| 00:24.47 | *** join/#harmattan rikanee (~rika@unaffiliated/rikanee) |
| 00:41.44 | *** join/#harmattan louisdk (~louis@static-5-103-130-65.seas-nve.net) |
| 00:44.15 | louisdk | I've attached my Nokia N9 to a wall charger for a couple hours. It will still not boot. |
| 00:46.16 | rikanee | louisdk: tried the 'flasher -i' trick? |
| 00:49.00 | louisdk | rikanee: Yes and 30+ mini charges. Source: http://everythingn9.com/death-by-battery-drain-charge-nokia-n9/ No recognition. |
| 00:50.22 | rikanee | Sounds like it isn't holding a charge anymore. |
| 00:51.43 | louisdk | I've disassembled the device. The battery has been laying for some mouths but I got the LED flash light when inserting it. I wonder if it could be some slots to the screen not connected correctly and this would keep the device from booting. |
| 00:52.31 | rikanee | The N9 (same with all other Nokia devices) will boot with just the board connected to the battery. |
| 00:53.59 | rikanee | Battery damage is a real possibility, and N9 batteries aren't particularly expensive. Might as well try a fresh one. |
| 00:54.12 | louisdk | rikanee: Anyway to charge the battery externally? |
| 00:55.28 | rikanee | louisdk: You can try to hook up a phone battery charger to the three contacts on the BV-5JW |
| 01:00.17 | louisdk | rikanee: Well. I have this charger: http://images.yetaa.com/mobilemate/nokia/nokia-original-dt-33-5230-e7-00-c7-n8-x6-c6-battery-charger_280.jpg But don't know how/if to charge the BV-5JW with that. |
| 01:03.46 | rikanee | Get some wire, and hook up the terminals. |
| 01:04.01 | rikanee | It may be a bit difficult to get them to stick to the tiny pins on that, though. |
| 01:04.22 | rikanee | Try to find one with the charging pins fat and exposed, so you can use electrical tape. |
| 01:08.00 | louisdk | Sounds a bit complicated. Maybe I should try to replace my N9 battery with a spare BV-5JW that I have laying round. |
| 01:18.24 | *** join/#harmattan jon_y (~enforcer@2002:db5c:6bdb::db5c:6bdb) |
| 01:58.46 | *** join/#harmattan mschlens (~mschlens@p5DE79B61.dip0.t-ipconnect.de) |
| 02:08.28 | *** join/#harmattan Morpog_Mobile (~Morpog@stgt-5d8427fa.pool.mediaWays.net) |
| 02:09.46 | dashin | http://ark.it.cx/qbalda/ anyone interested in these games? |
| 02:19.43 | *** join/#harmattan louisdk (~louis@static-5-103-130-65.seas-nve.net) |
| 02:43.21 | *** join/#harmattan bef0rd (~bef0rd@unaffiliated/beford) |
| 02:44.01 | dashin | (N9 deb pkg + screen) |
| 02:44.26 | bef0rd | hi |
| 02:46.06 | dashin | hi |
| 02:50.54 | *** join/#harmattan joecool (~joecool@no-sources/joecool) |
| 02:53.53 | louisdk | hi |
| 03:03.49 | *** join/#harmattan DocScrutinizer05 (~HaleBopp@openmoko/engineers/joerg) |
| 03:34.16 | *** join/#harmattan ugoub (~ugoub@110.188.53.228) |
| 03:38.47 | *** part/#harmattan ugoub (~ugoub@110.188.53.228) |
| 05:39.14 | *** join/#harmattan dashin (~dd@94.253.104.215) |
| 05:54.59 | *** join/#harmattan e-yes (~e-yes@95.73.165.86) |
| 06:21.58 | *** join/#harmattan bef0rd (~bef0rd@unaffiliated/beford) |
| 06:26.03 | *** join/#harmattan Guest79237 (Termana@outback.nsw.au.panicbnc.net) |
| 06:32.42 | *** join/#harmattan trigpoint_n9 (~trigpoint@91.183.109.166) |
| 06:39.43 | *** join/#harmattan e-yes (~e-yes@95.73.165.86) |
| 06:48.50 | *** join/#harmattan rikanee (~rika@unaffiliated/rikanee) |
| 06:51.42 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 07:39.41 | *** join/#harmattan e-yes (~e-yes@95.73.165.86) |
| 08:06.45 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 08:17.24 | *** join/#harmattan M4rtinK (~M4rtinK@mail.melf.eu) |
| 08:22.17 | *** join/#harmattan ced117 (~ced117@opensuse/member/ced117) |
| 08:27.56 | *** join/#harmattan xes (~xes@unaffiliated/xes) |
| 08:39.05 | *** join/#harmattan NIN101 (~NIN@p5DD29509.dip0.t-ipconnect.de) |
| 08:39.48 | *** join/#harmattan e-yes (~e-yes@95.73.165.86) |
| 08:57.58 | *** join/#harmattan rcg (~rcg@2a02:908:e250:3801:f588:6680:a189:775) |
| 09:06.17 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 09:15.52 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 09:50.53 | *** join/#harmattan ced117 (~ced117@opensuse/member/ced117) |
| 10:04.32 | *** join/#harmattan tonyoy (~tonyoy@cor13-3-88-164-147-63.fbx.proxad.net) |
| 10:05.15 | *** join/#harmattan sqrt7744 (~SillyPant@cable-86-56-69-193.cust.telecolumbus.net) |
| 10:10.18 | *** join/#harmattan suy (~alex@94.190.14.37.dynamic.jazztel.es) |
| 10:15.23 | *** join/#harmattan NIN101 (~NIN@p5DD29509.dip0.t-ipconnect.de) |
| 10:21.58 | *** join/#harmattan ced117 (~ced117@opensuse/member/ced117) |
| 10:25.47 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 10:31.03 | *** join/#harmattan Morpog_PC (~Morpog@stgt-5d8427fa.pool.mediaWays.net) |
