**** BEGIN LOGGING AT Thu Sep 13 03:00:00 2012 Sep 13 07:09:25 How to usb networking? I'm following the instruction on maemo wiki http://wiki.maemo.org/USB_networking, don't know how to install usb-otg-plugin, apt-get install can't find it Sep 13 07:27:24 Skry: is it fixable? Sep 13 07:41:41 Hello. How to usb networking? I'm following the instruction on maemo wiki http://wiki.maemo.org/USB_networking, don't know how to install usb-otg-plugin, apt-get install can't find it Sep 13 07:42:55 http://wiki.maemo.org/N900_USB_networking Sep 13 07:43:02 those instructions are for the N8x0 Sep 13 07:43:45 kerio, thank you, will read it Sep 13 07:44:01 anyway, there's really not much to it, if you just want to make the n900 and the computer communicate Sep 13 07:44:07 make sure g_nokia is loaded, ifup usb0 Sep 13 07:44:21 and then configure the network on the pc side Sep 13 07:44:26 if you're on windows, HAH Sep 13 07:45:29 yes, on the windows :) Sep 13 08:14:08 kerio, I have installed MADDE and edited the files of pcsuite-*.sh Sep 13 08:14:19 how to load g_nokia and ifup usb0 ? Sep 13 08:14:30 g_nokia is loaded whenever you enter pc suite mode on the n900 Sep 13 08:14:47 (you don't have to *be* in pc-suite mode, just that it should be the last mode you entered) Sep 13 08:14:52 I have connected n900 via pc suite mode, but nothing happened... Sep 13 08:14:53 lsmod | grep g_ Sep 13 08:15:04 it's going to be either g_nokia or g_file_storage Sep 13 08:15:15 (or g_ether, i think) Sep 13 08:15:27 kerio, yes, it has been loaded Sep 13 08:15:38 g_nokia 30376 6 Sep 13 08:15:40 then find a way to kick windows into shape Sep 13 08:15:45 not sure if there's one Sep 13 08:15:57 the n900 should appear as a network card on your computer's side Sep 13 08:16:00 kerio, what do you mean "kick windows into shape"? Sep 13 08:16:07 install drivers or something Sep 13 08:16:08 idk Sep 13 08:16:20 follow the wiki Sep 13 08:16:25 from the n900's side, sudo ifup usb0 Sep 13 08:16:46 I have installed MADDE and usb networking driver Sep 13 08:17:53 maybe I need a reboot? I will try Sep 13 08:20:02 I dont understand "This will only work if you got MADDE installed and configured for "Windows network" Sep 13 08:20:07 How to configure? Sep 13 08:20:47 I just installed MADDE Sep 13 08:20:48 No gui configure tools... Sep 13 08:32:13 ok, after restart windows, still no luck... Sep 13 08:37:42 I haven't done this step I think: got MADDE installed and configured for "Windows network" Sep 13 08:37:55 I have got MADDE installed, but I don't know how to conifgure it Sep 13 08:42:28 o/ Sep 13 08:58:34 I can ping 192.168.2.15, but I can't ping google.com, I have set up the resolv.conf Sep 13 10:55:02 hi, it is possible to get reason why/who stopping init.d/event.d script? Sep 13 10:55:35 I'd like to add option, if script is stopped by reboot/shutdown command or by user Sep 13 10:56:27 or I'd like to handle SIGINT/SIGTERM command and decide what to do if system is in shutdown/reboot state... Sep 13 10:56:36 DocScrutinizer05, see ^^^ Sep 13 11:06:00 /sbin/runlevel is shell script which print only "N 2" Sep 13 11:06:21 so it is unusable on upstart systems (like maemo) Sep 13 11:18:35 hm, what's the stuff in /etc/event.replace.d? Sep 13 11:35:57 kerio: is what fixable? Sep 13 11:36:16 Skry: the voltage confusions Sep 13 11:37:18 sure, for now it is just a matter of reverting one commit, so cpufreq can round voltages so it can find a match from OPP Sep 13 11:37:36 what would that entail? Sep 13 11:37:46 SR would be stable? Sep 13 11:38:32 No, SR is still unstable, it just makes cpufreq work without lockin up Sep 13 11:39:50 Skry: which kernel is that? Sep 13 11:40:14 freemangordon: 3.6 and 3.5 Sep 13 11:40:32 on n900? Sep 13 11:40:34 yes Sep 13 11:41:20 i see. well, according to what I know, SR cannot be made stable using what TI has programmed in efuse Sep 13 11:41:33 it is just garbadge Sep 13 11:42:14 :( Sep 13 11:42:22 true, with, n900 there is one certain voltage value which makes the device lockup Sep 13 11:42:54 Skry: the problem is that all devices have the same calibrations for OPP3-OPP5 Sep 13 11:43:57 Skry: the good news is that calibrations for OPP1 and OPP2 are ok ;) Sep 13 11:44:19 is it possible to install rootsh on n900 without internet? Sep 13 11:45:03 Skry: so the rest can be extrapolated. look at my SR patch in KP if you are interested Sep 13 11:45:10 akls: yeah Sep 13 11:45:13 but rootsh is bad Sep 13 11:45:14 any netsplit detected? Sep 13 11:45:16 :c Sep 13 11:45:23 DocScrutinizer05: kinda big netsplit, yes Sep 13 11:45:32 kerio, it's bad? then what should I use? Sep 13 11:46:10 freemangordon: I'm not that interested in SR, got the cpufreq fixed so that should be enough for now Sep 13 11:46:10 akls: well, to be fair the intended way is to just enter r&d mode Sep 13 11:46:13 * DocScrutinizer05 wonders whether to reboot infobot Sep 13 11:46:19 kerio, how? Sep 13 11:46:29 akls: nah, that's a silly thing Sep 13 11:46:56 * kerio ponders about making a tiny .deb that installs a file in /etc/sudoers.d/ Sep 13 11:47:11 kerio, sudo flasher --enable-rd-mode ? Sep 13 11:47:22 Skry: well, it is up to you, but SR is very very good thing to have working. Sep 13 11:47:39 freemangordon: wait, so is SR usable or not? Sep 13 11:47:48 kerio: where? Sep 13 11:47:53 on my n900 Sep 13 11:48:04 does it work? Sep 13 11:48:06 DocScrutinizer: infobot joined! Sep 13 11:48:21 it used to Sep 13 11:48:36 then i disabled it, because everyone says that it's useless and/or instable Sep 13 11:48:54 akls: yep Sep 13 11:48:58 kerio: now, who is that "everyone"? Sep 13 11:49:09 "sudo gainroot" is always doable, but it checks for rd-mode Sep 13 11:49:26 once you have a root shell, i'd manually add your own sudoers file Sep 13 11:49:31 and then disable rd-mode Sep 13 11:49:52 freemangordon: every sentient being in the universe Sep 13 11:50:04 or maybe someone on the wiki Sep 13 11:51:40 kerio: please stop trolling akl Sep 13 11:51:44 freemangordon: I have it enabled in "my" kernel so if user wants to use it he can enable it via sysfs, however, when engineers from both TI and Nokia say it is unstable and broken on 3430, it is unstable. Besides, I dont see any logic of using two separate methods to scale voltages at the same time. Sep 13 11:51:49 akls even Sep 13 11:51:51 DocScrutinizer05: no, seriously, it works Sep 13 11:52:14 check /usr/sbin/gainroot and /etc/sudoers.d/01sudo Sep 13 11:52:42 kerio: how would you enable r&d flag? Sep 13 11:52:46 the flasher? Sep 13 11:52:57 on device??? Sep 13 11:53:03 from a computer Sep 13 11:53:10 yeah :-/ Sep 13 11:53:16 so? Sep 13 11:53:30 everybody got rootsh installed Sep 13 11:53:33 not me Sep 13 11:53:43 it's the reason i won't install qtmobilehotspot Sep 13 11:53:48 it depends on that POS Sep 13 11:54:02 i really