 for example Aug 01 13:09:43 we never realized since the jobs doing the promotion been broken thoroughly by that other core bug Aug 01 13:10:08 now that the jobs work, we can spot those minor simple issues easily Aug 01 13:11:08 maybe there's more to fix around user credentials/accounts promoting from garage to the other VMs Aug 01 13:11:26 but we will fix such stuff too pretty soon now Aug 01 13:12:09 At least lesser need for maintanence overheads :D Aug 01 13:12:17 indeed Aug 01 13:12:34 takes a hell of a workload off techstaff's shoulders Aug 01 13:12:59 and sixwheeledbeast will be unemployed ;-P Aug 01 13:14:02 no more new recurring "old package version got removed by repo-management" Aug 01 13:14:48 Probably the same goes with those aching apache issues I presume as well? :) Aug 01 13:16:58 we will clean out quite exactly one million bogus package remove events which will speed up all cronjobs and stop heavy workload on apache for those cronjobs and pages like http://maemo.org/packages/view/pid/ Aug 01 13:18:19 you run heavy cron jobs on your web server machine? Aug 01 13:18:31 thus the cronjobs have a chance to actually finish and not abort with "out of RAM", and apache processes will not timeout and go zombie anymore Aug 01 13:18:57 jon_y: you want to run them for us on your machine? ;-P Aug 01 13:19:06 Yeah, hence those aching issues with apache and OOM. I remember that chunk of OOM errors you pasted the other day. Aug 01 13:19:16 yep Aug 01 13:19:28 DocScrutinizer05: you only have a single server? Aug 01 13:19:32 partially at least Aug 01 13:19:53 Any relief is better than none :D that's the bottom line. Aug 01 13:20:06 jon_y: please refer to http://wiki.maemo.org/Migrating_to_Community-driven_Infrastructure Aug 01 13:20:25 and related pages Aug 01 13:20:30 linked there Aug 01 13:21:32 I'm on DSL with about 20KB/s up link Aug 01 13:21:57 yeah, not much there to help host Aug 01 13:23:18 jon_y: I was just kidding you. we don't need more servers to run jobs on them Aug 01 13:24:26 we have two blades populated with a hexacore, on a supermicro server, and we gonna pupulate a third blade eventually Aug 01 13:24:32 populate even Aug 01 13:25:07 jon_y: see http://monitor.maemo.org/ganglia/?r=hour&cs=&ce=&m=load_one&s=by+name&c=maemo&h=&host_regex=&max_graphs=0&tab=m&vn=&sh=2&z=small&hc=4 Aug 01 13:25:09 90% of those resources end up feeding a php monstruosity, sadly :( Aug 01 13:25:29 ugh, I hate php after working with them Aug 01 13:25:40 the syntax the syntax! Aug 01 13:25:47 phpsadness Aug 01 13:26:01 or lack thereof Aug 01 13:26:09 :nod: Aug 01 13:27:20 I was actually bitten by the php "special" ternary Aug 01 13:27:27 typically we're idling a lot, nevertheless Aug 01 13:27:51 what we need is more storage and more memory Aug 01 13:28:05 and we will get both with a third blade eventually Aug 01 13:29:19 then we even got enough resources to joggle with to set up a development environment where we could test improvements to maemo.org Aug 01 13:30:04 if we find any maintainers/devels willing to take care and with enough free time for the workload this implies Aug 01 13:30:21 sixwheeledbeast: ooh, flopswap is in testing :D Aug 01 13:30:30 what's flopswap? Aug 01 13:30:33 hi freemangordon :-) Aug 01 13:30:44 DocScrutinizer05: hi Aug 01 13:31:06 (swapon swap2 && swapoff swap1) || (swapon swap1 && swapoff swap2)? Aug 01 13:31:17 kinda Aug 01 13:31:21 DocScrutinizer05: please post on TMO that autobuilder/repos are just fine, something in the line of "developers, developers..." :D Aug 01 13:31:47 DocScrutinizer05: asking EVERYBODY not needing extras-devel to disable it immediately Aug 01 13:31:55 eventually, freemangordon. I got an appointment with GF Aug 01 13:32:02 will kill me when I'm late Aug 01 13:32:16 DocScrutinizer05: I don't give a shit about your sexual life :P Aug 01 13:32:22 ok Aug 01 13:32:27 no hurry Aug 01 13:32:28 i just use the watchdog to reset my swap usage Aug 01 13:32:35 ~rape freemangordon Aug 01 13:32:35 * infobot takes freemangordon behind the WallMart and makes a few grunts and screams Aug 01 13:32:48 something tends to reboot my n900 before i've used my 2gb swap Aug 01 13:32:49 kerio: the point is that promotion works ;) Aug 01 13:32:53 yay Aug 01 13:33:33 2GB swap are insane nonsense Aug 01 