| 10:49.21 | *** join/#harmattan Pali (~pali@Maemo/community/contributor/Pali) |
| 10:55.09 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 11:11.38 | *** join/#harmattan NIN101 (~NIN@p5DD29682.dip0.t-ipconnect.de) |
| 11:19.12 | *** join/#harmattan stef_204 (~kvirc@unaffiliated/stef-204/x-384198) |
| 11:31.33 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 11:31.59 | *** join/#harmattan Necrosporus (~Necrospor@unaffiliated/necrosporus) |
| 11:37.03 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 11:44.02 | *** join/#harmattan Martix_ (~martix@static-84-242-103-180.net.upcbroadband.cz) |
| 11:44.58 | *** join/#harmattan trigpoint_n9 (~trigpoint@31.71.219.189) |
| 12:00.23 | *** join/#harmattan sqrt7744 (~SillyPant@cable-86-56-69-193.cust.telecolumbus.net) |
| 12:00.58 | *** join/#harmattan Sfiet_Konstantin (~sk@85.218.89.237) |
| 12:26.49 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 12:45.29 | *** join/#harmattan friese (~anonfries@p5481A1DE.dip0.t-ipconnect.de) |
| 12:50.18 | *** join/#harmattan NIN101 (~NIN@p5DD2885F.dip0.t-ipconnect.de) |
| 12:50.44 | *** join/#harmattan rashm2k1 (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 12:51.01 | *** join/#harmattan xsacha_ (~quassel@124.148.30.36) |
| 13:22.02 | Morpog_PC | is it save to replace files in this directory, or will aegis kill my device? /usr/share/themes/base/meegotouch/meegotouchhome/images/ |
| 13:35.31 | Morpog_PC | Sfiet_Konstantin, can you answer my above question? |
| 13:36.22 | *** join/#harmattan hardaker (~hardaker@mail.hardakers.net) |
| 13:36.25 | Sfiet_Konstantin | Morpog_PC: not sure if it is a good idea at all |
| 13:36.29 | Sfiet_Konstantin | why ? |
| 13:36.56 | Morpog_PC | trying to customize launcher a bit |
| 13:37.24 | Sfiet_Konstantin | I think it is aegis-protected |
| 13:37.27 | Morpog_PC | The file I want to edit seems to be added from homescreensettings app |
| 13:37.29 | Sfiet_Konstantin | so not a good idea |
| 13:37.36 | Morpog_PC | damn :( |
| 13:38.02 | Morpog_PC | is there an overview which areas are aegis-protected anywhere? |
| 13:38.13 | Morpog_PC | or how can we check if something is aegis protected? |
| 13:38.27 | Sfiet_Konstantin | hum |
| 13:38.37 | Sfiet_Konstantin | actually I don't know that much |
| 13:38.42 | Sfiet_Konstantin | because it is not lib |
| 13:38.47 | Sfiet_Konstantin | so you might be able to edit it |
| 13:38.58 | Sfiet_Konstantin | by rethinking it |
| 13:39.06 | Sfiet_Konstantin | I don't think that it will be dangerous |
| 13:40.02 | Morpog_PC | seems refhaslist has all files in it |
| 13:41.13 | Morpog_PC | but where the hell is it located.... |
| 13:41.17 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 13:42.49 | Morpog_PC | /var/lib/aegis/refhashlist |
| 13:44.57 | Morpog_PC | haven't found /usr/share in there at all |
| 14:11.36 | DocScrutinizer05 | Morpog_PC: |
| 14:11.41 | DocScrutinizer05 | ~may-i-edit |
| 14:11.41 | infobot | mayIedit () { grep "`basename $1`" /var/lib/aegis/refhashlist && echo "probably not" && return; echo "edit if you feel venturous"; } |
| 14:13.06 | Morpog_PC | ;-) |
| 14:14.28 | Pali | who and how kill you if you edit refhashlist? |
| 14:14.32 | *** join/#harmattan NIN102 (~NIN@p5DD2AC62.dip0.t-ipconnect.de) |
| 14:30.11 | DocScrutinizer05 | ~aegis |
| 14:30.11 | infobot | http://www.developer.nokia.com/Community/Wiki/Harmattan:Developer_Library/Developing_for_Harmattan/Harmattan_security/Security_guide , or "The purpose of this framework is: ... to make sure that the platform meets the requirements set by third party software that requires a safe execution environment.", or http://en.wikipedia.org/wiki/Trusted_Computing#Criticism, or http://en.qi-hardware.com/w/images/1/10/ME_382_LockedUpTechnology2.gif |
| 14:32.36 | DocScrutinizer05 | and how? |
| 14:32.39 | DocScrutinizer05 | ~malf |
| 14:32.39 | infobot | methinks malf is http://maemo.cloud-7.de/Aegis-kills-device.jpg |
| 15:13.58 | Pali | no, I mean how aegis now that that config file was changed? |
| 15:15.04 | Pali | if somebody edit /var/lib/aegis/refhashlist manually how aegis will now that? |
| 15:15.15 | Morpog_PC | no no |
| 15:15.41 | Morpog_PC | i was talking about a file to edit which is referenced in refhaslist |
| 15:16.11 | Pali | above is my question :-) |
| 15:16.16 | Morpog_PC | turned out that the file wasn't in refhashlist, so it's safe to edit it |
| 15:16.22 | Morpog_PC | aegis can't |
| 15:22.34 | *** join/#harmattan Milhouse (~Milhouse@Maemo/community/contributor/Milhouse) |
| 15:31.46 | Morpog_PC | result of my modding: http://abload.de/img/2013-05-04_16-56-00haiaa.png |