hope it doesn't pipe commands to it Sep 13 11:55:12 akls: you might want to sanitize rootsh by enabling proper password query for commands "root" and "sudo gainroot". See Sep 13 11:55:15 ~jrtools Sep 13 11:55:15 methinks jrtools is http://wiki.maemo.org/User:Joerg_rw/tools Sep 13 11:56:46 kerio: I hope you checked qtmobilehotspot and gave it proper rating in testing so it never makes it into maemo-extras Sep 13 11:57:02 hmm... no i did not Sep 13 11:57:52 plenty of packages depend on rootsh Sep 13 11:58:13 then plenty of packages deserve demoting Sep 13 11:58:25 including stuff like "optimizen900", "clean900", "nitroid-installer" Sep 13 11:58:28 * kerio lols Sep 13 11:58:44 * DocScrutinizer05 pukes Sep 13 11:59:08 lrn2sudo, fools!!! Sep 13 11:59:41 how do i vote down? Sep 13 12:00:03 I'm tempted to roll a new version of rootsh that does things proper, incl asking for password, and then watch all that crap go down epically Sep 13 12:00:24 once stuff is in extras, you can't Sep 13 12:00:24 chinese li-ion charger==USB 5V charger with 0.8v Zener across the output. Sep 13 12:00:35 stuff in testing has rating buttons Sep 13 12:00:41 qtmobilehotspot is in extras-testing and extras-devel Sep 13 12:00:53 Skry: well, we have at least 2 proves that what YI and Noika say is not alwasy true - USB hostmode and thumb2 Sep 13 12:01:00 Relying on USB port maximum current to control current. Sep 13 12:01:02 vi_: /o\ Sep 13 12:01:26 DocScrutinizer05: http://maemo.org/packages/package_instance/view/fremantle_extras-testing_free_armel/qtmobilehotspot/0.3.5/ Sep 13 12:01:28 vote down plskthx Sep 13 12:01:42 kerio: then go to extras-testing, add comment explaining why you give it thumbs-down Sep 13 12:01:43 Skry: what is the other method to control voltages? Sep 13 12:02:13 freemangordon: cpufreq does both Sep 13 12:02:19 kerio: I first need to check it before voting it down Sep 13 12:02:33 DocScrutinizer05: Depends: [...] rootsh [...] Sep 13 12:02:40 where? Sep 13 12:02:41 Skry: aah, I see. But SR does a totally different thing Sep 13 12:03:10 DocScrutinizer05: in that page Sep 13 12:03:14 qtmobilehotspot 0.3.5 Sep 13 12:03:18 o.O Sep 13 12:03:22 >Promotion unlocked, waiting for maintainer to promote Sep 13 12:03:24 oh hell naw Sep 13 12:03:29 I am late to the party, what is the argument? Sep 13 12:03:37 vi_: packages that depend on rootsh Sep 13 12:03:41 Also, SR and is stable and thumb rules. Sep 13 12:04:03 In addition to this, rootsh works fine. Deal with it. Sep 13 12:04:18 DocScrutinizer05: qradio depends on rootsh too, and that's in extras Sep 13 12:04:34 vi_: just a note: SR is stable with KP ;) Sep 13 12:04:35 vi_: rootsh is a massive kludge Sep 13 12:05:40 freemangordon: Thats what I meant! Sep 13 12:05:46 kerio: IDGAF. Sep 13 12:06:00 Maemo5==A MASSIVE KLUDGE. Sep 13 12:06:04 freemangordon: yeah, I know that, k hilman just convinced me that when cpufreq does the voltage scaling itself, benefits from SR would be minimal -> not worth the instability. Anyways, you were saying that problems with SR are caused by the same reason I had trouble with cpufreq? Sep 13 12:06:24 guys, I've just bought my phone Sep 13 12:06:31 and wireless is not working out of the box Sep 13 12:06:41 akls: it should Sep 13 12:06:49 akls: anyway, have you installed the latest PR from nokia yet? Sep 13 12:06:52 I simply can't see any networks. My friend bought a phone with me - everything is fine on his phone Sep 13 12:06:53 PR1.3 Sep 13 12:06:56 kerio, yes. Sep 13 12:07:00 akls: Is that supposed to be a question or an allusion to disgruntlement? Sep 13 12:07:12 vi_: oh shush Sep 13 12:07:31 akls: which region is your n900 from? Sep 13 12:08:10 friggin testing pages always make me want to click "send" again, after 60s of waiting Sep 13 12:08:30 akls: Re-flash to stock pr1.3 firmware to eliminate the possibillity of software and try again. If it still does not work, take it back. Sep 13 12:08:33 akls: tried rebooting already? Sep 13 12:08:35 kerio, how could I know, bought it from ebay Sep 13 12:08:46 kerio, of course! Sep 13 12:09:22 akls: have you done a full reflash? Sep 13 12:09:23 kerio, what's a "FULL" reflash? Sep 13 12:09:26 flashed both VANILLA and COMBINED with flasher-3.5 Sep 13 12:09:28 any supertester around to vote down http://maemo.org/packages/package_instance/view/fremantle_extras-testing_free_armel/qtmobilehotspot/0.3.5/ ? Sep 13 12:09:34 SpeedEvil: ^^^? Sep 13 12:11:57 DocScrutinizer05: how did rootsh end up in extras anyway? Sep 13 12:12:08 it has all the signs of a bad, bad, bad package Sep 13 12:12:09 that's historical Sep 13 12:12:29 see who did it, and when Sep 13 12:13:02 Maintainers: Faheem Pervez, Pali Rohár Sep 13 12:13:05 Pali: oh you Sep 13 12:13:12 tbh there's worse crap around: sudser etc Sep 13 12:13:32 yes, I promoted new patched version to extras Sep 13 12:13:42 kerio: pali kinda fixed worst borkage of it Sep 13 12:13:47 iirc Sep 13 12:13:47 in extras was old version which has broken postrm Sep 13 12:15:39 hm Sep 13 12:16:46 kerio: Pali: damage done by Faheem Pervez. All we can do now is roll a better-rootsh that also secures HAM via hildon-su and thus allows to really kinda get some little bit of sanity & security back on fremantle Sep 13 12:17:30 hell, it can even conflict with rootsh (original) Sep 13 12:17:40 no Sep 13 12:17:44 no? Sep 13 12:17:46 original rootsh is not packages Sep 13 12:17:49 *packaged Sep 13 12:17:50 :D Sep 13 12:17:57 errr Sep 13 12:18:06 original rootsh is only in FIASCO rootfs image Sep 13 12:18:19 you mean it's not a proper package with uninstall etc? Sep 13 12:18:25 preinst script moving existing rootsh to new location Sep 13 12:18:28 meh Sep 13 12:18:34 and postrm moving old roosh back Sep 13 12:18:39 I mean the rootsh from repo Sep 13 12:18:51 DocScrutinizer05: you mean gainroot? Sep 13 12:18:52 this doing extras rootsh package Sep 13 12:19:04 the package called rootsh Sep 13 12:19:26 rootsh is package which provide gainroot Sep 13 12:19:28 http://maemo.org/packages/package_instance/view/fremantle_extras-testing_free_armel/rootsh/1.8/ Sep 13 12:19:44 not really, gainroot is provided by sudo Sep 13 12:21:17 I'd really appreciate some help on creating a clean way to secure HAM, preferrably in a generic way by using some hildon-su Sep 13 12:21:39 DocScrutinizer05: are we allowed to modify HAM? Sep 13 12:21:56 why not Sep 13 12:22:11 cssu already did it, hell - pali did it Sep 13 12:22:38 DocScrutinizer05: we need a better sudo Sep 13 12:23:02 I want a fremantle system with proper root password, and no way to install shit like rootsh that neuters it Sep 13 12:23:28 DocScrutinizer05: better sudo with askpass support (and maybe sudoers.d support so we don't need update-sudoers) Sep 13 12:23:49 I'm all for getting better (newer) sudo, that relieves us from pita of