| rinigus | mal: thank you very much for looking into it! | 03:48 |
|---|---|---|
| rinigus | piggz: fingers crossed | 03:49 |
| T42_ | <adampigg> https://irc.thaodan.de/.imgstore/2b0bc74c/file_2651.jpg | 09:05 |
| T42_ | <adampigg> https://irc.thaodan.de/.imgstore/27ac09fe/file_2652.jpg | 09:06 |
| piggz | rinigus: guess i have some debugging to do | 09:09 |
| T42_ | <b100dian> You do have telnet hopefully:) | 09:09 |
| piggz | nope, not on lvm volume | 09:09 |
| piggz | rinigus: | 09:25 |
| piggz | Sep 14 12:24:27 VollaPhone sailfish-device-encryption-community-wizard[883]: DevEnc::Dev:227: Failed to activate device. Device: /dev/sailfish/home_open | 09:25 |
| piggz | Sep 14 12:24:27 VollaPhone sailfish-device-encryption-community-wizard[883]: DevEnc::Dev:132: Failed to encrypt and format device. Device: /dev/sailfish/home_open | 09:25 |
| T42_ | <edp_17> Hi all! Can I build 4.2.0.21 with tooling&target 4.2.0.19 or there will be available 4.2.0.21 version for both? | 10:10 |
| piggz | im sue lbt will have 4.2 targets avaialble by lunch time ;) | 10:11 |
| T42_ | <edp_17> @adampigg : Good news! I'll wait then. Thanks! 👍 | 10:13 |
| T42_ | <edp_17> Another question. 😊 Building 4.1 I needed to use a specific commit of droidmedia, gst-droid and gmp-droid. Do I need to use the same for building 4.2 or I can use the latest from each? | 10:15 |
| lbt | rofl | 10:17 |
| piggz | :) | 10:18 |
| piggz | im sure there should be a voting process where the community can dictate jolla's immediate priorities | 10:19 |
| rinigus | piggz: sorry for delay - was afk. have you checked that you have device /dev/sailfish/home_open? | 10:55 |
| piggz | rinigus: yes, also found this.... | 10:55 |
| piggz | https://paste.mozilla.org/4XC7i1m9 | 10:56 |
| rinigus | piggz: can you check /etc/fstab on whether you have /home mount there? | 10:57 |
| rinigus | I wonder if it was mounted when accessed by wizard | 10:58 |
| piggz | rinigus: i dont | 10:58 |
| rinigus | piggz: `df -h` ? | 10:58 |
| rinigus | ls -l /dev/sailfish | 10:59 |
| rinigus | and let's check also home_open size with lvm. maybe it was not resized | 10:59 |
| piggz | https://paste.mozilla.org/oncsiDBF | 10:59 |
| rinigus | lvdisplay, I think | 10:59 |
| rinigus | I have a meeting in 0 minutes, will be back as soon as I can - maybe 1h? sorry | 11:00 |
| rinigus | piggz: ^ | 11:00 |
| piggz | rinigus: np :) | 11:01 |
| piggz | https://paste.mozilla.org/8AsNyksA | 11:03 |
| T42_ | <elros34> @edp_17 most likely you can use 4.2.0.19. IIRC gmp-droid should be 0.2 and other packages also can't be latest. For gst-droid I used https://github.com/sailfishos/gst-droid/commits/upgrade-4.2.0 without latest commit which breaks video recording for me | 11:59 |
| T42_ | <elros34> for droidmedia I think 9450a66a133ffa4df4af5f1d2af5b09ff794606d | 12:01 |
| T42_ | <edp_17> I have used these: | 12:01 |
| T42_ | <edp_17> droidmedia - 9450a66a133ffa4df4af5f1d2af5b09ff794606d | 12:01 |
| T42_ | <edp_17> gst-droid - 364c13e3f6b3d99a87b58beb526bd13f377ce6f4 | 12:01 |
| T42_ | <edp_17> gmp-droid - b61b740987a9880fccf5401220644c7dcf184be2 | 12:01 |
| T42_ | <edp_17> for building 4.1 | 12:02 |
| T42_ | <adampigg> i think for 4.2, you dont need to package gst-droid/gmp-droid because there is a new common repo which contains droidmedia-devel .... however, there was an issue with aarch64, so i think for that arch you have to wait until 4.3 for the benefit ... mal can confirm | 12:02 |
| T42_ | <elros34> @adampigg where do you see such a repo, I have 4.2.0.19 and I do not see droidmedia-devel when searching via zypper | 12:08 |