| 15:31.51 | *** join/#harmattan Milhouse (~Milhouse@Maemo/community/contributor/Milhouse) |
| 15:35.49 | robertjw | Doc: that .../Security_guide link is old. It redirects to the generic Platform Guide. |
| 15:43.47 | DocScrutinizer05 | robertjw: well, you noticed that *every* link on that page gets redirected to <self> ? |
| 15:44.42 | robertjw | my parnoid part wants to assign a meaning to this |
| 15:45.15 | DocScrutinizer05 | this is Nokia's outstanding awesome Meego/Harmattan support |
| 15:46.29 | robertjw | I like Richard Stallman's rename of "Trusted computing" to "Treacherous computing" |
| 15:47.28 | robertjw | http://www.gnu.org/philosophy/can-you-trust.html |
| 15:47.50 | DocScrutinizer05 | well |
| 15:48.07 | DocScrutinizer05 | RMS is not exactly the reference for me |
| 15:49.03 | robertjw | he's been controversial, but he often makes some good points. I'll cut him some slack. :) |
| 15:51.58 | robertjw | this whole issue of "locked hardware" really pushes my buttons. The vendors call it "trusted computing", to imply something good, but in fact it is the vendor not trusting the customer. |
| 15:53.00 | SpeedEvil | it depends. |
| 15:53.21 | robertjw | expand your thought please |
| 15:53.24 | DocScrutinizer05 | http://www.youtube.com/watch?v=0cbS_lDJuJg |
| 15:53.32 | SpeedEvil | there are valid arguments for trusted computing at least somewhat opaque from the user |
| 15:53.53 | SpeedEvil | for example, running credit card secret algorithms |
| 15:54.09 | robertjw | I'm listening, all the arguments I've heard so far favor the vendor at the expense of the user. |
| 15:55.08 | robertjw | the algorithms don't have to be secret in order to be secure. Reference: ssh |
| 15:55.23 | SpeedEvil | also, owner and person with the device at the moment are not the same |
| 15:55.31 | SpeedEvil | ssh does a different thing |
| 15:55.47 | SpeedEvil | being able to clone your credit card is bad |
| 15:56.22 | robertjw | in my view, openness improves security because of all the eyes watching it |
| 15:56.52 | SpeedEvil | umm... that doesn't work, if there is a secret key that needs to remain secret |
| 15:57.11 | robertjw | you were discussing algorithms |
| 15:57.59 | SpeedEvil | the code that normally runs on the credit card to do with |
| 15:59.17 | SpeedEvil | auth |
| 16:00.58 | *** join/#harmattan jluisn (~quassel@187.115.172.24) |
| 16:03.31 | robertjw | Doc: enjoyed your youtube link. That's a bookmarker... |
| 16:06.32 | *** join/#harmattan arcean (~arcean@aadb188.neoplus.adsl.tpnet.pl) |
| 16:16.25 | DocScrutinizer05 | if there's a secret in something that *I* must not know, then what a basket is handing that object with the embedded secret to me anyway? |
| 16:18.59 | DocScrutinizer05 | anything with embedded secrets has a wrong and broken business model |
| 16:19.20 | DocScrutinizer05 | or usecase concept |
| 16:20.00 | DocScrutinizer05 | in the end it's always about depriving me from my natural control over what I can do with any such object |
| 16:20.48 | robertjw | I agree, and I would even put the credit card example into this. In the case of the credit card, the secret is the money, and the money belongs to someone. The one who owns the money should hold its secret so that NOone can take his money. |
| 16:20.53 | DocScrutinizer05 | be it copying of mp3/movie data, or dongling of sw to hw, or whatever |
| 16:21.14 | DocScrutinizer05 | yep |
| 16:21.31 | DocScrutinizer05 | what's wrong with cloning my credit card? |
| 16:21.44 | DocScrutinizer05 | it's dirt simple |
| 16:22.20 | DocScrutinizer05 | what you also can clone but better don't share is the secret PIN to approve any transaction with the credit vard |
| 16:22.34 | DocScrutinizer05 | you keep that in your wetware |
| 16:22.35 | *** join/#harmattan bef0rd (~bef0rd@unaffiliated/beford) |
| 16:23.14 | DocScrutinizer05 | and it's absolutely irrelevant of there's one or 200 card that have same data on their magnet stripe |
| 16:23.23 | DocScrutinizer05 | if* |
| 16:23.44 | robertjw | so long as the act of cloning does not "create" additional money |
| 16:24.09 | DocScrutinizer05 | then your architecture is terribly broken |
| 16:24.16 | robertjw | right |
| 16:26.02 | robertjw | the issue of locks in general comes down to one entity locking something from another so you have to ask who is doing the locking and who is being locked out |
| 16:26.47 | DocScrutinizer05 | right |
| 16:27.21 | DocScrutinizer05 | aegis locks out users, telling us "it's a benefit for you, since no malware can do what you can't do" |
| 16:27.32 | DocScrutinizer05 | that's BS |