update-sudoers abomination Sep 13 12:24:04 and then make HAM call sudo -A appropriate-graphical-password-asking-program apt-worker whatevs Sep 13 12:24:27 :nod: Sep 13 12:24:37 so proficient users can remove the passwordless sudo on apt-worker, and then it'll just ask for the password Sep 13 12:24:40 Pali: you're still with us? :-) Sep 13 12:25:00 Moo,Doc Sep 13 12:25:04 Ah, fun, fun Sep 13 12:25:06 moo RST38h Sep 13 12:25:23 DocScrutinizer05: actually it can be done without modifying HAM Sep 13 12:25:29 we do need a password-asking binary though Sep 13 12:25:44 call that hildon-su Sep 13 12:26:02 DocScrutinizer05: no, just a password asking Sep 13 12:27:14 password asking, email sending binary Sep 13 12:27:36 DocScrutinizer05: in fact, with a good sudo that knows about the graphical askpass we don't need *any* modifications to HAM apart from the removal of NOPASSWD Sep 13 12:28:04 what can be better than that? Sep 13 12:28:34 kerio: we discussed all this some days ago. There's zenity for now, and we concluded we'd need some modifications to HAM (to include -a, no hildon-su parameter needed since that can get passed by ENV), and we need sudo that supports -a Sep 13 12:28:57 can zenity ask for a password? Sep 13 12:29:05 yes Sep 13 12:29:11 brilliant Sep 13 12:29:24 even hides input Sep 13 12:30:14 zenity --help-entry Sep 13 12:30:48 it's trivial to make a zenity-askpass that does what sudo expects, yes Sep 13 12:30:55 zenity --entry --hide-text --text="please enter root password" Sep 13 12:31:40 kerio, ham alternative that directly uses apt? Sep 13 12:31:45 what sudo *would* expect - if sudo wasn't prehistoric crap Sep 13 12:32:47 but you already remarked a few days ago that this method would ask for root password on each new interactive operation user does on HAM Sep 13 12:33:06 DocScrutinizer05: nope Sep 13 12:33:13 we exploit sudo's timeout Sep 13 12:33:21 and it still needs HAM to use "sudo -a apt-worker" instead of "sudo apt-worker" Sep 13 12:33:38 kerio: ok, feasible Sep 13 12:34:00 DocScrutinizer05: modern sudos automatically call askpass if they're not on a tty Sep 13 12:34:12 OOOH! Sep 13 12:34:16 :-) Sep 13 12:34:30 and you can specify askpass in sudoers Sep 13 12:34:35 yep Sep 13 12:34:44 it's just an envvar Sep 13 12:34:46 sorry I confused that with ENV Sep 13 12:35:45 so who's going to provide sudo update for cssu? Sep 13 12:35:50 cssu Sep 13 12:35:52 easy Sep 13 12:36:25 meh, cssu is a project. it doesn't magically breed new packages Sep 13 12:36:39 ok, the last person to say "not me" on irc Sep 13 12:36:40 not me Sep 13 12:36:41 we need somebody willing to create and maintain new sudo Sep 13 12:38:23 we also need further review of that concept. I'm not that self-confident to claim I already figured each possible security flaw it might hide Sep 13 12:39:18 kerio: you already mentioned things like direct invocation of sudo apt-worker when I pondered to run HAM under root permissions, with hildon-su wrapper Sep 13 12:39:42 DocScrutinizer05: well, it would be the same Sep 13 12:39:54 so at worst we have the same security Sep 13 12:40:12 we need proper review of our concept, by a few other sensible devels Sep 13 12:40:52 kerio: I'm not going to mess with all that shit to end with "same security as before", you know ;-) Sep 13 12:41:19 DocScrutinizer05: well, the security will increase if the user removes NOPASSWD Sep 13 12:41:28 the purpose of this effort is to improve security Sep 13 12:41:41 yeah but there's not even the *concept* of a user password in fremantle Sep 13 12:41:46 by default Sep 13 12:41:53 root password Sep 13 12:42:05 yeah but there's not even the *concept* of a password in fremantle Sep 13 12:42:18 in hildon, right Sep 13 12:42:26 in fremantle, sure there is Sep 13 12:42:40 DocScrutinizer05: not as presented to the user Sep 13 12:42:58 sure, see ssh root@n900 Sep 13 12:43:20 WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! Sep 13 12:43:21 lol wat Sep 13 12:43:30 (ssh root@localhost) Sep 13 12:43:52 kerio: rogue boy! Sep 13 12:43:57 wtf happened Sep 13 12:43:59 no, seriously Sep 13 12:44:07 you reflashed? Sep 13 12:44:38 i never reflash only the root Sep 13 12:45:24 well, for sume reason root's publickey doesn't match user's list of known_hosts anymore Sep 13 12:45:47 it's not root's publickey, it's sshd's keypair Sep 13 12:45:49 either of both has changed Sep 13 12:46:14 fool, sshd has no own key pair Sep 13 12:46:15 second weird thing: this is the first time i ssh localhost Sep 13 12:46:19 i'm kinda sure of this Sep 13 12:46:37 DocScrutinizer05: /etc/ssh/ssh_host_key[.pub] Sep 13 12:46:40 it's the publickey of root's ssh keypair Sep 13 12:46:46 those are keys, and they're two Sep 13 12:46:59 wtf?? Sep 13 12:47:15 today on #maemo: DocScrutinizer learns about sshd Sep 13 12:47:28 meh Sep 13 12:47:39 * DocScrutinizer05 goes getting a coffee Sep 13 12:47:46 there's not even a /root/.ssh/ Sep 13 12:47:54 ooh? Sep 13 12:48:10 i don't have it, i never ran ssh as root \_o_/ Sep 13 12:48:17 IroN900:~# ls -l ~root/.ssh/ Sep 13 12:48:19 insgesamt 8 Sep 13 12:48:20 -rw-r--r-- 1 root root 1115 2007-11-15 21:07 authorized_keys Sep 13 12:48:22 -rw-r--r-- 1 root root 2136 2012-04-22 12:50 known_hosts Sep 13 12:48:39 sure, but that's unrelated to what sshd does Sep 13 12:49:12 sure, that's why the folder is named .ssh Sep 13 12:49:22 ssh*d* Sep 13 12:49:35 pff Sep 13 12:49:56 guess where sshd will look for authorized_keys when I log in a s root Sep 13 12:50:30 guess how related *that* is to the identification of the server by ssh Sep 13 12:50:51 ssh's identification goes both ways Sep 13 12:53:58 shit Sep 13 12:54:09 guess where sshd will look for authorized_keys when I log in a s root Sep 13 12:54:13 you're right that's not the host authentication key Sep 13 12:54:32 [Notice] -christel- [Global Notice] Hi all, not only would it seem that Christmas came early this year, we must also have been much nicer than we thought! We're receiving a steady stream of packets at present and apologise for any inconvenience whilst we try work around the issue! Again, sorry and thank you for using freenode! Sep 13 12:56:34 xmas packets Sep 13 12:58:56 yeah, kicked me out Sep 13 12:59:15 kerio: still alive? Sep 13 12:59:23 nope :( Sep 13 12:59:26 a sniper killed me Sep 13 12:59:29 hehe Sep 13 12:59:32 now i inhabit my laptop Sep 13 12:59:51 * DocScrutinizer05 sends hypersnyper Sep 13 13:00:10 anyway, yeah, you were wrong :3 Sep 13 13:00:31 yeah, publicly admitting you were right about auth key Sep 13 13:00:55 ~kerio Sep 13 13:00:56 you are probably [...] kerio is correct [...] Sep 13 13:01:19 the power of ellipses! Sep 13 13:02:07 you however were wrong about ~/.ssh being unrelated to sshd Sep 13 13:02:25 well of course, authorized_keys is to login with a keypair Sep 13 13:02:41 since ~/.ssh/authorized_keys evidently is used by sshd Sep 13 13:03:47 now for something completely new and different: hildon-su and contemporary proper sudo binary Sep 13 13:04:46 for sudo binary I bet any increase in size will be ballanced by reduced size of /etc/sudoers messed up file Sep 13 13:04:49 hildon-askpass Sep 13 13:04:57 crap Sep 13 13:05:42 I already told you (and you noticed it) that hildon-askpass has no option to switch it to only one password textbox Sep 13 13:06:03 shall I scan the chanlog? Sep 13 13:06:15 no, that's maemo-ask-password or something Sep 13 13:06:23 ooh Sep 13 13:07:17 for hysterical raisins that class of programs is named -su, or su: KDEsu Sep 13 13:07:22 k Sep 13 13:07:48 can't we just get gksu compiled for maem? Sep 13 13:07:50 *maemo Sep 13 13:08:14 i still don't get which toolkit we're supposed to use on the n900 Sep 13 13:09:50 i'd like to know that too Sep 13 13:10:45 obviously on fremantle we're free to use gtk or qt or qtm Sep 13 13:11:12 yeah but which is the recommended one Sep 13 13:11:18 with each of them supported by a number of toolkits, like pyqt Sep 13 13:11:44 recommended? seems qt is most preferred one, generally Sep 13 13:15:06 Pali: would you see any chance you throw "new" sudo into maemo make machine? Sep 13 13:15:26 sorry, no time for other project Sep 13 13:15:31 k Sep 13 13:17:32 * DocScrutinizer05 idly wonders if it will work to simply scp user@oldbox/scratchbox / Sep 13 13:17:47 rather, if scratchbox will work on new machine Sep 13 13:18:30 (given both machines use basically same distro) Sep 13 13:23:29 hm Sep 13 13:23:36 just updated my n900 Sep 13 13:23:46 and now it doesnt want connect to internet Sep 13 13:23:49 :/ Sep 13 13:27:22 says No connections availible Sep 13 13:31:30 dhbiker: reboot? Sep 13 13:31:42 already did Sep 13 13:32:17 updated from ? to ? Sep 13 13:33:15 cssu testing to 21.2011.38-1Tmaemo5.1 Sep 13 13:33:54 dhbiker: you observerd [general notice]? Sep 13 13:34:28 [GENERAL NOTICE] if you're using CSSU-T, do NOT use "update all" in HAM, you need to first install/update the opername widget, only *then* go update the rest. Read http://talk.maemo.org/showthread.php?p=1264963#post1264963 ff Sep 13 13:35:21 dhbiker: your "settings -> about_CSSU" shows 21.2011.38-1Tmaemo5.1? Sep 13 13:37:07 i did so Sep 13 13:37:12 opername widget Sep 13 13:37:14 then cssu Sep 13 13:37:19 and yes Sep 13 13:39:08 so, after dis-assembling this chinese USB charger I have found the 5V output is wired backwards. Sep 13 13:39:33 vi_: lol Sep 13 13:39:36 Yes, wired backwards. Sep 13 13:39:40 :D Sep 13 13:40:06 What 'engineer' comes up with this shit? Sep 13 13:40:21 DocScrutinizer05: so this means restore the backupmenu backup ? -.- Sep 13 13:40:27 the PRC didn't get this far by double-checking the connectors Sep 13 13:40:31 How can they look their wives in the face and say, 'yes, I had a good day today honey'. Sep 13 13:40:39 dhbiker: not really Sep 13 13:41:00 i have no idea how to diagnose an ICD2 problem Sep 13 13:41:14 probably the easiest wa Sep 13 13:41:15 y Sep 13 13:41:19 dhbiker: I don't see why your system has problems Sep 13 13:41:47 anyway restoring BM backup and restarting all over again seems a sane idea Sep 13 13:42:00 k Sep 13 13:42:11 copying the backup to phone Sep 13 13:42:46 dhbiker: you know this should be done in BM as well? Sep 13 13:42:52 I just threw out a rattly 18650 cell. Sep 13 13:42:53 oh ? Sep 13 13:42:57 Yes, rattly. Sep 13 13:43:11 They had put a tiny battery in an 18650 size tube. Sep 13 13:43:24 LOL Sep 13 13:43:36 seen similar crap every now and then Sep 13 13:44:05 very popular: use AAA NiMH cell in a A case Sep 13 13:44:17 vi_: that's hilarious Sep 13 13:44:22 lolol Sep 13 13:44:58 I knew these batteries were utter shit as soon as I saw the '3800mAH' written on the side. Sep 13 13:45:01 But this? Sep 13 13:45:30 My brain cannot handle bullshit of this magnitude. Sep 13 13:45:36 lol Sep 13 13:46:07 they used to wrap it in paper to avoid rattling Sep 13 13:46:18 DocScrutinizer05: backupmenu only charges with .5A, right? Sep 13 13:46:23 There is an economic depression on you knw. Sep 13 13:46:30 Cannot waste good paper these days. Sep 13 13:46:32 kerio: probably Sep 13 13:46:52 yeah, otherwise there would be no way to use usb MSM and stuff like that Sep 13 13:47:21 vi_; some people would argue anyone buying a "3800 mAh" 18650 deserves to have it blow off their hand Sep 13 13:47:54 vi_: your brain refuses to acknowledge they pulled a fast one on you Sep 13 13:47:59 * kerio buys an AAA in a 18650 case marked as "very original fusion power infinite battery" Sep 13 13:48:04 I did not specifically buy it, it came with some random tat I bought from China. Sep 13 13:48:06 I BET THIS WILL WORK GUISE Sep 13 13:49:04 But I do recognise them as the batteries that have exploded from a thread on a flashlight forum. Sep 13 13:49:17 In that case the flashlight had 3 18650 in series. Sep 13 13:49:32 One cell had reversed polarity. Sep 13 13:49:33 err the one that nuked the balkony? Sep 13 13:49:51 DocScrutinizer05: No, it nuked some old guys hands. Sep 13 13:49:55 :c Sep 13 13:49:58 poor dude Sep 13 13:50:00 :-(( Sep 13 13:50:32 Severe burns, nothing more. Sep 13 13:50:32 vi_: link? Sep 13 13:50:38 let me try. Sep 13 13:51:03 http://www.candlepowerforums.com/vb/showthread.php?280909-Ultrafire-18650-3000mA-exploded Sep 13 13:53:22 Whenever I see something like that in a forum, I'm reminded of the alleged samsung s3 that exploded Sep 13 13:53:29 FUUUUUUUU...ck, sftp oldbox /scratchbox to newbox. oldbox /scratchbox: 4.7G. newbox /scratchbox: 5GB and counting :-/ Sep 13 13:53:33 when really, the guy dunked it into the water. Sep 13 13:53:42 DocScrutinizer05: i heard you like symlinks Sep 13 13:53:48 AND THEN thought it would be a good idea to connect it to a powerline Sep 13 13:53:49 or hardlinks Sep 13 13:53:52 and bindmounts :) Sep 13 13:53:59 teotwaki: Seriously? Sep 13 13:54:09 aye Sep 13 13:54:14 He accused samsung of selling faulty devices Sep 13 13:54:17 blogged about it Sep 13 13:54:29 Samsung sent a VP over to give him a new one Sep 13 13:54:40 no bindmounts, scratchbox not started Sep 13 13:54:58 And then, samsung, looking closely at the device, said "Wait a minute, this device didn't explode" Sep 13 13:55:08 I know people in general do not have a clue about electronics but you think they would grasp electricity+water==mega fail. Sep 13 13:56:45 and >> -r Recursively copy entire directories when uploading and downloading. Note that sftp does not follow symbolic links encountered in the tree traver- sal. Sep 13 13:56:53 there we go Sep 13 13:57:00 restored backup Sep 13 13:57:08 now to the update part again xD Sep 13 14:02:49 DocScrutinizer05: \_o_/ Sep 13 14:02:57 indeed Sep 13 14:03:08 transparent compression? Sep 