13:33:42 and yet Aug 01 13:33:58 watchdog will reboot device when kernel tries to swap large amounts of it Aug 01 13:34:09 DocScrutinizer05: once we cleanup the events table, I suspect we'll have some RAM freed Aug 01 13:34:10 even .7GB are already too large Aug 01 13:34:19 actually, the only time the watchdog triggered was when i was connected to a really flaky wifi hotspot Aug 01 13:34:29 so i just assumed it was a powersaving issue Aug 01 13:34:35 that caused a panic Aug 01 13:34:40 (re infra resources) Aug 01 13:34:52 but idk if a kernel panic ends up as a watchdog reboot in bootreason Aug 01 13:35:20 the only reason why they used 768MB of swap space, except for marketing blabla, been that swap fragmentation kicks in later Aug 01 13:35:38 yeah Aug 01 13:35:41 and 2gb is even later Aug 01 13:35:57 you can't reasonably use >maybe 500MB swap on a OMAP3 device with flash Aug 01 13:36:09 of course you can't Aug 01 13:36:22 i doubt you can get to 256mb of swap sanely Aug 01 13:36:41 that would be what I'd use for swap size Aug 01 13:36:43 even if it doesn't crash, it'll perform horribly Aug 01 13:37:00 500mb of swap means you have to cycle swap partitions a lot more often than me Aug 01 13:37:02 and then use floopswap or sth like that, frequently Aug 01 13:37:46 2 swap partitions of 256MB each Aug 01 13:37:56 absolutely sufficient Aug 01 13:38:18 keep one active, one to swapswap Aug 01 13:38:37 (swapon swap2 && swapoff swap1) || (swapon swap1 && swapoff swap2) Aug 01 13:38:56 assuming swapon fails if the partition is already enabled Aug 01 13:39:18 yeah, won't prolly work, but nice oneliner anyway Aug 01 13:39:27 ;-) Aug 01 13:39:35 DocScrutinizer05: maemo will complain if only 256MB of swap is set Aug 01 13:39:44 oooh, why? Aug 01 13:39:57 i would actually make it barely more than 256mb Aug 01 13:39:59 OMM killer settings Aug 01 13:40:04 really? Aug 01 13:40:09 hm? Aug 01 13:40:11 but... there's 256mb Aug 01 13:40:13 it's not out Aug 01 13:40:28 check in /proc/sys/vm/... Aug 01 13:40:57 IIRC OOM killer kicks in too early if anything < 512MB is set Aug 01 13:41:43 I recently reswapped to a tiny swap partition I happened to have on my uSD, and maemo actually unloeded some preloaded processes, but recovered just fine after reswapping again back to 768MB Aug 01 13:41:59 and device been usable though laggy on that tiny swap as well Aug 01 13:42:14 laggy on stuff like opening dialer etc Aug 01 13:42:21 not generally laggy Aug 01 13:42:49 of course, since it needs to swap out or even stop other processes to start up dialer when needed Aug 01 13:43:31 but all that been completely transparent to user and not really had any negative impact on fremantle operation Aug 01 13:43:49 So then make four "slabs" of swap with 256MB each, whilst simultaneously enabling two extra swaps (of 256MB) disable the other two swaps (of 256MB)? Aug 01 13:43:58 psycho_oreos: why? :s Aug 01 13:44:09 ...hell, i think i might have *two* 2gb swap partitions on the usd Aug 01 13:44:11 it just introduced some lag when opening new stuff like dialer etc, and of course some loglines in syslog from OOM Aug 01 13:44:16 kerio, trying to kill two birds with one stone :) Aug 01 13:44:21 OHM? Aug 01 13:44:28 HD even? Aug 01 13:45:58 HD OOM? Aug 01 13:46:13 actually 3 swap partitions of 192MB each, of which always 2 are active, that might be worth a try Aug 01 13:46:38 HD is the process managing the preload stuff Aug 01 13:46:55 Ahh. Aug 01 13:46:59 there are some rather nifty scripts somewhere Aug 01 13:47:21 or was it actually c source? Aug 01 13:48:06 can't recall, I looked into it when trying to figure what parameters the .desktop X-MAMEO-PRELOAD= key takes, except 'always' Aug 01 13:48:17 'never' Aug 01 13:48:26 'sometimes' Aug 01 13:49:13 actually it doesn't support any other values, but once it been planned to support other stuff that hen didn't pan out Aug 01 13:49:35 s/ hen/ then/ Aug 01 13:49:35 DocScrutinizer05 meant: actually it doesn't support any other values, but once it been planned to support other stuff that then didn't pan out Aug 01 13:52:12 hello everyone Aug 01 13:52:43 sorry >>X-Maemo-Prestarted=always<< Aug 01 13:52:56 On second thoughts of 3x192MB swaps. It probably will still trigger OOM killer since the threshold was < 512MB swap. Aug 01 13:53:33 