| mal | @adampigg what issue? | 12:10 |
| T42_ | <adampigg> mal: no issue ... was just mentioning the new common repo for the media stuff | 12:11 |
| mal | @adampigg not sure what you meant with "there was an issue with aarch64" | 12:13 |
| T42_ | <adampigg> mal: you gave me the link, where aarch64 didnt make it to 4.2 | 12:13 |
| T42_ | <adampigg> i think! | 12:13 |
| mal | no idea what you are talking about :) | 12:14 |
| T42_ | <adampigg> *sigh* searching logs! :D | 12:14 |
| T42_ | <adampigg> mal: https://github.com/mer-hybris/community-adaptation/pull/4/files | 12:15 |
| T42_ | <edp_17> Does somebody know when we can start building 4.2 on OBS? | 12:21 |
| piggz | @edp_17 keep calm an wait on lbt :) | 12:23 |
| T42_ | <edp_17> Okay 😊 | 12:23 |
| T42_ | <edp_17> I didn't mean to urge, just asked. 😊 | 12:24 |
| T42_ | <elros34> usually after particular sfos version is released, 4.2.0. is still in EA | 12:27 |
| T42_ | <edp_17> Oh, I though with the 4.2.0.21 the EA is over. I though the EA is always the first version of of an upgrade (like 4.2.0.19) and thought the following update is the final (like 4.2.0.21). | 12:30 |
| mal | I have the new adaptation-common repo in 4.2.0.21 | 12:33 |
| T42_ | <adampigg> mal: i just searched that repo, and it doesnt have the media stuff in it ... is that correct? is that changing? | 12:37 |
| T42_ | <XAP2P> When ban? (re @💝💝💝 ГPUППA_ZНAК0MCTV_18+: ) | 12:41 |
| mal | @adampigg then what does it have? | 12:42 |
| T42_ | <adampigg> https://irc.thaodan.de/.imgstore/95ce9f73/file_2658.jpg | 12:42 |
| T42_ | <adampigg> mal: ^^ | 12:43 |
| mal | @adampigg now I remember, droidmedia and friends only appear there after next release | 12:53 |
| T42_ | <adampigg> ok | 12:53 |
| mal | other stuff is there already | 12:53 |
| T42_ | <adampigg> so, @elros34 i was a release early :) | 12:54 |
| rinigus | piggz: this all looks as it should be. can you paste devices.ini? | 13:01 |
| piggz | https://paste.mozilla.org/qNV86awv | 13:02 |
| rinigus | this one is fine as well. hmm | 13:03 |
| rinigus | piggz: if you reboot, I expect that you will be asked to setup the device again. would you mind testing without encryption? | 13:05 |
| piggz | sure | 13:05 |
| rinigus | I presume you exited wizard through development Quit button | 13:06 |
| piggz | correct | 13:06 |
| rinigus | I start to wonder whether cryptsetup is installed ... but let's see if it works without encryption | 13:07 |
| piggz | rinigus: also found this Sep 14 16:05:51 VollaPhone sailfish-device-encryption-community-wizard[916]: device-mapper: reload ioctl on failed: Operation not permitted | 13:07 |
| piggz | and | 13:08 |
| rinigus | piggz: that I will have to check at home and see whether I have something like it on dev device | 13:08 |
| piggz | Sep 14 16:05:51 VollaPhone kernel: (7)[916:sailfish-device][dm-crypt] dev path: /dev/sailfish/home_open, type: -1 | 13:08 |
| piggz | Sep 14 16:05:51 VollaPhone kernel: (7)[916:sailfish-device][dm-crypt] dev path: /dev/sailfish/home_open, type: -1 | 13:08 |
| piggz | Sep 14 16:05:51 VollaPhone kernel: (7)[916:sailfish-device]device-mapper: table: 253:2: crypt: Unknown error | 13:08 |
| piggz | Sep 14 16:05:51 VollaPhone kernel: (7)[916:sailfish-device]device-mapper: ioctl: error adding target to table | 13:08 |
| piggz | rinigus: i should point out, i was also unable to get stock encryption working on this device :) | 13:10 |
| piggz | maybe some weird kernel bug? | 13:10 |