| 16:27.57 | robertjw | the issue of locks has been cloaked in clever terminology, vendors use locks to claim "security" and users feel good, while in fact they are talking about their own security from the user |
| 16:28.18 | DocScrutinizer05 | such type of lock finally *always* is about depriving users of their natiral powers |
| 16:28.40 | DocScrutinizer05 | yes, exactly |
| 16:29.08 | robertjw | with a lock there is always and inside and an outside. In this case the vendor is on the inside and the user is on the outside. Few users see this clearly. |
| 16:29.24 | DocScrutinizer05 | there's definitely ZILCH that can be done by aegis that couldn't be done by a decent traditional permissions handling |
| 16:29.44 | DocScrutinizer05 | except depriving user from his right to poke in his data |
| 16:30.01 | DocScrutinizer05 | aka DRM |
| 16:31.00 | robertjw | I secure my gnu/linux systems with packet captures for analysis/monitoring and firewall for protection, that coupled with use of heavily peer-reviewed open software. |
| 16:31.06 | DocScrutinizer05 | "no, dude! This movie won't get played back via AV output! embedded screen sure, HDMI, maybe. Not AV though" |
| 16:31.45 | DocScrutinizer05 | "Sorry, you have the permission to listen to this song 3 times. We think you used up that quota" |
| 16:32.46 | robertjw | in another sense, proprietary closed software is a form of lock. Only the developer of that software knows what it does. The proprietary developer holds they key, the user is on the outside. |
| 16:33.13 | DocScrutinizer05 | "what, you want to install a software that does same as our ACME shite, but without nagware banners aka commercials? Sorry, we won't allow that, even while it's your hw" |
| 16:35.07 | DocScrutinizer05 | "your government found the last year's report from greenpeace contains confidential material. Sorry we had to delete it from your library" |
| 16:36.20 | robertjw | Now you are raising my blood pressure. :) |
| 16:36.26 | DocScrutinizer05 | "your new browser has a builtin protection function to save you from watching irritating content like wikileaks etc. Please install the update now, as your OS will stop working until you did" |
| 16:37.57 | DocScrutinizer05 | sure, all for our protection |
| 16:39.18 | DocScrutinizer05 | not a single business/company will go for those nonsensical TPM chips, when they really bother about own assets |
| 16:39.48 | robertjw | Having said all this, my N9 is the closest I've been able to get to a smartphone that I can secure. I need three things to secure a system, packet-capture tools for analysis/monitoring, firewall for protection, and a source of open (peer-reviewed) software. With N9 I have (almost) all three. |
| 16:41.19 | robertjw | too many Android apps are closed and locked to make it securable, and I won't even mention iOS. |
| 16:49.30 | *** join/#harmattan lucido (~chris@49.49.233.121) |
| 16:50.09 | lucido | Hi, are there any new harmattan devices, or is there a successor with new hardware for the OS? |
| 17:09.14 | mschlens | lucido: no |
| 17:09.27 | lucido | :( |
| 17:23.43 | *** join/#harmattan ced117 (~ced117@opensuse/member/ced117) |
| 17:38.45 | *** join/#harmattan rashm2k1 (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 18:08.50 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 18:12.49 | *** join/#harmattan trigpoint_n9 (~trigpoint@31.71.200.71) |
| 18:15.18 | *** join/#harmattan ortylp (~ortylp@HSI-KBW-078-042-249-136.hsi3.kabel-badenwuerttemberg.de) |
| 18:22.01 | *** join/#harmattan rcg (~rcg@2a02:908:e250:3801:f588:6680:a189:775) |
| 18:27.38 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 18:50.11 | *** join/#harmattan Milhouse (~Milhouse@Maemo/community/contributor/Milhouse) |
| 18:57.41 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 18:59.08 | *** join/#harmattan stroughtonsmith (~steven@86-42-129-122-dynamic.b-ras1.bbh.dublin.eircom.net) |
| 19:02.15 | *** join/#harmattan rcg (~rcg@2a02:908:e250:3801:f588:6680:a189:775) |
| 19:05.32 | *** join/#harmattan Pali (~pali@Maemo/community/contributor/Pali) |
| 19:16.48 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 19:28.00 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 19:30.02 | *** join/#harmattan fmunozs (~fmunozs@186.154.34.98) |
| 19:31.23 | *** join/#harmattan piggz (~piggz@host-92-18-185-170.as13285.net) |
| 19:32.41 | *** join/#harmattan juiceme (~juice@a88-115-90-185.elisa-laajakaista.fi) |
| 19:33.05 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 19:38.16 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 19:46.22 | *** join/#harmattan m3joooog (~m3jog@213.139.60.68) |
| 19:46.49 | *** join/#harmattan rashm2k (~rm@188-221-216-100.zone12.bethere.co.uk) |
