**** BEGIN LOGGING AT Thu Jun 14 03:00:05 2018 Jun 14 05:48:47 anyone here who ordered with czech keyboard and received his gemini or at least have order locked? Jun 14 07:40:00 Is there some usable way to get a Google free Android on the Gemini apart from the inofficial Lineage ROM? Jun 14 07:55:00 matthegap[m]: You could use yalp Jun 14 07:55:01 (store which gets the apks from google store without needing a google account) Jun 14 07:56:12 Bmifritscher1 But that only makes sense, if I can get rid of the preinstalled Google Apps and services first. Jun 14 07:59:50 morning all, has anyone noticed a time discrepancy between the real system time & the one android displays at the top right corner? Jun 14 08:00:07 lock screen & time on the appbar are correct, the top corner one lags behind Jun 14 08:01:32 matthegap: i deactivated all google apps i could and blocked all connections with NetGuard Jun 14 08:02:01 you could also kill the apps directly but it could damage the stock android... so the only clean way is the lineage rom Jun 14 08:02:18 (i wait till there is a release on which wwan works... then i switch) Jun 14 08:03:25 devyl: That leaves me with the Play Services running with root rights. Jun 14 08:03:25 My dream would be an official release from Planet Computers of their regular image with the Google apps removed. Jun 14 08:03:52 did you email them to ask for such a release? Jun 14 08:04:28 you can still kill the google play services (after you uninstalled the other google apps) Jun 14 08:04:29 mom Jun 14 08:06:02 https://android.stackexchange.com/questions/91628/how-to-remove-all-google-specific-apps-from-a-rooted-android here is a list/howto Jun 14 08:37:15 so guys i'm waiting for all the people telling me they would buy my gemini the moment i got it to step forward and, you know, buy it now that i have it Jun 14 09:37:56 everyone in here either has one or is waiting for one I imagine Jun 14 10:16:34 Hmm i read that people have had bad experiences with cha Jun 14 10:16:51 rging the gemini with any other charger than the one provided. Jun 14 10:17:24 I tried my cheapo battery pack and a super cheap usb-a to usb-c cable and it seems to charge fine. Jun 14 10:52:00 razumihin: My original Planet charger is still in it's box. I've been using my OnePlus USB-C charger at home and an Eveready multisocket USB adaptor with a generic USB-C cable at work. Haven't had charging issues either....(yet!) Jun 14 10:52:29 Okay, s Jun 14 10:52:35 o people exa Jun 14 10:52:40 ggerate. Jun 14 10:52:53 ... i'm not still used to this enter. Jun 14 10:53:04 but yeah, have seen a few charging posts on FB Jun 14 10:53:39 Yeah, i think I'll try my car charger and my qc3.0 usb hub charger next. Jun 14 10:53:52 my original charger and cable are still in the box too. i tried my google pixel charger, and that didn't work, but any other changers work fine Jun 14 10:54:11 i think it has to do with the amount of current some chargers output Jun 14 10:54:15 My car charger is just a simple regulator so I can't imagine how it could fail. Jun 14 10:54:34 If you want to do more than basic USB charging, you need one that works right Jun 14 10:55:08 I don't like currents... much prefer raisins and sultanas.... (I'll get my coat...) Jun 14 10:56:24 as in, you need one which uses the correct fast charging protocol (that is, one that uses MediaTek's fast charge thing) Jun 14 10:58:02 BTW - just got an Indigogo update email saying they've released a new FOTA update. Not sure I want to be the first to try it though... don't want it to break my dual-boot setup. Who wants to be the guinea-pig? ; ) Jun 14 10:58:29 It did not work on my dualboot firmware Jun 14 10:59:00 Tries to look at the partition size and it's not the same as in non-dualboot so it fails. Jun 14 10:59:03 did it break the dual-boot, or just not update, razumihin ? Jun 14 10:59:18 Just did not work, nothing broke. Jun 14 10:59:38 aah...cheers - you're braver than me! Jun 14 11:45:35 how am i supposed to take the microsd out of this thing... sandisk ultra Jun 14 11:46:02 canny[m]: I had titaniumbackup