| rinigus | piggz: you have this strange error in the beginning of https://paste.mozilla.org/8AsNyksA | 13:10 |
| rinigus | maybe there is some delay before LVM volumes get recognized properly? | 13:10 |
| rinigus | piggz: cryptsetup should be there though - its required by the main package | 13:12 |
| rinigus | let's see if non-encrypted version will work | 13:12 |
| piggz | rinigus: i posted to dm-crypt ML ages ago about cryptsetup nt working on this device | 13:13 |
| piggz | didnt get very far ... you can imiagine how much sympathy a vendor kernel gets | 13:13 |
| rinigus | piggz: you mean that cryptsetup command doesn't work at all on volla? | 13:14 |
| piggz | rinigus: one moment, the ML wesite is down, searching mail | 13:14 |
| piggz | rinigus: this is the post ... it was using the older android 9 port , this is 10 https://paste.mozilla.org/sZYr5VWs | 13:16 |
| piggz | rinigus: non-encrypt worked ok | 13:16 |
| rinigus | piggz: ... reading your email ... | 13:17 |
| T42_ | <TheVancedGamer> Now but my client doesn't support bans (re @XAP2P: When ban?) | 13:19 |
| rinigus | piggz: looks like the same we have in our case. | 13:20 |
| piggz | it does | 13:20 |
| rinigus | on device, if you now unmount /home and try to run that cryptsetup luksFormat /dev/sailfish/home_open - would that fail the same way? | 13:21 |
| T42_ | <adampigg> i suppose i should stop user session if i can! | 13:22 |
| rinigus | then you can reset settings by removing /etc/sailfish-device-encryption/config.ini and removing home.mount from /etc/systems/system. | 13:22 |
| rinigus | if you remove or rename devices.ini it will boot without /home mounted and wizard bothering you | 13:23 |
| rinigus | piggz: ^ | 13:23 |
| piggz | rinigus: errr, no, it seeme to work! | 13:23 |
| rinigus | piggz: "it"? | 13:24 |
| piggz | https://paste.mozilla.org/85Zv1jxC | 13:24 |
| piggz | whats command to mount? | 13:25 |
| rinigus | piggz: luksOpen makes it available as device via mapper | 13:26 |
| rinigus | nice that it works, btw! | 13:26 |
| rinigus | after luksOpen your would have to format it as ext4 and then mount | 13:26 |
| piggz | rinigus: ah, but the open failed | 13:27 |
| piggz | [root@VollaPhone ]# cryptsetup luksOpen /dev/sailfish/home_open home_open | 13:28 |
| piggz | Enter passphrase for /dev/sailfish/home_open: | 13:28 |
| piggz | No key available with this passphrase. | 13:28 |
| piggz | Enter passphrase for /dev/sailfish/home_open: | 13:28 |
| piggz | device-mapper: reload ioctl on failed: Operation not permitted | 13:28 |
| piggz | wrong passwd first | 13:28 |
| rinigus | piggz: but have you had home unmounted? | 13:28 |
| piggz | yes, it was unmounted | 13:28 |
| rinigus | command is cryptsetup luksOpen /dev/sailfish/home_open home_encrypted | 13:29 |
| rinigus | you need to have different mapper name | 13:29 |
| piggz | same | 13:29 |
| piggz | in other terminal it shows | 13:30 |
| piggz | [ 815.026278] (7)[3901:cryptsetup][dm-crypt] dev path: /dev/sailfish/home_open, type: -1 | 13:30 |
| piggz | [ 815.026326] (7)[3901:cryptsetup][dm-crypt] dev path: /dev/sailfish/home_open, type: -1 | 13:31 |
| piggz | [ 815.026346] (7)[3901:cryptsetup]device-mapper: table: 253:2: crypt: Unknown error | 13:31 |
| piggz | [ 815.026363] (7)[3901:cryptsetup]device-mapper: ioctl: error adding target to table | 13:31 |
| rinigus | piggz: this is quite bad. no idea and indeed, looks like something is wrong on kernel/driver side | 13:32 |
| rinigus | I wonder what this error type: -1 means | 13:32 |