| 19:58.08 | *** join/#harmattan loin (~al@79.113.83.229) |
| 19:58.23 | loin | hello, do i need any special permissions to make my .sh files executable ? |
| 20:05.42 | *** join/#harmattan juiceme (~juice@a88-115-90-185.elisa-laajakaista.fi) |
| 20:07.15 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 20:11.35 | Jonni | loin: normal unix +x bit, but ofcourse if you want to be able to execute them without aegis-exec, then you need to add the sha1sum of the script in refhashlist |
| 20:13.00 | Jonni | (usually you just package the scripts in deb file, and let dpkg installer add the sha1sums for you) vs you resign the refhashlist yourself with opensh or similar |
| 20:13.44 | Jonni | or you just continue calling your scripts with aegis-exec prefix and you dont have to do anything. |
| 20:16.32 | loin | Jonni, i want to execute them without aegis-exec |
| 20:16.41 | loin | but they wont work |
| 20:17.28 | loin | Jonni, i have a script in the project, but it won't allow me to execute it |
| 20:17.28 | Jonni | if you want to execute without aegis-exec, then you need to package them inside a .deb file and allow unknown sources, and install the scripts with dpkg -i yourscriptpackage.deb |
| 20:18.04 | loin | Jonni, i am pretty much doing that but it wont allow me |
| 20:18.06 | loin | sadly |
| 20:18.22 | Jonni | or if you have opensh with all capas, you can take sha1sum of your script, manually add the line in refhashlist and resign the list. |
| 20:18.40 | Jonni | if you try to run scripts without installing them 1st, aegis wont let you run them |
| 20:19.22 | Jonni | loin: so whats the error message when you dpkg -i ? |
| 20:19.32 | loin | Jonni, no error, the package installs and everything |
| 20:19.48 | loin | its just that when i want to execute the script, i get a not nice error message |
| 20:19.53 | loin | "/opt/feelcache/bin/feelcache.sh: line 36: can't create /proc/sys/vm/drop_caches: Permission denied" |
| 20:20.21 | loin | it is the exact same script as the one from another project |
| 20:20.29 | Jonni | well thats most likely because vm dir is root owned and your script is running as user |
| 20:20.30 | loin | and that one can "create" /proc/sys/vm/drop_caches |
| 20:20.47 | loin | <PROTECTED> |
| 20:20.48 | loin | / app.start("/opt/dropcache-mdn/bin/dropcache.sh", QStringList() << QLatin1String("--3")); |
| 20:20.54 | loin | if i run the first line i get that error |
| 20:20.59 | Jonni | (unless you added aegis manigest in your deb package to make script run as root) |
| 20:21.02 | loin | if i run the second line, everything is ok |
| 20:22.08 | Jonni | did you add aegis manifest to allow root in your deb file? |
| 20:22.42 | loin | where should i do that? |
| 20:24.04 | loin | Jonni, http://paste.ubuntu.com/5633434/ http://paste.ubuntu.com/5633435/ |
| 20:24.10 | loin | the manifests for the different files |
| 20:25.43 | loin | the scripts are the exact same file |
| 20:25.53 | loin | same md5sum, same permissions same owner |
| 20:26.13 | loin | but still one executes and another one doesn't ... |
| 20:26.15 | Jonni | so your script is named feelcache and not feelcache.sh |
| 20:26.42 | loin | the script is feelcache.sh |
| 20:27.00 | Jonni | well that manifest is not giving any permissions to feelcache.sh |
| 20:27.10 | Jonni | its only giving permissions to feelcache |
| 20:27.34 | loin | but why does dropcache execute then? |
| 20:28.02 | Jonni | do you have full sources to both of the applications? |
| 20:28.43 | loin | yes |
| 20:32.56 | loin | any ideas? |
| 20:35.21 | Jonni | yes I have the idea :) |
| 20:36.23 | Jonni | loin: it works because of this http://paste.ubuntu.com/5633476/ |
| 20:36.38 | Jonni | so your missing aegis manifest for your script :) |
| 20:36.51 | Jonni | (I just clone dropcache sources and checked) |
| 20:36.57 | Jonni | cloned even |
| 20:38.12 | loin | ??? |
| 20:38.17 | loin | how did that get there? |
| 20:38.42 | Jonni | so looks like you pasted the wrong manifest :) |
| 20:40.37 | loin | Jonni, that is so strange |
| 20:40.48 | loin | when i opened it in qt creator |
| 20:40.59 | Jonni | and at this point, I also have to say that dropping caches doesnt make any sense. As it just makes N9 slower. |
| 20:41.01 | loin | it just created a regular manifest |
| 20:41.29 | loin | Jonni, i thought so too, but it makes my phone much faster |
| 20:41.38 | Jonni | yes qt creator assumes and autocreates crap, you should really devel in scratchbox :) |
| 20:42.55 | Jonni | but anyways your solution is in here http://code.google.com/p/dropcache/source/browse/debian/dropcache-mdn.aegis |