anyway and I'm on a vacation ;) Jun 14 11:55:45 razumihin: you've just reminded me to install that - now that I'm rooted... Jun 14 11:56:00 :) Jun 14 12:11:31 sb244: i've read you need to lift it a little with some sharp tool to remove it Jun 14 12:13:57 cos-: the tool is well-hidden inside one of black envelopes you received with the Gemini Jun 14 12:14:09 sb244: ^ Jun 14 12:16:27 it's black but has a big label engraved onto it Jun 14 12:27:08 Gemini has come but no accessories :-( Jun 14 12:27:23 good job I prepared myself to be disappointed Jun 14 12:29:13 kilobyte: you mean the cover removal tool or is there some sd removal tool also somewhere? Jun 14 12:30:41 there's also some plastic resin on the top edge of the keyboard :-/ Jun 14 12:31:49 well, the cover is the hard part. SD cards sometimes need some manipulation, but it's not _that_ hard. Jun 14 12:32:15 at least compared to replacing emmc in some devices... Jun 14 12:52:25 yeah i used it to remove the cover, ill try prescision tweezers Jun 14 13:15:12 just put a precision screwdriver in the beveled part of that card and pulled it out Jun 14 13:59:21 question. i have the display brightness set the way i want, low, which makes the keyboard more visible. Jun 14 14:00:03 but my setting doesn't survive the sleep cycle and i have to Fn-B again. have i done something wrong here? Jun 14 14:00:29 how to i make the display level persisent through screen blanking and coming back? Jun 14 14:00:38 this in linux, of course. Jun 14 14:03:59 do I get to send my Gemini back if it's not up to scratch? Jun 14 14:04:24 there's a load of resin on the top edge of the keyboard frame Jun 14 14:04:37 if I'd bought it from a normal retailer, I'd be sending it back Jun 14 14:04:44 do I have any recourse with Indiegogo? Jun 14 14:04:50 * rah is guessing not Jun 14 14:04:57 "you get what you're given"? Jun 14 14:05:26 rah : if you got it from a retailer, it would be a purchase. on igg it's a gift they give you for donating. Jun 14 14:06:14 is it just a cosmetic issue or something else? Jun 14 14:11:32 TheKit: exactly Jun 14 14:13:06 pfft Jun 14 14:13:18 rah: they seem to respond well to nice queries, so its probably worth dropping them an email with a photo to see what they say Jun 14 14:15:20 dep: I did look into it breifly and the code is calling the 'set normal brightness' but it dosn't seem to work out as the right value being set, patches welcome : ) Jun 14 14:17:46 Huzzah - got my unit working with our company's Citrix farm! Hopefully I'll never need to lug around the company laptop when I'm on-call again... Jun 14 14:18:46 Just give me access to the company network and a copy PuTTY and that's all I need! Jun 14 14:20:04 adam_b_, I think I know what happens Jun 14 14:20:30 hwcomposer driver sets brightness to 0 when screen turns off, and it has to set it to some value when restoring Jun 14 14:20:31 (copy ^of^ PuTTY) Jun 14 14:21:06 implementing any interface to set brightness via xrandr should help Jun 14 14:21:11 TheKit: yea and then just after that repowerd sets it to 'normal' which is the user selected value Jun 14 14:21:34 its no longer set to max brightness which is what was happening before we had repowerd in the mix Jun 14 14:22:19 hm, nice then Jun 14 14:23:13 adam_b : i was just wondering if the value is written to a config file somewhere, in which case i'd just set it there. Jun 14 14:23:36 dimming the keyboard makes the keyboard effectively much brighter. Jun 14 14:23:47 probably good for the battery, too. Jun 14 14:25:20 dep: if it was losing the value over a reboot then it would be a saving the setting type fix, but as its only over a open/close i think its something else Jun 14 14:25:21 dimming the display makes the keyboard effectively brighter, and i need more coffee. Jun 14 14:25:49 adam_b: it doesn't read a value from someplace? Jun 14 14:26:29 no the code really is setting a 0.4 or whatever the user setting has picked back, only its not doing it for some reason Jun 14 14:27:03 so this is a known thing. Jun 14 14:27:35 yes I did investigate it but gave up, more important