find the code that hanles this, and you also found the code that handles unloading of processes when short on RAM Aug 01 13:53:59 mhm Aug 01 13:54:07 * psycho_oreos would rather look at the shortcomings of quick and dirty hacks :D Aug 01 13:54:42 Kupe: hi! here to stay this time? Aug 01 13:55:24 [2013-08-01 14:46:15] Aug 01 13:55:26 [2013-08-01 14:46:19] <-- kupe has left this channel ("Leaving"). Aug 01 13:55:45 ;-) Aug 01 13:56:00 need help on uboot Aug 01 13:56:13 :( Aug 01 13:56:13 mhm Aug 01 13:56:17 It seemed almost like kupe "ragequit" *ducks* Aug 01 13:56:39 Kupe: so shoot! Aug 01 13:56:54 Kupe, did you do what kerio said earlier on? there were two things you needed to do and you mentioned you did one. Nothing about the other. Aug 01 13:57:24 http://pastebin.com/ELeEGd3p and postimg.org/image/r35tdlvlv/70e3b7a9/ Aug 01 13:57:47 I already did both Aug 01 13:57:49 good focus Aug 01 13:59:18 I have u-boot flasher , fiasco-image_update ask and uboot tools installed Aug 01 14:00:16 I installed it using ham Aug 01 14:00:56 kupe: have you installed the bootimg? <--- I think that was left unanswered. Aug 01 14:01:20 ya it is installed Aug 01 14:02:06 * DocScrutinizer05 waves and vanishes Aug 01 14:02:50 sorry guys it seems like noone is going to help Aug 01 14:04:03 I thought I should find out what I did wrong but... Aug 01 14:04:05 bye Aug 01 14:04:08 ~uboot Aug 01 14:04:08 N900 uBoot is a siamese twin binary [uBoot+stockMaemoKernel] that resides in kernel NAND partition /dev/mtd3 aka "kernel". You can't uninstall it, rather you'll nuke it when you flash/install another kernel like stock maemo kernel or powerkernel. To start other than stock maemo kernel via uBoot, you have to provide the according kernel image files Aug 01 14:05:37 o.O Aug 01 14:07:38 "nobody solving my problem, despite I already waiting 10min. See ya" *cough* Aug 01 14:07:49 Can't help the helpless. We're not glued permanently to our devices 24x7 with complete wizardry-like level of skills when it comes to maemo devices. Aug 01 14:08:07 annoying Aug 01 14:08:47 It in itself is probably also a sign that this impatient user may soon part the N900. Aug 01 14:09:01 :nod: Aug 01 14:09:06 * psycho_oreos senses probably another N900 will be available to buy soon enough. Aug 01 14:09:16 ;-) Aug 01 14:09:20 o/ Aug 01 14:11:05 Besides.. here isn't paid/priority/priviledged support. Here is community driven free help, nobody is obliged to help out regardless of whatever they were doing beforehand. Aug 01 14:11:34 what if he dumps the n900 into the recycling bin? Aug 01 14:11:38 nooooooo!!! Aug 01 14:12:28 I don't think he would be that dumb to do that but who knows what he'll do with it. Hopefully the more sane way is to sell it to someone who would be more patient enough. Aug 01 14:13:17 do not underestimate what angry people do Aug 01 14:13:34 or the impatient Aug 01 14:14:21 do not overthink about the situation either. Its not your device in the first place :) Aug 01 14:14:49 yeah, too bad Aug 01 14:17:13 I'm sure the majority of us would like to think that he would have a more sane approach. He was evidently too impatient and is pretty lucky considering how many testers before him risked their devices to get a functional uboot. Aug 01 14:28:09 DocScrutinizer05: At least you have a girlfriend. I barely even have any local female true friends. ☹ Aug 01 14:33:24 Incidentally, I want ckermit for Maemo 5 but I do not want to sidetrack myself to try building and packaging ckermit for Maemo 5. ☹ Aug 01 14:37:48 well Yes i was impatient I have been trying to get working uboot on my n900 since morning and couldn't find what I am doing wrong Aug 01 14:38:26 but that doesn't mean I am going to sell my n900 :) Aug 01 14:39:07 * brolin_empey will have to settle for minicom for the few times he needs a handheld serial terminal emulator. ☹ Aug 01 14:40:43 maybe its better to wait for newer version uboot 2013.7 Aug 01 14:44:03 Or remember to finally try using GNU screen as a serial terminal emulator. Aug 01 14:44:51 phsyco_oreos: your senses are wrong in kupes case.. :) Aug 01 14:45:19 ya Aug 01 14:49:54 Heh, nice cobweb site: http://kermit.wwarthen.com/ Aug 01 14:51:32 http://www.tox.im/ Aug 01 14:57:08 The walls have ears.. o.O Aug 01 14:59:55 At least I wasn't