| 20:43.28 | loin | Jonni, thank you, i found the file using the file manager |
| 20:43.42 | loin | i really wasn't expecting qt creator to do this to me :) |
| 20:43.44 | Jonni | that way script is running as root and it has write permission to vm dir |
| 20:44.13 | loin | Jonni, thank you |
| 20:45.29 | loin | Jonni, if i don't regularly drop the cache |
| 20:45.39 | loin | it happens too often that my phone just starts swapping crap |
| 20:45.42 | loin | and it lags like hell |
| 20:45.53 | *** join/#harmattan wella (~wella@197.132.75.56) |
| 20:46.03 | *** join/#harmattan lucido (~chris@49.49.233.121) |
| 20:47.07 | Jonni | well ofcourse if you happen to run that low on memory then your out of luck, ofcourse you could just disable the ramswap, that way phone will never swap anything :) |
| 20:47.55 | loin | i wonder what would the phone do in such a case |
| 20:47.59 | loin | when it ran out of memory |
| 20:48.10 | loin | just kill whatever app is requesting more ram? |
| 20:48.22 | Jonni | then out of memory killer would start killing processes |
| 20:48.48 | joecool | loin: thanks btw, your log of bricking your n9 trying to remove facebook saved me from bricking mine |
| 20:48.54 | joecool | :D |
| 20:49.01 | loin | joecool, :) |
| 20:49.09 | loin | joecool, yw |
| 20:49.25 | loin | joecool, i bricked it once trying to rm facebook |
| 20:49.32 | loin | then bricked it much harder a second time |
| 20:49.48 | joecool | i got facebook removed, just not libqt-facebook, aegis would not let it go |
| 20:49.57 | joecool | better than nothing i guess |
| 20:50.03 | loin | ever since i did not attempt to rm any "unremovable" apps |
| 20:50.08 | loin | i will most likely remove them again |
| 20:50.24 | loin | thing is, if you remove things that aegis doesn't want you to remove |
| 20:50.30 | loin | never attempt to reset to factory settings |
| 20:50.34 | loin | or you will get bootloop |
| 20:50.59 | joecool | good to know |
| 20:51.23 | loin | but i think it's safe to remove facebook and stuch |
| 20:51.25 | loin | such* |
| 20:51.38 | joecool | yeah that was really the only one that bugged me being there |
| 20:52.29 | Jonni | most of the N9 slowness is because of smart search, most efficient way to make N9 faster and to increase battery life is just to do "chmod a-x /usr/bin/smartsearch" |
| 20:52.50 | loin | well, i removed skype / tracker / user info / twitter / facebook / music store / drive / youtube / google / caldav / flickr / picassa / sip |
| 20:53.13 | loin | Jonni, when is smartsearch used? |
| 20:53.31 | joecool | damn goin light there |
| 20:53.47 | Jonni | loin: it runs always in background when you idle. End user only uses it if you use that global search fuctionality (I never use that) |
| 20:53.54 | joecool | yeah i probably could do without music store and flickr, the rest i use a lot of |
| 20:53.55 | loin | joecool, i don't use all that stuff and i didn't want it so why not throw it away? |
| 20:54.07 | joecool | oh and twitter, i don't use that |
| 20:54.29 | loin | Jonni, i don't use global search |
| 20:54.32 | loin | who wants that? |
| 20:54.38 | Jonni | my N9 only eats 7mA when idle, mostly because I disabled smartsearch. |
| 20:55.00 | loin | 7ma? not bad at all |
| 20:55.03 | loin | mine uses 11 |
| 20:55.10 | joecool | how do you measure? |
| 20:55.26 | loin | joecool, "battery usage" |
| 20:55.33 | Jonni | joecool: I have a script that updates image on low powermode screen |
| 20:55.44 | Jonni | and I just read it on the screen |
| 20:55.47 | joecool | ah cool |
| 20:56.13 | loin | Jonni, isn't that a bit too extreme? |
| 20:56.26 | joecool | only had a few days so far with the N9, a lot to pick up, overall pretty happy, i got it primarily for battery life |
| 20:56.37 | joecool | sip eats a TON of battery on android |
| 20:57.17 | loin | joecool, i got the n9 because of the awesome design |
| 20:57.28 | loin | the physical device design and the swipe interface design |
| 20:57.34 | loin | it's just so stylish :) |
| 20:58.38 | loin | i knew it was sluggish like hell |
| 20:58.39 | joecool | well that too |
| 20:58.48 | joecool | tbh, it's not actually that sluggish |
| 20:58.53 | loin | that was what was stopping me from buying one |
| 20:59.00 | loin | joecool, did you ever have a nexus device? |
| 20:59.18 | joecool | loin: no, i've had some extensive time demoing them though |
| 20:59.39 | loin | joecool, well, my previous phone was a nexus |
| 20:59.42 | joecool | my two other current devices are the xperia ion and the htc hd2 |
| 20:59.44 | loin | and it loaded really fast |
| 20:59.46 | loin | on the n9 |
| 20:59.47 | loin | you click |