things to work on first Jun 14 14:28:01 whew! it wasn't my screwup! a first! Jun 14 14:29:04 it's also no big deal. and i only break really important things. Jun 14 14:29:51 adam_b_: respond after how long? :-) Jun 14 14:55:32 rah: well yea, it can take a few days/weeks/etc Jun 14 14:59:43 :-( Jun 14 15:00:16 in the meantime my Gemini sits there doing nothing ;-( Jun 14 15:00:20 instead of soldering wires onto the main board to get me a serial console, I thought I'd try for a different approach, and followed the instruction here: http://www.stevenhoneyman.co.uk/2014/11/mtk-mediatek-debug-cable.html Jun 14 15:00:42 rah: I thought it was a cosmetic issue Jun 14 15:01:25 it seems MTK exposes serial output on the USB port during bootup Jun 14 15:02:42 I managed to together a cable, but no output so far :( Jun 14 15:03:00 * to put together Jun 14 15:05:48 is Tx connected? Jun 14 15:06:10 yes it is Jun 14 15:06:43 using GND, RX and TX only... no VCCconnected Jun 14 15:06:53 https://bpaste.net/show/5ba56c35f6c3 Jun 14 15:06:57 that's the start of boot process Jun 14 15:08:01 yeah... if only i got _any_ output... but I see nothing so far, so the cable isn't working yet :) Jun 14 15:09:42 https://bpaste.net/show/4e9f2adbece1 - check is_uart_cable_inserted Jun 14 15:24:18 adam_b_: why did you think that? Jun 14 15:24:23 Omegamoon, in Linux, you can try echo 1 > /sys/devices/soc/11270000.usb3/musb-hdrc/portmode Jun 14 15:24:54 then echo something to /dev/ttyMT* Jun 14 15:27:26 I see... will try that Jun 14 15:28:37 Omegamoon: did you tried both left and right USB-C ports? Jun 14 15:28:49 yes, I tried both ports Jun 14 15:28:58 exit Jun 14 15:29:10 also switched RX and TX... I also get those 2 mixed up :S Jun 14 15:29:25 never mind -- focus on my desktop machine wasn't in the terminal anymore! Jun 14 15:29:36 also=always Jun 14 15:29:59 left port for sure Jun 14 15:30:20 (or, well, pretty likely) Jun 14 15:30:34 rah: just the answer you gave to TheKit's question (I guess the answer was ambigous) Jun 14 15:30:53 did you directly connected the output from USB-C port to UART adapter or via some converter? Jun 14 15:31:38 ah, right... forgot to mention that... I use a usb adapter Jun 14 15:31:53 since I don't have too many usb-c cables Jun 14 15:32:03 yeah... USB-C cables are still expensive Jun 14 15:32:25 UART and USB signals are not electrically compatible so adapter might make some issues imho Jun 14 15:33:55 also I think the dmesg of Gemini is very noisy by default Jun 14 15:34:11 yeah, known issue with all MediaTek kernels Jun 14 15:44:04 hehe i can't even gett accepted into the fb group so have given up (re oesf post) Jun 14 15:44:46 adam_b_: I was saying "exactly; *is* it just a cosmetic issue or something else?" Jun 14 15:47:33 rah: probably best to ignore my comments, I've clearly got no idea what we are actually talking about, I thought you had a device with 'resin' on the top edge and it was just cosmetic etc, so all worked fine but just had a lumpy look Jun 14 15:51:33 guess it depends on just how much resin we're talking about here; big keyboard-disabling glob, or just a little doodle? Jun 14 15:55:58 the blob issue raises its ugly head once again! Jun 14 16:07:23 http://settrans.net/~rah/misc/gemini-resin-01.jpg http://settrans.net/~rah/misc/gemini-resin-02.jpg http://settrans.net/~rah/misc/gemini-resin-03.jpg Jun 14 16:07:26 * rah -> away Jun 14 16:12:48 Anyone try the other-the-air firmware update yet? Jun 14 16:12:53 Over the air rather Jun 14 16:41:39 well now, Ive had the gemini a few hours now, and this is me connected from debin Jun 14 16:41:43 debian Jun 14 16:41:50 :) Jun 14 16:55:03 moon127, congrats on surviving flashing steps :) Jun 14 17:18:48 TheKit: looks pretty good for my use case so far Jun 14 17:19:16 I mean LibreOffice is hellish slow - but I can get a lot of what I need done already Jun 14 17:19:57 just playing for now though, for work use I need to get at least homedir encrypted. Jun 14 17:26:31 I can say for sure my quassel IRC runs nice though ;) Jun 14 17:26:56 and after the EGL steps for chromium