[00:50] still no beta 2? [00:52] vincelaw: Nope. No beta2 [00:58] soon [01:58] ls /usr/share/initramfs-tools/hooks/klibc^i-t [01:58] do you also have this file ? [02:49] allquixotic, keep me posted if you test that script out === littlebunnyfufu is now known as SonikkuAmerica [04:35] http://cdimage.ubuntu.com/ubuntu/releases/16.04/beta-2/ damn, only powerpc and IBM Z builds for now :( [04:37] whops, they are on http://releases.ubuntu.com/16.04/ :) [04:42] pfoo: or, FWIW, you could just use the daily image :) [05:10] I noticed in the RC for ubuntu 16.04 vim is compiled without python support. Will this be changed once released? [05:11] Although, maybe there aren't that many of us who use that. [05:15] It causes some plugins not to work correctly [05:24] ubuntu-desktop beta2 iso not out yet? http://cdimage.ubuntu.com/ubuntu/releases/16.04/beta-2/ [05:32] They are on http://releases.ubuntu.com/16.04/ :) [05:32] ryan_turner: ah tnx [05:33] ryan_turner: did you test? [05:34] i mean install [05:50] Waiting on my new pciex m.2 drive first [05:50] ryan_turner: wich brand? [05:52] Samsung 950 pro [05:52] niceee [05:54] I hope so! Just got the xps 13 today [05:54] Came with 256, which IMO isn't enough. So I was shopping 512 GB drives and that just kept glaring at me [05:55] lol [05:55] it will rocknroll with xenial [06:08] Link to beta 2? [06:09] http://cdimage.ubuntu.com/ubuntu/releases/xenial/beta-2/ [06:09] oh wait thats not right [06:09] http://cdimage.ubuntu.com/ubuntu/daily-live/current/ [06:09] * Ben64 shrugs [06:12] Ben64: ok found [06:12] http://releases.ubuntu.com/16.04/ [06:13] Ben64: I would have thought that link would be on here https://wiki.ubuntu.com/XenialXerus/Beta2 ! [06:14] Some one should stick it in this channels message :) [06:17] hmm cant we update daily to beta2? [06:17] daily is better than beta2 [06:17] ill stick on daily then [06:18] no other changes that i can see on softpedia [06:32] Right now Ubuntu 16.04 is in Final Beta Freeze along with todays release of the Final Beta. From what I understand there will be no more changes or public releases until the FinalRelease April 21st except for the ReleaseCandidate. [06:34] Thats for Ubuntu with unity. All others have more phases I belive [07:03] http://www.ostechnix.com/upgrade-ubuntu-16-04-lts-final-beta/ [07:03] seems interesting === Ian__ is now known as Ian_Corne [09:42] Hai. i'm trying to install Kubuntu Xenial beta 2 but the installer crashed all the time. Any workaround for this? thx [10:17] I got this error when I tried to install mysql on Ubuntu 16.04, http://imgur.com/yPiZwnF. Anyone? [10:17] xenialxerus: did you already have a mysql database on the host ? [10:17] Nope [10:18] Fresh install from this http://cdimage.ubuntu.com/ubuntu-server/daily/current/xenial-server-amd64.iso [10:18] ok - so remove the package and remove the database data files, re-install the package and see if you get the same problem [10:18] How I can remove the database files? [10:19] just delete them off the file system AFTER you've removed the package [10:22] This command right sudo rm -rf /var/lib/mysql/ and sudo rm -rf /etc/mysql/ [10:23] I already removed btw using apt-get [10:23] Anything missing? [10:24] that should do it [10:25] Reinstalling mysql-server now [10:33] Lol, I got this error after reinstalling, http://imgur.com/AU3jssP [10:33] thats not an error [10:33] thats asking you to confirm your password for the root user for mysql (not root ubuntu user) [10:34] this very much looks like your machine is missconfigured rather than a problem with 16.04 [10:34] eg: you must have had an existing mysql database before [10:34] Yeah seem like one [10:35] Where the directory for the database is located? [10:35] I assume this is a test box ? [10:35] /var/lib/mysql [10:35] Yes [10:35] ok - re-install, first package you then install is mysql-server [10:35] Running on VMware Workstation 12 Player [10:35] confirm the bug on a clean machine [10:35] as the machine did not appear as clean as you thought