| 20:59.50 | Jonni | but anyways its just a hint that if you dont use global search icon, then chmod a-x will make a big improvement in your life :) |
| 20:59.53 | loin | wait like 5 seconds then it loads |
| 21:00.18 | loin | Jonni, thank you for that hint, i will probably not chmod -x it but replace the content with exit 1 |
| 21:00.19 | loin | or something |
| 21:00.23 | joecool | loin: oh you mean load time, ok.. UI lags and FC's are a part of android life |
| 21:00.28 | joecool | that and bad battery life |
| 21:00.57 | Jonni | loin: chmod a-x will not malf your device, replaceing with exit 1 will screw refhashlist and make your device not to boot up :) |
| 21:01.09 | joecool | i have not had the N9 go unresponsive yet |
| 21:01.24 | Jonni | and you can always a+x if you want to use smartsearch later. |
| 21:01.30 | loin | Jonni, thanks for that, that's one less reflash i'll have to do on my device |
| 21:02.02 | loin | joecool, maybe your n9 is better than mine :) |
| 21:02.04 | Jonni | smartsearch is the root of all evil, it indexes everything and makes everythinhg slow, it indexes all your audio, video, pictures, emails, facebooks, sms'es etc. |
| 21:02.23 | loin | Jonni, but i don't have any filees on the phone |
| 21:02.26 | loin | so it doesn't really index anything |
| 21:02.47 | Jonni | basicly it spies everything that you do on the phone and puts it on the database. All the incoming and outcoming calls and messages. |
| 21:02.49 | joecool | Jonni: where's it located at so i can chmod it? |
| 21:02.57 | joecool | (android has same issue, called mediascanner or something) |
| 21:03.24 | Jonni | "chmod a-x /usr/bin/smartsearch" |
| 21:04.44 | joecool | thanks |
| 21:05.52 | loin | Jonni, i added that info to my manifest and it still doesn't run as root |
| 21:06.54 | Jonni | loin: did you make the deb file with creator or scratchbox? :), by default creator doesnt include aegis manifests in deb file :) |
| 21:07.03 | *** join/#harmattan lucido (~chris@183.89.147.126) |
| 21:07.16 | Jonni | yet another nice feature that took me a while to notice |
| 21:07.17 | loin | qt creator, of course |
| 21:07.23 | loin | i don't really enjoy scratchbox |
| 21:07.39 | Jonni | well I bet that aegis file is missing from the deb package and script isnt runned as root |
| 21:07.54 | loin | i only used scratchbox when compiling my systemui |
| 21:08.34 | joecool | Jonni: curious, how long can you stretch on the battery? |
| 21:09.03 | Jonni | loin: you can easily check if root permissions are missing by " tail -20 /var/lib/aegis/restok/restok.conf" |
| 21:10.10 | Jonni | if it doesnt have request UID::ROOT then your creator didnt create valid deb file |
| 21:11.26 | Jonni | loin: if I remember right the trick was to remove AutoGenerateAegisFile line from manifest or otherwise creator just didnt do anything. |
| 21:11.27 | loin | Jonni, nothing on the feelcache |
| 21:12.20 | loin | thanks Jonni |
| 21:12.24 | loin | i'll try that |
| 21:12.31 | loin | saved me a lot of time on debugging :P |
| 21:12.47 | Jonni | joecool: it depens on usage, if I dont do anything with it and keep it on low powermode, it can last something like 4-5 days, but usually when I actively use it, I keep it on charger every night, and even on working hours on my workdesk. |
| 21:15.52 | loin | Jonni, joecool http://i.imm.io/14S3M.png |
| 21:15.55 | Jonni | joecool: and on long travels I have this with me http://www.amazon.co.uk/10000mAh-Multi-voltage-Recorders-Lightning-Sensation/dp/B006WHMKEE/ makes live your easy to have extra 10000mAh, nice bonus for the default 1400mAh phone battery. |
| 21:17.11 | loin | Jonni, for that small change i needed inception |
| 21:17.19 | loin | to override systemui |
| 21:19.18 | Jonni | I find using inception to overrote sustemui a bit bad practice, since it changes the origin of the package. Personally I just override packages with com.nokia.maemo origin, that way you can easily replace systemui back to stock systemui with just apt-get command. (where it would fail if you just incept the package) |
| 21:20.13 | loin | i never intend to go back to systemui |
| 21:20.36 | loin | also, removing auto generated aegis worked |
| 21:20.42 | loin | boy, this qt creator is a pain |
| 21:21.13 | Jonni | well yes, but if someday there is some system package updated that is depending on some incepted package, end users might be in pain. |
| 21:21.29 | joecool | Jonni: yeah i got a similar battery unit because my ion only gets ~4 hours of battery life when on cell network with vpn |
| 21:21.55 | joecool | with the backup i can stretch it through the day, it's just annoying to lug around |