13 14:03:28 4.7gb -> 6.5GB and counting Sep 13 14:04:08 how many hardlinks could be in there that du actually notices? Sep 13 14:04:35 idk Sep 13 14:04:37 a lot, apparently Sep 13 14:04:51 most importantly, how many hardlinks are in there that you're not remaking here? Sep 13 14:05:12 aiui sftp doesn't even know about hardlinks Sep 13 14:05:29 hardlinks are pretty hard to handle Sep 13 14:05:32 ;-) Sep 13 14:06:28 there's no pointer on inode entry A->(somefile) to tell about arbitrary B->(samefile) Sep 13 14:07:10 basically you have to check *all* inode entries on that volume, for those that have same data inode number Sep 13 14:07:52 only clue is counter about total number of inode entries pointing to that file Sep 13 14:09:18 so, to copy, your program needs to enter each inode number of a file that has inode count >1 to a list, and before that check if the list already has same inode number for another file Sep 13 14:09:44 in that case don't copy data but rather create hardlink to the already existing other file Sep 13 14:11:59 does scp do that? Sep 13 14:12:57 tar knows about them Sep 13 14:13:01 FAIL, maemo system can be unbootable after calling reboot/halt/poweroff/shutdown Sep 13 14:13:42 problem is that dsme is exited abnormal if is killed by init in shutdown process Sep 13 14:13:58 DocScrutinizer05: btw, why are you using scp? Sep 13 14:14:03 this is a local transfer, right? Sep 13 14:14:26 I saw this problem, sometimes when I called sudo reboot phone was rebooted and then somebody turned it off before X started Sep 13 14:14:30 Pali: damn, are you sure about that? Sep 13 14:14:52 dsme sending broadcast messages that system going offline Sep 13 14:15:03 kerio: sftp Sep 13 14:15:15 and some processes like hildon-* doing some stuff on it Sep 13 14:15:24 and dsme storing STATE file Sep 13 14:15:26 DocScrutinizer05: what happened to tar jv dir | nc foo 31337? Sep 13 14:15:49 when soembody tell that DSME should shutdown system Sep 13 14:15:51 by the way, why isn't /tmp a tmpfs? Sep 13 14:16:01 it fell apart due to lack of inner coherence Sep 13 14:17:01 kerio: (tmpfs) maybe because we're already short on free memory? Sep 13 14:17:04 I'm suggesting to create wrapper around UPSTART /sbin/shutdown utility which will call DSME reboot/shutdown function Sep 13 14:17:13 DocScrutinizer05: and we can't spare 100K? Sep 13 14:17:15 how do you enable front facing camera to be on for surveillance Sep 13 14:17:34 right now my /tmp is 84K Sep 13 14:18:00 open a pdf in browser and see what your tmp is like, then Sep 13 14:18:13 1349 < divVerent> in busybox it's a forkbomb too Sep 13 14:18:13 1350 < divVerent> change the 1024 to 1048576 for nice forkbomb effect ;) Sep 13 14:18:13 1356 < divVerent> #!/bin/busybox sh Sep 13 14:18:13 1356 < divVerent> S=`yes | head -n 32769`; while :; do exec < 1356 < divVerent> $S Sep 13 14:18:15 1356 < divVerent> EOF Sep 13 14:18:18 1356 < divVerent> here it is, the busybox forkbomb :P Sep 13 14:20:05 hmm, nice, but... what it's good for now? Sep 13 14:20:37 and which 1024 to change? Sep 13 14:20:45 max-processes? Sep 13 14:21:10 http://dpaste.com/799895/ Sep 13 14:23:04 I like my forkbomb better Sep 13 14:23:36 sftp finished at du -sh /scratchbox ->6.5GB Sep 13 14:24:22 wtf is /sys/kernel/debug Sep 13 14:24:34 what's nodev? Sep 13 14:24:39 no device nodes allowed Sep 13 14:24:43 DocScrutinizer05, what do you think? /sbin/shutdown is upstart ELF utility. We can move this program to /sbin/shutdown.real and implement wrapper which call dsme shutdown/reboot function only once. Next time will be called real ulility Sep 13 14:24:47 jaska: nodev on /sys/kernel/debug type debugfs (0) Sep 13 14:24:52 you cant make things like tty1 or sda etc there Sep 13 14:25:06 debugfs is used for some twiddling of kernel stuff, not sure of the use on n900 myself Sep 13 14:26:16 fsckng sftp has a bug Sep 13 14:26:47 cough http://i.imgur.com/KfDoU.jpg Sep 13 14:39:01 LOL Sep 13 14:39:05 but kinda true Sep 13 14:39:40 * DocScrutinizer05 wonders what type of touchpanel digitizer they use on hooker920 Sep 13 14:55:12 it's driving me nuts.trying to update the cssu and i can't seem to get passed making a backup. then when i go back to the update,it prompts me to make a backup Sep 13 14:55:30 read GENERAL NOTICE :) Sep 13 14:55:38 tl;dr: update the operator name widget first Sep 13 14:55:47 oh Sep 13 14:57:27 tadzik: thanks Sep 13 14:57:35 you're welcome Sep 13 15:04:45 [GENERAL NOTICE] if you're using CSSU-T, do NOT use "update all" in HAM, you need to first install/update the opername widget, only *then* go update the rest. Read http://talk.maemo.org/showthread.php?p=1264963#post1264963 ff Sep 13 15:05:07 * tadzik wonders if the trailing "ff" is related to "ffs" Sep 13 15:05:39 it's a common (latin?) tla(?) Sep 13 15:06:05 f: the one that follows. ff: the following (more than one) Sep 13 15:07:16 it's of same class like etc, et al ... Sep 13 15:07:36 nb Sep 13 15:07:49 sic Sep 13 15:09:46 http://en.wikipedia.org/wiki/List_of_Latin_abbreviations Sep 13 15:09:57 folio Sep 13 15:10:22 >>This abbreviation is used in citations to indicate an unspecified number of following pages following, Example: see page 258ff.<< Sep 13 15:17:24 tadzik: "ffs" is "for fuck's sake" Sep 13 15:17:33 completely unrelated to pages Sep 13 15:18:05 also fast file system, completely unrelated too Sep 13 15:24:23 this should fix reboot requests: https://gitorious.org/community-ssu/upstart/commit/a5a91a61aa9cba8af960b6a0dff083a233bc6200 Sep 13 15:24:46 dsme will see reboot request and send broadcast signal Sep 13 15:50:03 kerio: I know what it is, I wondered if "ff" is related to it Sep 13 16:24:00 Pali: ??? Sep 13 16:24:11 ??? Sep 13 16:24:23 your last 2 posts Sep 13 16:24:40 patch for that dsme Sep 13 16:24:40 (( this should fix reboot requests...)) Sep 13 16:25:04 see my previous posts in log Sep 13 16:25:09 I didn't know there's anything wrong with reboot Sep 13 16:25:27 problem is that DSME storing state to state file Sep 13 16:25:48 and if dsme got SIGINT/TERM signal then it does not store state Sep 13 16:26:16 and then you call reboot then dsme does not handle it and init daemon will kill dsme Sep 13 16:27:02 when you use dsmetool -b or other hildon method for reboot, it tell DSME that system going to shutdown/restart Sep 13 16:27:31 dsme will send broadcast messages to all client to save data and prepair for shutdown/reboot Sep 13 16:27:55 reboot command is deprecated Sep 13 16:28:11 where? Sep 13 16:28:39 reboot/poweroff/halt are still used on unixes Sep 13 16:28:49 not on fremantle Sep 13 16:30:03 it's known that reboot does a hard reboot Sep 13 16:32:05 that's why it's called mini-reboot I guess Sep 13 16:32:15 err minishutdown Sep 13 16:34:11 actually, if anything, dsme -b should call minishutdown (after