| rinigus | or that next line. with type: -1 I guess I can check at home and see if I get the same | 13:37 |
| piggz | rinigus: does android use basically the same method for its encryption? | 13:39 |
| piggz | also, what encryption method is used? | 13:39 |
| rinigus | piggz: https://source.android.com/security/encryption/full-disk#how_android_encryption_works | 13:40 |
| rinigus | piggz: does encryption work on volla android? | 13:42 |
| piggz | rinigus: i dont know .... i presume so, as i have to format /data to install on it | 13:43 |
| T42_ | <eriki73> piggz: are you using the same encryption algorithm on SFOS compared to Android? | 13:44 |
| rinigus | piggz: do you have recovery working on volla? so, if you install and encrypt on android, would you be able to check out what is /data? | 13:44 |
| T42_ | <adampigg> @eriki73 algo is https://paste.mozilla.org/s8LiBsP2 | 13:45 |
| T42_ | <eriki73> @adampigg , try aes-cbc-essiv:sha256 | 13:46 |
| T42_ | <eriki73> That is what Android claims to be using... | 13:47 |
| T42_ | <eriki73> In Android docs... | 13:47 |
| piggz | same https://paste.mozilla.org/JPanLah0 | 13:50 |
| piggz | @eriki73 do you have a working v10 install to check there? | 13:50 |
| T42_ | <eriki73> piggz: No, I am daily-driving my Volla on Droidian. | 13:51 |
| piggz | not sfos 4.1??? | 13:51 |
| rinigus | @eriki73: with cryptsetup? | 13:51 |
| T42_ | <eriki73> rinigus: it is not encrypted. | 13:52 |
| piggz | @eriki73 you could try creating a file though | 13:52 |
| T42_ | <eriki73> piggz: a file? | 13:53 |
| rinigus | yes, file would work fine for testing | 13:53 |
| rinigus | create sparse file and then format that with luksFormat | 13:53 |
| piggz | make it big enough to be your ~, then mount it and secure your data :D | 13:54 |
| rinigus | piggz: I guess for you file fails the same way, right? | 13:55 |
| piggz | rinigus: yes, same error | 13:57 |
| piggz | https://paste.mozilla.org/L7SSJcBZ | 13:58 |
| rinigus | piggz: android it is then :) . assuming that you can somehow pull android encrypted data from the phone... is there some kind of dump that let's you copy it as an image? | 13:59 |
| piggz | :/ | 14:00 |
| rinigus | piggz: any other device where encryption is known to work? | 14:01 |
| piggz | rinigus: pro1 works using stock.... | 14:01 |
| piggz | not sure I want to force everyone to reflash for communitty | 14:01 |
| rinigus | :) | 14:03 |
| rinigus | piggz: https://forums.ubports.com/topic/6076/device-mapper-crypto-not-working-on-volla-phone/2 | 14:12 |
| piggz | nice find | 14:14 |
| rinigus | sounds like you need to patch kernel and add "sailfish" to the list... or try to get rid of these "acceleration" hacks | 14:14 |
| T42_ | <adampigg> @NotKit ^^^ as ubports maintainer, thoughts? | 14:15 |
| piggz | rinigus: at least maintainer is here :) | 14:16 |
| T42_ | <NotKit> first time I see this myself, so no idea, but could just modify it to treat unrecognized storage as external storage | 14:20 |
| T42_ | <NotKit> so int type = 1; by default | 14:20 |
| piggz | yeah, i see a kernel mod coming :) | 14:22 |
| T42_ | <eriki73> When I saw the error, I was thinking that it looks like there is some MediaTek coding quality in the kernel. I think I was correct... | 14:29 |
| T42_ | <adampigg> rinigus: @NotKit @eriki73 https://paste.mozilla.org/RZ9YYujZ | 15:28 |
| piggz | rinigus: can i reset everything so the wizard starts again? | 15:30 |
| rinigus | piggz: yes. remove config.ini in /etc/sailfish-device-encryption . I think that is sufficient | 16:04 |