completely wrong in my speculations :p N900 is an old device no matter how you look at it and I didn't think there would still be people that are impatient and not willing to part their beloved N900s :D Aug 01 15:01:25 I'm sure t.m.o. would have a fair few threads about ubooting. There you might be able to pick up some clues as to why it may not work in your case. If all else fails, it doesn't hurt to ignite the same old threads about uboot again and wait for someone to give you maybe some directions. Aug 01 15:37:24 phew. got crossed out sim icon, but no errors in dmesg. Reboot didn't help. Turned out the sim card was actually dislodged Aug 01 15:37:37 phew indeed Aug 01 15:38:05 kupe left again, wtf Aug 01 15:47:04 I'm sure whoever relayed messages to kupe about what was said after he left again will inform him either way. Aug 01 15:50:06 omg omg omg omg omg pypy for ARM Aug 01 15:52:55 freemangordon: omg omg omg omg Aug 01 15:55:55 brolin_empey: i always found minicom to be better than screen to act as a serial terminal Aug 01 15:56:14 screen always fucked up the pseudographical characters during the debian installs on my sheevaplug Aug 01 16:19:48 sorry to bother you guys, but I just cant find that tiny little software (work around) for driver installation for flashing in win 7 64. Aug 01 16:20:14 It works great, and easy, but couldnt find it anymore. any help is appreciated. Aug 01 16:21:21 SAiF: https://wiki.maemo.org/Updating_the_tablet_firmware#Issues_with_64-bit_Windows_and_Workarounds Aug 01 16:25:00 thanks, and .. Aug 01 16:25:38 Okay by reading that thread again I now know it was libusb-win32, again any links? Aug 01 16:29:00 again did you read it properly? The link is right there: http://sourceforge.net/projects/libusb-win32/files/latest/download Aug 01 16:29:09 ;) Aug 01 16:31:40 There is something really wrong with me, I still couldnt find it, may be I am not looking well enough, the second google link get me to this same page. Aug 01 16:31:57 but thank you for your assistance.. Aug 01 16:46:09 RzR: ffx phone was released in Venezuela today :) twitter.com/mozillaVe Aug 01 18:05:05 clear Aug 01 18:34:49 bef0rd, yes and http://rzr.online.fr/q/firefoxos# Did @mozilla actually said that in #Brazil there is still no date for the release or price ? #FireFoxOs ? Aug 01 18:56:17 kerio: hmm? Aug 01 19:01:32 pypy for ARM :D Aug 01 19:01:40 build pls Aug 01 19:01:59 wtf is that? Aug 01 19:02:25 python? Aug 01 20:26:07 yup Aug 01 20:26:27 a JITted python implemented in a subset of python Aug 01 20:54:11 *burp* Aug 01 21:55:19 freemangordon: DocScrutinizer05: sysop-team: Top marks on fixing autobuilder :) thank you. Aug 01 21:55:54 I have successfully promoted two packages to testing. Aug 01 21:56:30 Also ML was a good idea. Aug 01 22:32:51 hmm... the internal resistance of my battery has increased enough to trigger low battery on high load... then after some rest, the battery gauge goes back to where it was before the spike... Aug 01 22:52:44 any n900 news i should know about, ive not looked in here in years Aug 01 22:52:57 still using my n900 though Aug 01 22:53:37 im interested in speed related tweeks Aug 01 23:08:14 Psi: cssu-thumb Aug 01 23:16:15 freemangordon: you running it? Aug 01 23:16:37 Psi: but ofc, I am the maintainer :D Aug 01 23:16:49 ~cssu-thumb Aug 01 23:16:49 [cssu-thumb] [thumb2 microb] indeed seems to render like mad, subjectively, or http://talk.maemo.org/showpost.php?p=1220597 Aug 01 23:16:51 ah, hehe Aug 01 23:18:37 i think i really need to reinstall everything, its been running for years and all sorts of crap has been installed Aug 01 23:42:42 dont cry when I'm nor answering Aug 01 23:42:49 I'll take an outtime Aug 01 23:43:52 maybe a week, or a month, or... Aug 01 23:44:09 DocScrutinizer05: hmm? Aug 01 23:44:17 we have a query to execute on mo :) Aug 01 23:44:53 no day off before that :P Aug 01 23:44:55 sorry Aug 02 00:03:01 Well.. that was rather abrupt. Aug 02 01:22:06 DocScrutinizer51: you there? Aug 02 01:36:41 DocScrutinizer05: you there? Aug 02 01:53:54 anyone.. here? Aug 02 02:05:28 * nox- hopes doc isnt burned out Aug 02 02:24:42 freemangordon: cssu-testing consistently does not work with me. :-( Aug 02 02:24:53 that means no camera-ui2 :-( **** ENDING LOGGING AT Fri Aug 02 02:59:59 2013