finishing whatever dsme does), not the other way round Sep 13 16:42:34 no Sep 13 16:42:44 dsmetool -b doing "proper" reboot Sep 13 16:43:22 reboot tell upstart to start killing running daemons (like on desktop) Sep 13 16:43:42 but dsme badly handle SIGTERM and SIGINT signals Sep 13 16:44:02 and more application depends on dbus broadcast signal sent by dsme Sep 13 16:44:41 mini-reboot/mini-shutdown is called when reboot application cannot talk to upstart Sep 13 16:45:14 we only need to handle situation when somebody call reboot/shutdown to tell this also dsme Sep 13 16:45:23 and my patch fixing it for reboot Sep 13 16:45:51 reboot calling telinit 6 and here is added line dsmetool -b Sep 13 16:46:27 dsmetool does not have flag for "shutdown", but we can find dbus menthod which is used by "power off" button Sep 13 16:57:06 reboot is deprecated Sep 13 16:57:28 on fremantle Sep 13 16:57:53 this patch should fix it Sep 13 16:57:54 the hierarchy is dsme > upstart > init Sep 13 16:58:13 reboot is *meant* to do hard shutdown Sep 13 16:58:20 DocScrutinizer05: but upstart *is* init! Sep 13 16:58:41 kerio: I meant init.d/* scripts Sep 13 16:58:46 i see Sep 13 16:59:24 dsmetool -b doing this: change dsme state to reboot, send application dbus signal, tell upstart to reboot system Sep 13 16:59:37 reboot doing this: tell upstart to reboot system Sep 13 17:00:04 Pali: is upstart smart enough to ignore the command to reboot while it's already rebooting? Sep 13 17:00:05 yes, it's one level lower in hierarchy Sep 13 17:00:29 kerio, new upstart Sep 13 17:00:31 yes Sep 13 17:00:32 upstart is not aloowed to kill dsme Sep 13 17:00:51 upstart killing dsme Sep 13 17:00:53 unless you actually *want* hard shutdown Sep 13 17:03:10 ok Sep 13 17:03:29 Pali: yes, and a kill -SIGKILL 1 is even more hard than dsme getting killed by upstart. Yet it's not the recommended way to do things, like using reboot command is not recommended to do polite reboot Sep 13 17:03:32 no, hard shutdown is /etc/init.d/minishutdown Sep 13 17:03:54 wait, i can't "sudo reboot" and have it do a clean reboot? Sep 13 17:03:56 wtf Sep 13 17:04:06 yes Sep 13 17:04:17 you can test my patch Sep 13 17:04:21 kerio: reboot is deprecated Sep 13 17:04:30 use dsme -b Sep 13 17:04:32 DocScrutinizer05: it's fscking **reboot** Sep 13 17:04:46 is shutdown the same thing? Sep 13 17:04:48 I find it somewhat sick that you need a patch to make reboot work like it should Sep 13 17:04:56 it's pretty much same as on any desktop Sep 13 17:05:16 KDE won't handle nice any init 6 call Sep 13 17:05:42 or reboot, for that matter Sep 13 17:06:27 kill, kill 'em all! Sep 13 17:06:53 it's aways 'highest' system that's supposed to do gentle shutdown/reboot. NOT low level systems like init/upstart telling KDE/dsme/hildon-desktop to fsckng quit now Sep 13 17:08:22 DocScrutinizer05: at the very least i expect KDE not to refuse to launch correctly after it's been SIGTERMed/SIGKILLed by init Sep 13 17:08:54 haha, good luck Sep 13 17:09:29 that's exactly what will NOT happen, if you for example configured KDE to restore last session Sep 13 17:09:37 :s Sep 13 17:09:44 what will happen? Sep 13 17:09:50 undefined Sep 13 17:10:49 init sends all processes sigint, then after 10s kills all porcesses hard. Whataver state they reached meanwhile Sep 13 17:11:09 err sigterm? Sep 13 17:11:38 depends on your session what KDE got done during that time Sep 13 17:11:58 it's not meant to be the normal usecase anyway Sep 13 17:12:26 and you won't see much appreciation for fixing init to wait for KDE to finish Sep 13 17:26:13 NAME Sep 13 17:26:14 halt, reboot, poweroff - stop the system. Sep 13 17:26:16 DESCRIPTION Sep 13 17:26:17 Halt notes that the system is being brought down in the file /var/log/wtmp, and then either tells the kernel to halt, reboot or poweroff the system. Sep 13 17:27:02 nuttin else, nuttin beyond, no gentle bells & whistles Sep 13 17:27:53 the way this got implemented changed slightly with upstart and then again with systemd, the semantics stay the same Sep 13 17:35:35 also see man 8 shutdown: >>... All processes are first notified that the system is going down by the signal SIGTERM. This gives programs like vi(1) the time to save the file being edited, mail and news processing programs a chance to exit cleanly, etc. ...<< NB there's NO feedback about any of those programs actually finished their thing to die. There's no (official) way to stop shutdown from killing all processes hard after 5 Sep 13 17:35:36 or 10 seconds, or actually just shut down telinit Sep 13 17:37:06 that's why KDE calls shutdown, not shutdown quits KDE, if you're trying to use recommended way to shutdown system. Same for fremantle and dsme/hildon/whatever Sep 13 17:38:32 dsme invokes lower level where initctl and reboot live, NOT reboot tells dsme and waits for it to finish Sep 13 19:34:03 and hildon-home/whatever GUI in turn invokes dsme -b (or similar high level function) when you select "power down" or "reboot" on desktop in menu Sep 13 19:35:31 Hello! I have exactly the same problem: http://talk.maemo.org/showthread.php?t=81760 Sep 13 19:35:34 but there's no fix... Sep 13 19:36:00 is there a way to see if a wlan chip is broken or something else is causing it? Sep 13 19:36:12 As it is stated in the thread Sep 13 19:36:20 it's a brand new phone Sep 13 19:36:57 I bought it from ebay so sending it back may cost a lot, so I'm looking for a way to fix it Sep 13 19:38:07 akls: ouch :c Sep 13 19:38:10 does dmesg say something? Sep 13 19:39:15 it says a lot Sep 13 19:39:19 what should I look for? Sep 13 19:39:39 the very end Sep 13 19:39:44 anything related to wlan Sep 13 19:39:52 hmm Sep 13 19:40:16 the worst part is that I got two phones Sep 13 19:40:21 another one is working :( Sep 13 19:40:58 make sure you did a _successful_ _full_ reflash, and maybe don't use ifconfig to play with wlan, rather use settings->internet-connections Sep 13 19:41:16 h, wrong discussion :P Sep 13 19:41:18 hmm, no, nothing in dmesg Sep 13 19:41:19 oops Sep 13 19:41:43 ignore everything I said in the last few lines ^_^ Sep 13 19:42:23 DocScrutinizer05, there's nothing in these settings Sep 13 19:42:27 nothing useful at leas Sep 13 19:42:34 t Sep 13 19:42:58 bluetooth is not working as well Sep 13 19:43:37 about reflashing... I will reflash as soon as I get another battery or charger, because my charger stopped working somehow Sep 13 19:43:42 and it doesn't want to charge from usb Sep 13 19:44:38 so, is there any way to confirm that my wireless chip is broken? Sep 13 20:00:20 akls: was it a used n900? Sep 13 20:00:36 it's a lot more likely that your usb port is busted Sep 13 20:04:38 or system general problem (ram, NAND, fs corruption, swap missing...) Sep 13 20:06:30 akls: you should do a full reflash (vanilla and combined), post