| rinigus | if something fails, let's remove /etc/systemd/system/home.mount . but it should be done automatically by wizard | 16:05 |
| rinigus | piggz^ | 16:05 |
| piggz | it worked | 16:07 |
| piggz | :) | 16:07 |
| rinigus | piggz: regarding reset - we can later discuss whether it should be in GUI. I have a scenario where it can happen by accident with a kid holding your phone | 16:07 |
| piggz | heh | 16:07 |
| piggz | rinigus: should we offer different algorithms? | 16:10 |
| rinigus | piggz: for encryption? not sure. gentoo howto had a statement similar to don't touch algos unless you really know what you do | 16:21 |
| piggz | rinigus: ok, lets leave that idea for now | 16:41 |
| piggz | rinigus: anything i can test for you? | 17:10 |
| piggz | rinigus: found an odd bug | 17:12 |
| rinigus | piggz : impossible! Should be no bugs... | 17:16 |
| piggz | heh | 17:17 |
| piggz | i set password to qwertyuiop | 17:17 |
| piggz | i can see on the keyboard that it is detecting all the keypresses | 17:17 |
| piggz | but, it nearly always misses off the last p | 17:17 |
| piggz | but i can see on the keyboard it is pressed | 17:17 |
| rinigus | piggz that's this keyboard on device boot, right? I wonder whether some other signal has to be used.. currently it's onclicked | 17:20 |
| piggz | rinigus: yes, the password entry keyboard | 17:21 |
| piggz | rinigus: i think its all the edge keys | 17:22 |
| piggz | eg, 1q0p that are worst | 17:22 |
| rinigus | piggz as we can see them fine, would probably handle onpressed then... | 17:27 |
| piggz | rinigus: you want a PR or have you got it covered? | 17:36 |
| rinigus | piggz: PR would be wonderful. especially a tested one - then I can continue with hwcrypt :) | 17:37 |
| piggz | ok | 17:38 |
| rinigus | its the same keyboard in wizard and ask password - 2 PRs? | 17:38 |
| itex | hello guys, I am still struggling to get the local manifest file working, I have used the right version of cm like written on the lineage wiki and used the right depo but it fails to find them when i try to repo sync https://www.3hg.fr/CHATONS/Pastebin/?21ade830ab9f9cd2#55GuKvdHfCrSjWrHCwfAS5omYsBUhtzgMWSWSxUirmoM | 17:41 |
| itex | I don't know where my problem is | 17:41 |
| piggz | rinigus: something weird on my device tooo.... | 17:51 |
| piggz | after unlock, boot continues | 17:51 |
| piggz | lipstick shows up, but seems to stutter at start... | 17:52 |
| piggz | but, main thing | 17:52 |
| piggz | backlight doesnt turn off when i press power to off the screen | 17:52 |
| rinigus | piggz: if you can - check whether systemd-ask-password-gui is running. service should be OK, but process should be killed | 17:53 |
| rinigus | and anything else odd in systemctl status? any odd services hanging? | 17:53 |
| rinigus | piggz: I also had to add https://github.com/sailfishos-sony-tama/droid-config-sony-tama-pie/blob/hybris-10/sparse/etc/systemd/system/systemd-ask-password-gui-stop.service.d/50-vendor-hwcomposer-2-3.conf | 17:54 |
| T42_ | <A_T_R> Your branch is cm-14.1 | 17:55 |
| T42_ | <A_T_R> not "stable/cm-14.1" (re @SailfishFreenodeIRCBridgeBot: <itex>hello guys, I ...) | 17:55 |
| piggz | rinigus: only the tty ask passwd is running | 17:56 |
| T42_ | <A_T_R> @itex - set revision="cm-14.1" | 17:56 |
| rinigus | piggz: good. now question is whether that hwcomposer restart is needed for you as well | 17:57 |
| rinigus | I can see in your github sources that one is missing | 17:58 |
| itex | I think it worked thanks A_T_R | 18:01 |
| piggz | rinigus: grrrr, i cant get the backlight off! | 18:33 |