videos eg on BBC news are flawless. Jun 14 17:27:10 kudos to all who have worked on the Linux side so far!! Jun 14 18:37:21 TallTim: worked flawlessly Jun 14 18:46:57 moon127, how is Quassel recently? I switched from it to HexChat for some reason, but can't remember why now Jun 14 18:47:18 Thanks mifritscher Jun 14 18:47:28 was curious if they had gotten it right ;) Jun 14 19:21:07 I'll get a second one in a few days, so I can be a bit more brave :-D Jun 14 19:21:50 and the update was completely fluent, I didn't even hav to do manuel presses in recovery (in fact I didn't even see the recovery) Jun 14 19:22:50 btw, does anybody know how to force the "touch symbols" to the right side again? they are at the botton after I decreased the dpi so that the gui elements are not such big any longer) Jun 14 20:07:35 someone know how can i flash manualy only the linux partition of my dualboot ? Jun 14 20:09:45 ju, after it has been partitioned before, yes Jun 14 20:12:57 TheKit: what do you mean ? Jun 14 20:16:00 hi Jun 14 20:20:23 ju, I mean if you have installed dualboot before and keept same scatter file Jun 14 20:44:28 OTA is only for android, or can i switch to multiboot with? Jun 14 20:56:36 TheKit: I ve installed lyke that https://geminiplanet.com/how-to-install-debian-linux-on-the-gemini/ Jun 14 20:56:51 I can take the scatter file given Jun 14 20:57:38 and link the new debian, check only the linux partition and voila ? Jun 14 21:58:03 ju: you can't. Current "Debian" is only a full-screen chroot on an UI-less Android; the latter is needed to boot, access all peripherals, etc, and all the time you have the full set of Android processes running. Jun 14 22:00:40 ju: one guy tried using mainline X20 with added Device Tree parts, the kernel sort of booted but it's only a serial console only affair for now Jun 14 22:03:32 ju: so for now you need to keep the Android partition; you can generate a scatter file that grants only a few gigs of space to Android if you're not going to use it yourself Jun 14 22:08:14 kilobyte: I think you did not understand what I meant Jun 14 22:08:38 kilobyte: i want keep my dualboot Jun 14 22:09:41 kilobyte: only reflash independant versions like newer android without ota Jun 14 22:10:36 not sure if it can be done, the flash tool is extremely quirky Jun 14 22:12:38 kilobyte: so no upgrade possible for dualboot user's ? Jun 14 22:14:22 have you borked your Debian partition somehow? If not, upgrades can be done with apt; there's an additional repository configured. Jun 14 22:15:15 by the way, you really want to add Debian's security, there's many packages with fixed security holes but not on what's flashed Jun 14 22:16:53 adam_b_: could you please update the instructions and tell people to 「echo "deb http://security.debian.org/ stretch/updates main contrib non-free" >>/etc/apt/sources.list」 ? Jun 14 22:18:44 would be good to tell people to purge and reinstall openssh-server, too -- otherwise they're using a publicly known secret key for sshd Jun 14 22:29:13 kilobyte: and for upgrading the android image ? Jun 14 22:30:43 no idea Jun 14 22:30:53 my debian isn't broken, just want to start from good start Jun 14 23:08:45 kilobyte: its a wiki, have at it Jun 14 23:09:47 isn't the github page at https://github.com/gemian/gemini-keyboard-apps/wiki/DebianTP2 the most often linked to one? Jun 14 23:10:30 (where I indeed can make a pull request to) -- what docs would you prefer to be official? Jun 14 23:10:33 if ju returns then my advice would be to wait for the new images from planet and then do a partial flash (known to work with latest tools) Jun 14 23:11:35 that DebianTP page (not the n2 one you mention, thats based on an early release of the next version and some of it shouldn't apply) Jun 14 23:12:27 but sorry my point is that the whole of the gemian wiki (inc DebianTP page) is world writable to anyone with a github account Jun 14 23:13:04 so yes that is as offical as it gets but its no more my responsibility as it is yours Jun 14 23:13:04 ah right Jun 14 23:15:54 though you might want to test a device with