the protocol of this to pastebin and link here for checking Sep 13 20:06:42 kerio, no, it was a new telephone Sep 13 20:06:50 DocScrutinizer05, did a full reflash, nothing Sep 13 20:07:00 protocol.. what's a protocol? Sep 13 20:07:01 then boot up device and pastebin output of dmesg Sep 13 20:07:10 and linl here Sep 13 20:07:37 link* Sep 13 20:07:54 thats a hard task to do without the internet Sep 13 20:08:07 the complete display on your terminal screen Sep 13 20:08:12 he might have a point Sep 13 20:08:22 DocScrutinizer51, I see that BME setting charge current sense voltage to 550mV, but shadowjk charge21.sh script 950mV Sep 13 20:08:27 it is safe? Sep 13 20:08:56 and BME set termination current sense voltage to 100mV and charge21.sh 150mV Sep 13 20:09:10 sorry, can't check right now Sep 13 20:09:30 DocScrutinizer51, ok write later Sep 13 20:09:33 akls, copy to microsd card? Sep 13 20:09:45 I don't know about mV for current Sep 13 20:10:16 DocScrutinizer51, only these 2 differences are in bme and charge21.sh script Sep 13 20:10:37 there's no such thing Sep 13 20:10:37 now bq2415x_charger.ko using same values as charge21.sh Sep 13 20:11:58 you probalygot confused about the unit uVblabla for termination and charge current Sep 13 20:12:08 here is small shell script which write bq2415x chip info: http://atrey.karlin.mff.cuni.cz/~pali/charger_print.sh Sep 13 20:12:20 compatible with and without bq2415x kernel driver Sep 13 20:12:20 see footnote 1 and 2 in datasheet Sep 13 20:12:52 err sorry thats bq27200 Sep 13 20:13:28 bbl Sep 13 20:24:51 how to cd to ".documents" folder? Sep 13 20:27:16 or any other folder that is visible from "Mass Storage" mode Sep 13 20:27:25 without sd card Sep 13 20:28:49 that should be in ~/MyDocs Sep 13 20:31:52 right Sep 13 20:34:13 DocScrutinizer05, kerio https://dl.dropbox.com/u/71236259/dmesg.txt Sep 13 20:34:18 anything else? Sep 13 20:35:43 hm, the CMT messages don't sound healthy Sep 13 20:36:05 kerio, and what does it mean? Sep 13 20:36:16 akls: does the phone work? Sep 13 20:36:17 "* CMT rst line change detected (0)" Sep 13 20:36:31 kerio, yes, as I stated Sep 13 20:36:38 everything works except for the wireless Sep 13 20:36:43 wifi Sep 13 20:36:51 and probably bluetooth, but I'm not sure Sep 13 20:36:57 "misc iphb: Module registered in 2.6.28-omap1, built Oct 28 2009 15:33:55" Sep 13 20:37:02 seems like PR1.0 Sep 13 20:37:16 Pali, yep, that's the oldest one. Sep 13 20:37:32 I got this phone today, wifi was not working Sep 13 20:37:39 then I updated to the latest 1.3 Sep 13 20:37:42 still nothing Sep 13 20:37:51 now I reflashed it to the oldest one Sep 13 20:37:56 and it doesn't work.. Sep 13 20:38:00 anything else I can do? Sep 13 20:38:26 akls: flash the latest FW Sep 13 20:38:42 for your region Sep 13 20:38:56 for Global, right? :D Sep 13 20:39:08 akls: NFC, you should know :P Sep 13 20:40:05 well, ok, I'll flash it now Sep 13 20:40:15 but still, how can I verify that wifi is totally broken? Sep 13 20:40:29 ifconfig wlan0 shows a mac address but nothing else Sep 13 20:40:35 [ 80.588653] wl1251: loaded Sep 13 20:40:42 [ 80.589416] wl1251: initialized Sep 13 20:40:52 no more about wl1251 Sep 13 20:40:55 yep Sep 13 20:41:14 akls: what happens if you try to connect to wifi network? Sep 13 20:41:21 freemangordon, oh Sep 13 20:41:26 (i am late to the party) Sep 13 20:41:33 Hello! I have exactly the same problem: http://talk.maemo.org/showthread.php?t=81760 Sep 13 20:41:34 but there's no fix... Sep 13 20:41:42 $ iwlist scan Sep 13 20:41:47 that guy has 100% same problem Sep 13 20:41:56 I don't think that I have wireless-tools installed... Sep 13 20:42:25 akls: yeah, I see Sep 13 20:42:35 but as stated in the forum thread using "ifconfig wlan0 up" locks and leads to other problems like file manager not working Sep 13 20:43:24 akls: anyway, Pali is more experienced than me re wifi Sep 13 20:44:14 that problem from TMO post is in hw Sep 13 20:44:18 WTF Sep 13 20:44:27 Pali: yeah, seems so Sep 13 20:44:31 why does enigma believe it's ok to take 3mb in the rootfs for a fucking game? Sep 13 20:44:35 ifconfig wlan0 up should not freeze system Sep 13 20:44:56 Pali, so it's broken and I can throw it away? Sep 13 20:45:01 damn Sep 13 20:45:11 akls: connect external wifi :P Sep 13 20:45:17 usb dongle that is Sep 13 20:45:22 you can try to reflash pr1.3 + vanilla emmc image Sep 13 20:45:26 awesome... Sep 13 20:45:29 yeah Sep 13 20:45:43 Pali, that's what I had Sep 13 20:45:47 some time ago Sep 13 20:45:50 Pali: though I wonder why the phone still works Sep 13 20:46:08 freemangordon, yep, it's working just fine for the exception of file manager Sep 13 20:46:12 and root console Sep 13 20:46:25 oh, and default console is not working as well.. Sep 13 20:46:34 akls: something is wrong there, you should not be able to use phone if you have downgraded from pr13 Sep 13 20:46:49 why not? Sep 13 20:46:51 freemangordon, why? Sep 13 20:46:56 something with CMT Sep 13 20:47:05 I don't know which firmware I had by default Sep 13 20:47:07 complete flash also flash cmt firmware Sep 13 20:47:08 but I upgraded it to 1.3 Sep 13 20:47:26 Pali: there is some remnants IIRC Sep 13 20:47:29 and them completely reflashed to the oldest one hoping that something is going to change Sep 13 20:47:35 then* Sep 13 20:47:40 ok Sep 13 20:47:56 I'm going offline Sep 13 20:48:05 Pali: CAL, CMT FW or something which cannot be downgraded Sep 13 20:48:13 alright, so I'll contact a guy who sold this one to me. Sep 13 20:48:14 ok Sep 13 20:48:22 thank you very much for your help Sep 13 20:48:27 akls: try to flash pr13 first Sep 13 20:48:30 freemangordon, all bme packages are pushed to gitorious project Sep 13 20:48:41 only libbmeipc is not fully yet Sep 13 20:48:52 I will done it later Sep 13 20:48:53 Pali: gret, unfortunately I am in process of doing a new -thumb update Sep 13 20:48:59 freemangordon, alright. But I'll go offline. If it will work - then I'll come back and tell you how happy I am Sep 13 20:49:05 bye Sep 13 20:49:07 but I'm pretty sure that it's not going to happen ;) Sep 13 20:49:10 won't have time in couple of days :( Sep 13 20:49:26 bye Sep 13 20:49:29 ;) Sep 13 20:51:11 how do i kill browserd? Sep 13 20:51:21 with a gun? Sep 13 20:51:26 is it dsme that keeps resurrecting it? Sep 13 20:51:39 vi__: dsme -k or something Sep 13 20:51:47 or dsmetool -k Sep 13 20:51:55 * vi__ shoots n900, solves all his problems with 1 bullet. Sep 13 20:52:23 vi__: a better idea: break it using your n950 as a hammer :P Sep 13 20:54:02 downgrading on cmt doesn't usually work! Sep 13 20:54:42 so ffs flash PR1.3 combined and latest(!) vanilla Sep 13 20:55:00 DocScrutinizer51: yes, that is why I wonder what did he upgrade Sep 13 20:55:12 sorry dude, support for PR1.0 discontinued Sep 13 20:55:31 DocScrutinizer51: he