| rinigus | piggz: even after that hwcomposer restart? | 18:34 |
| piggz | yup | 18:35 |
| piggz | rinigus: found something | 18:37 |
| rinigus | piggz: was just going to type that I have no idea | 18:37 |
| piggz | rinigus: /sys/class/leds/ | 18:37 |
| piggz | contains lcd-backlight/brightness | 18:38 |
| piggz | echoinf 0/255 into that does nothing | 18:38 |
| piggz | also has mt6370_pmu_bled/brightness | 18:38 |
| piggz | echoing into that works as expected | 18:38 |
| piggz | so, must be a config problem | 18:38 |
| piggz | but, i dont know whyu ive just noticed it | 18:38 |
| rinigus | piggz: but has it worked before? | 18:38 |
| piggz | im _sure_ it worke before | 18:39 |
| piggz | thought, the lvm build, which is new, does have that brighness config in it, which is wrong in my case | 18:39 |
| rinigus | piggz: it could be that the boot sequence is somewhat altered as well. as encryption depends on some android bits (in getting ready to hwcrypt as well as hwcomposer). so, in theory, some services could start earlier than before now | 18:40 |
| rinigus | but it could be error in config that got triggered somehow now | 18:41 |
| piggz | rinigus: i think its a config issue, added 99-backlight.ini with correct path, it seems fine | 19:06 |
| piggz | now, back to fixing keyboard | 19:06 |
| piggz | rinigus: what line endings do you have on that file? | 19:22 |
| rinigus | piggz: ******** no idea why crlf sneaked in. feel free to change to unix | 19:23 |
| piggz | :D | 19:24 |
| piggz | my git diff was showing ^M ... which got me wondering :) | 19:24 |
| rinigus | piggz: thanks! | 19:39 |
| rinigus | did it help? | 19:39 |
| piggz | rinigus: yes, it fixed it for me | 19:41 |
| rinigus | piggz: excellent! now same for wizard? | 19:43 |
| piggz | yup... | 19:43 |
| rinigus | piggz: updates required for hwcrypt are all in. we can now create hw backed key from settings GUI and use it during a boot | 19:56 |
| rinigus | you would have to update hwcrypt and 3 other packages (see what's landed today). assuming that systemd-ask-password is updated already with your fix | 19:58 |
| rinigus | closing hwcrypt issue - open a new one if it fails for you :) | 19:59 |
| piggz | ok ... | 20:00 |
| piggz | bug confirmed in wizard | 20:00 |
| rinigus | piggz: new bug? or something with keyboard? | 20:02 |
| rinigus | I'll be off till tomorrow (+2h) ... if new, maybe should get that at github | 20:03 |
| piggz | rinigus: same kb bug | 20:03 |
| rinigus | piggz: great - then you have the fix :) | 20:04 |
| itex | if i use lineage os manifets I cannot use others from non lineage os? "breakfast $DEVICE" asks me for some repos but the ones from lineageos don't seem to work https://www.3hg.fr/CHATONS/Pastebin/?041c15af56f887ce#3U51HzRWgUaEne2NTUuim85XX2JK9AV6o5epxHLospi4 | 20:30 |
| Keto | and seems it came back online... | 20:54 |
| T42_ | <XAP2P> Google translator xD (re @SailfishFreenodeIRCBridgeBot: <Keto>PSA: the sailb...) | 21:28 |
| T42_ | <edp_17> Good evening all! I need a little help. SFOS 4.2.0.21, treltexx, CM14.1 base, device stays on the yamui splash screen because lipstick doesn't start. Journal: https://paste.ubuntu.com/p/sPBVXNMqND/ | 21:54 |
| T42_ | <edp_17> I've updated tooling&target to 4.2.0.19 and all three submodules. | 21:55 |
| T42_ | <edp_17> Can you help to find out what is causing the issue, please? | 21:56 |
| T42_ | <edp_17> Oh, the initial setup couldn't run, therefore the device is only accessible with telnet. | 21:57 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