whatever updates you propose just incase we need to update a branched item Jun 14 23:16:51 hmm right, good to at least compare versions Jun 14 23:17:59 I just mean that we've patched things like sddm/keyboard layouts etc so when upstream bump the numbers it can revert our version so then you go back to not having those tweaks Jun 14 23:19:06 yeah, obviously. This is probably a nasty thing for you in buster (it is in any of my patched projects...); lemme take a look at stretch. Jun 14 23:19:59 yea on that basis I'm not inclined to recommend buster for general use Jun 14 23:25:07 best way I found is to upstream changes as soon as possible :) Jun 14 23:25:59 wonder how easy that is Jun 14 23:26:13 keyboard stuff might be accepted Jun 14 23:27:26 though it would be nice to have confirmation from at least one person per layout that they find them acceptible Jun 14 23:32:31 wrt security: there's not a single package (source or binary) that's both in security/stretch and in gemian Jun 14 23:34:11 sounds good, just the ssh-server purge/re-install bit we need to mention too I guess Jun 14 23:35:08 as for stretch, you're up to date for all modified packages (last point release was a while ago) Jun 14 23:36:13 some branched things have gone direct to upstream rather than just stretch version Jun 14 23:36:14 sanity check in case I'm doing something wrong: they're lxqt-common, lxqt-globalkeys, ofono, pulseaudio, sddm, xkeyboard-config Jun 14 23:36:43 and the qt stuff Jun 14 23:37:08 I mean packages that exist in both stretch and gemian Jun 14 23:37:27 yea we've re-built qt to use egl Jun 14 23:38:50 not so major if that reverts, just all goes a bit slow Jun 14 23:44:31 got it: the source is missing in gemian Jun 14 23:45:03 all gemian source should be in the gemian repo, I've done the full branching for easy searching Jun 14 23:45:43 I don't see for example qtbase-opensource-src; looking at stretch Jun 14 23:46:06 github.com/gemian Jun 14 23:46:08 deb-src http://apt.gka.thinkglobally.org stretch main Jun 14 23:47:47 uncertain why its not in the deb-src, but I prefer to look when I can see history Jun 14 23:49:42 just added the security updates line and who-ee, there is a lot of stuff! Jun 14 23:51:06 though sudo /usr/lib/aarch64-linux-gnu/vlc/vlc-cache-gen /usr/lib/aarch64-linux-gnu/vlc/plugins segfaults Jun 14 23:52:15 and, oh joy, video acceleration has disappeared. Jun 14 23:54:13 and the packages listed for autoremove include: Jun 14 23:54:26 libcdio13 libdirectfb-1.2-9 libgles1-mesa Jun 14 23:54:51 libiso9660-8 libvcdinfo0 libvlccore8 Jun 14 23:54:55 yea thats the kind of problem we don't want Jun 14 23:55:48 yes. so it seems. damn. i don't guess there's a hitherto unknown apt-get undo . . . Jun 14 23:56:33 looking at binaries, I still don't see a single package that's outdated in gemian, so security updates shouldn't cause a regression unless there's an unpackaged change somewhere Jun 14 23:56:46 some folk on oesf were talking about timeshift, but that would require thinking ahead Jun 14 23:57:25 kilobyte: its more that there are some packages that pull others in and push some out (as per dep's autoremove suggestios) Jun 14 23:57:32 * kilobyte uses btrfs for snapshot goodness everywhere, but that's a bad idea on old kernels. Jun 14 23:58:33 dep: in theory you should be able to look at https://github.com/gemian/gemian-multistrap-config to see what to reinstall/patch Jun 14 23:58:42 most security updates are minimal, backporting only the minimal fix Jun 14 23:59:49 in this case it included a lot, such as Jun 15 00:00:35 bsdutils chromium dirmngr firefox-esr and all of libreoffice Jun 15 00:01:02 prepare for complete utter madness in near future with firefox, though... Like chromium before, this time it's firefox that forces backporting multiple compilers, a buttload of rust packages, plenty of misc libraries, throws away XUL extensions, and so on Jun 15 00:09:34 looks as if slowness is primarily manifested in chromium -- firefox was already painful. Jun 15 00:10:25 i didn't autoremove, btw. **** ENDING LOGGING AT Fri Jun 15 03:00:10 2018