is not here :D Sep 13 20:55:35 anyway, I am off Sep 13 20:55:37 night Sep 13 20:55:46 night fmg Sep 13 20:56:48 freemangordon: ! Sep 13 20:59:05 DIE BROWSERD ASSWIPE Sep 13 20:59:50 nooo :( Sep 13 21:01:51 what does libstartup.so do?? Sep 13 21:02:29 from memory, its a statistics thing Sep 13 21:04:10 what for? Sep 13 21:04:50 to measure process startup time to first ui idle call Sep 13 21:05:48 wtf is the point? Sep 13 21:07:03 optimising CPU use Sep 13 21:10:33 easy debian chromium-browser... Sep 13 21:10:37 ...not bad Sep 13 21:11:39 is there a proper chromium for maemo? Sep 13 21:13:39 there was but it got abandoned due to legal issues. Sep 13 21:14:14 legal issues that everyone but us dodged? Sep 13 21:16:01 dunno, ask jacekowski Sep 13 21:32:43 SpeedEvil: duh :-o Sep 13 21:33:18 * DocScrutinizer05 learnt something new \o/ Sep 13 21:33:48 this is very hazy memory Sep 13 21:33:56 just when I was about to start thinking "this might get boring" Sep 13 21:35:36 http://maemo.org/development/tools/doc/chinook/sp-startup-time/ Sep 13 21:37:12 LD_PRELOAD=/usr/lib/sp-startup-time/libstartup.so app-to-run Sep 13 21:37:34 vi_: ^^^ Sep 13 21:38:03 OMFG chinook Sep 13 21:38:45 this been one or two releases before diablo? Sep 13 21:39:03 C = D-- Sep 13 21:39:09 so one before diablo Sep 13 21:39:21 what's e? Sep 13 21:39:26 undef Sep 13 21:39:35 :( Sep 13 21:39:40 for some reasons some chars are banned Sep 13 21:39:41 but that's the same letter as my name! Sep 13 21:41:59 * DocScrutinizer05 wonders if RC0 been named "atrocity" or "atrophy" Sep 13 21:42:34 you mean maemo 1 or maemo 2 Sep 13 21:42:36 ? Sep 13 21:43:18 1 doesn't have a codename, 2 is mistral/scirocco/gregale, 3 is bora, 4 is chinook/diablo, 5 is fremantle Sep 13 21:43:27 according to wikipedia Sep 13 21:44:31 LOL Sep 13 21:44:33 (1.0.0) Gregale (2.0.0) Bora (3.0.0) Chinook (4.0.0) Diablo (5.0.0) Fremantle Sep 13 21:44:42 what's that? Sep 13 21:44:50 http://wiki.maemo.org/Maemo-releases Sep 13 21:45:16 how do i vote for deletion? Sep 13 21:46:05 http://wiki.maemo.org/Codenames Sep 13 21:46:48 hehe, who added CSSU there? Sep 13 21:47:04 Starting from Bora the release names will be alphabetical, E and G have been excluded from this progression for various reasons (see Presentation by Quim Gil at LinuxTag 2007). Sep 13 21:47:26 o.O Sep 13 21:47:29 CSSU? Sep 13 21:47:34 in Codenames Sep 13 21:47:41 LOL Sep 13 21:47:52 WE ARE FAMOUS!!! \o/ Sep 13 21:48:17 dafuq? Sep 13 21:49:16 Elephanta is an awesome name Sep 13 21:49:18 ;-P Sep 13 21:49:50 it was thp Sep 13 21:49:58 http://wiki.maemo.org/index.php?title=Template:Release_history_table&oldid=43575 Sep 13 22:28:34 wtf libmeegotouch on N900? http://maemo.org/packages/view/applauncherd-libmeegotouch-perf-tests/ Sep 13 22:29:51 DocScrutinizer05: http://maemo.org/packages/search/?org_maemo_packages_search[1][property]=name&org_maemo_packages_search[1][constraint]=LIKE&org_maemo_packages_search[1][value]=libmeegotouch&org_maemo_packages_search[2][property]=title&org_maemo_packages_search[2][constraint]=LIKE&org_maemo_packages_search[2][value]=same Sep 13 22:31:12 merlin1991: :nod: - but... any application using it yet? Sep 13 22:31:19 and how? Sep 13 22:31:45 don't know, the guy who uploaded libmeegotouch is working on a harm -> fremantle compatability layer Sep 13 22:31:45 I mean... is it a toolkit like qt? Sep 13 22:33:09 MTF date/time-picker for sure more sane than fremantle qt standard Sep 13 22:33:29 aka F*CKING slotmachine Sep 13 22:33:36 libmeegotouch is a widget library on top of qt widgets Sep 13 22:33:58 which can be styled with css files Sep 13 22:34:14 ohnoes, so even the MTF timepicker will be a slotmachine on fremantle?? :-o Sep 13 22:34:26 wtf? Sep 13 22:34:31 how should that work? Sep 13 22:34:47 dunno, just shocked Sep 13 22:35:02 fremantle timepicher is hildon/gtk, it would have to be a very very weird coincidence that it looks similar in any way Sep 13 22:35:04 when I heard "on top of qt" Sep 13 22:35:38 well *nokia* fremantle apps are 90% hildon/gtk Sep 13 22:35:51 actually I don't know a single core application that would be in qt Sep 13 22:36:06 HARM qt(m) apps have slotmachines worse than fremantle Sep 13 22:37:05 harm has a really nice datepicker Sep 13 22:37:15 in MTF, yes Sep 13 22:37:30 well you're talking about the harm qml apps Sep 13 22:37:43 those have bad timepickers, that's true Sep 13 22:38:01 though someone released a proper qt component to be used from qml that works like the mtf one Sep 13 22:38:14 qtm/qml wtf idunno Sep 13 22:38:29 http://va-sorokin.blogspot.co.at/2012/02/presentation-of-meego-harmattan-native.html would be the thing Sep 13 22:38:50 yeah, seen it and already forgot that awesome work Sep 13 22:39:11 since I'll probably never touch HARM again Sep 13 22:39:26 though it's kinda sad how they changed the stock timepicker in pr1.2 Sep 13 22:40:00 before that you practically had 2 wheels which you draged around, now it jumps to the point you click at Sep 13 22:40:42 * merlin1991 should find the offending commit and revert it in the lib Sep 13 23:46:06 http://kuwaitiful.com/tech-gadgets/iphone-5-vs-fake-iphone-5/ Sep 14 00:06:04 Pali: bme might start at 550mA, but it goes up eventually (unless dc adapter is crap, unless battery bsi reading is crap) Sep 14 00:07:48 Pali: bme set 100mA termination in original firmware, after original firmware, around PR1.3, a kernel change was made which increased minimum power consumption while on charger by 60mA, thus the bump by 50mA in charge script. This kernel change is reverted in power kernel, so there I actually use 50 myself. Sep 14 00:31:19 yeah, ang guess what's eating those 60mA! musb-hdrc staing active while on charger, since.. duh, it might turn out it's a USB host that only took like 3h to start ENUM (don't ask me why such host has D+/- short though) Sep 14 00:31:34 s/ang/and/ Sep 14 00:31:35 DocScrutinizer05 meant: yeah, and guess what's eating those 60mA! musb-hdrc staing active while on charger, since.. duh, it might turn out it's a USB host that only took like 3h to start ENUM (don't ask me why such host has D+/- short though) Sep 14 00:34:53 iirc it was for some weird charging usb hub Sep 14 00:35:05 that doesnt exist in real life Sep 14 02:25:57 any n900 coming out??? Sep 14 02:26:04 ...? Sep 14 02:26:05 some kinda version Sep 14 02:26:35 Uh, why would you think that? Sep 14 02:27:45 N900 was/is awesome. And would own windows phone if it had a bit of polish... which means Nokia can't release it(their loyalty being bought by Microsoft) Sep 14 02:49:20 whaaaaaat Sep 14 02:49:55 crazies if they only knew how much tech ppl would love to tweak there phones Sep 14 02:52:55 well, there is still a lot of tweaking to be done with n900, so get tweaking :) **** ENDING LOGGING AT Fri Sep 14 02:59:59 2012