| 21:22.13 | Jonni | personally I just have added my own pgp keys are com.nokia.maemo trusted keys, and when I sign my debian packages with my pgp key, system thinks that they are coming from com.nokia.maemo origin |
| 21:22.17 | loin | Jonni, any tutorial on your method of overriding systemui? |
| 21:23.29 | Jonni | so I can just add my own repo to sources.list and just apt-get my overriding systemui :) |
| 21:23.51 | loin | Jonni, sounds like a nice plan |
| 21:24.18 | loin | Jonni, but i used whatever method i found at the time |
| 21:24.34 | loin | i don't like having inception |
| 21:24.53 | Jonni | well yes, most people are just using inception because they dont want to learn how aegis really works. |
| 21:25.16 | loin | Jonni, is it documented anywhere? |
| 21:25.33 | Jonni | loin: http://harmattan-dev.nokia.com/docs/library/html/guide/html/Developer_Library_Developing_for_Harmattan_Harmattan_security.html |
| 21:25.57 | Jonni | yes its documented |
| 21:27.02 | loin | Jonni, did you write any tutorials on that replacing system ui without inception thing? |
| 21:27.32 | joecool | hm so it seems like for openvpn i'd need openmode kernel? |
| 21:27.48 | Jonni | joecool: openvpn works without openmode kernel. |
| 21:28.20 | Jonni | joecool: I have openvpn running on stock kernel, but yes I used opensh to get all the required credentials to run it |
| 21:28.53 | Jonni | loin: sorry, not that many tutorials around, most of that info is just in my head :) |
| 21:29.07 | joecool | ok, will need this working by next week so phone is usable in canada |
| 21:29.46 | loin | Jonni, too bad, i'd really love to get rid of that inception thing |
| 21:30.33 | loin | Jonni, how do you override package origins? |
| 21:30.49 | Jonni | lets see if I have time some day, I was planning to write triple boot tutorial soon also. |
| 21:31.08 | Jonni | basicly I've just patched dpkg to allow AEGIS_FIXED_ORIGIN export. |
| 21:32.08 | loin | that's it? |
| 21:32.15 | fmunozs | but only in openmode? |
| 21:32.26 | Jonni | so I just run "AEGIS_FIXED_ORIGIN=com.nokia.maemo/ovi dpkg -i foobar.deb" if I want package installed as ovi credentials, and "AEGIS_FIXED_ORIGIN=com.nokia.maemo dpkg -i package.deb" if I want to package appear as nokia system pacakge |
| 21:32.38 | Jonni | fmunozs: I have enabled that in secure mode :) |
| 21:33.44 | loin | Jonni, don't you need open mode to AEGIS_FIXED_ORIGIN ? |
| 21:34.22 | Jonni | so basicly all the benefits of open mode export, but without need to ever go into open mode :). So you dont need openmode :) |
| 21:35.00 | Jonni | by default you need to be in openmode to use that export, it just requires a bit of paching to enable that in secure mode. |
| 21:35.20 | fmunozs | but you can't "easily" replace dpkg with a patched one, or can you? |
| 21:35.43 | Jonni | I find using export much cleaner solution than inception :) |
| 21:35.56 | loin | Jonni, any commits you can point to for the dpkg thing? |
| 21:36.43 | Jonni | fmunozs: well define easily, I just have a 3-line shell script that does the patching, so its semi-easy. :) |
| 21:37.50 | fmunozs | not sure, I was just guessing that dpk is protected by aegis so you couldn't replace it without causing MALF on reboot |
| 21:38.21 | fmunozs | dpkg* |
| 21:39.14 | Jonni | fmunozs: well yes simple replace causes malf. basicly line 1 just does echo trick to selected place, line 2 calculates new sha1sum for reflashlist and line 3 just resigns refhashlist with opensh capabilities. |
| 21:39.50 | Jonni | ie if you patch things, you need to make corresponding changes to refhashlist to keep system happy |
| 21:40.09 | fmunozs | yep |
| 21:41.38 | Jonni | but after the patch life is easy and you never need to use inception again :) |
| 21:56.11 | fmunozs | :/ can't run QtSDK on new ubuntu |
| 21:56.59 | Jonni | 13.04? |
| 21:57.04 | fmunozs | seems i found a fix http://www.andreagrandi.it/2012/01/12/nokia-qtsdk-installer-crash-on-ubuntu-how-to-fix-it/ |
| 21:58.24 | fmunozs | yea, 13.04 and that fix -style cleanlooks worked |
| 21:59.33 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 22:56.53 | *** join/#harmattan ToAruShiroiNeko (~admin@wikimedia/ToAruShiroiNeko) |
| 22:59.34 | *** join/#harmattan Morpog_PC (~Morpog@stgt-5d8427fa.pool.mediaWays.net) |
| 23:04.10 | *** join/#harmattan xsacha (~quassel@124.148.30.36) |
| 23:30.51 | *** join/#harmattan Chen (~108406587@117.136.14.52) |
| 23:37.11 | *** join/#harmattan loin (~al@79.113.83.229) |
| 23:37.31 | loin | how can i set context properties? |
| 23:40.37 | *** join/#harmattan mag0g (~magog@2001:8b0:bedb:0:6ee9:7ff:fe8c:518c) |
| 23:48.55 | *** join/#harmattan tonyoy (~tonyoy@cor13-3-88-164-147-63.fbx.proxad.net) |