it was [10:38] I think so, i will take a look later on. Thanks for your help ikonia [11:10] I installed xenial in December, and I still have software-center, wasn't that to be replaced by gnome-software? [11:10] Do new xenial installations have it? Should I remove it manually? [11:12] alkisg: new ones should have it [11:12] flocculant: thank you, so, we now have 2 GUIs to manage software? [11:12] well [11:12] you do :) [11:12] as far as I know ubuntu only has g-s [11:12] I know for fact that xubuntu does :) [11:13] (01:10:32 μμ) alkisg: Do new xenial installations have it? Should I remove it manually? [11:13] I meant software-center there [11:13] (01:12:03 μμ) flocculant: alkisg: new ones should have it [11:13] I know what you meant [11:13] I suppose you meant gnome-software there? [11:13] yea [11:13] in the daily you would just see g-s [11:14] Cool, so I'll remove software-center manually then [11:14] Thanks! [11:14] remove usc - in fact if you've not got g-s, then maybe installing that removes usc - not sure [11:14] Using only apt-get dist-upgrade since November, I now have both of them [11:15] I only dist-upgrade if it's really necessary [11:15] dist-upgrade is the normal upgrade, the one we get from the upgrade ui as well [11:15] *update ui [11:15] yea - I usually either cli or synaptic :) [11:16] Haha [11:16] Old guy like me, I suppose... :D [11:16] :) [11:34] guys is 16.04 stable for use? [11:34] It's beta, not stable [11:34] But it works fine here [11:35] that doesn't mean it will work for other people [11:37] alkisg, ikonia tnx again. === clivejo_ is now known as clivejo === ren0v0 is now known as Guest52990 [12:08] Hi, any mods here that can find out why i was just banned from #ubuntu ? [12:09] whoever banned me needs removing as a mod [12:09] ren0v0_: I removed you and you're welcome to join #ubuntu-ops to discuss it === ren0v0_ is now known as ren0v0 [12:10] ty [12:16] Hiyas all === Guest52990 is now known as ren0v0 [13:01] I'm thinking of trying to move from windows to ubuntu. Would it be a good idea to hold off till 16.04 comes out? Or would it be roughly the same to go into 14.04 and upgrade in a week or two? [13:02] ubuntu218: use 15.10 and upfrade from april to july [13:04] Alright. I might should try that. I do have the 14.04 currently, but I figured for a new user to linux, sitting on a lts version would be a good idea [13:05] Unbuntu makes me feel stupid sometimes.... most of the time. Since I've been on windows since XP when I was like, single digits years old. Switching is like trying to use a unicycle. [13:06] ah you are already on 14.04? then use that [13:06] I installed it a few days ago, and I'm still trying to figure out how to use it. [13:06] and the LTS to LTS upgrade to 16.04 will be opened in july when 16.04.1 is released. like the first "servicepack" [13:07] Like, I can install steam and skype, get them to run on bootup. But then anything that wants me to use Mono or Wine seems to act wonkey [13:07] I really should finish watching that ubuntu tut video [13:08] some things are different to windows. but rule of thumb is: use the service the package-system offers you. and most things are common sense :) dont look at the windows way, but at the "what would be general clever way" :) [13:09] Actually, I do things the windows way so far. In by that I mean "google.com/search/howtoinstallgzfiles" and "google.com/search/howtofeellessstupid" [13:09] nonono. dont install stuff from websites [13:10] if you want to install $program, first look into the package system (softwarecenter or apt on cli) if ubuntu doesnt ship already the program made working with ubuntu. [13:11] some of the programs I want aren't in the software center. I know some things I'm wanting to run req some forms of compatability help. Wine/Mono/etc [13:12] and some things flat out won't work. So I'm dualbooting until I can figure out how to move my windows into an ISO to throw it into a vm. [13:12] because some things just aren't made for Linux [13:12] demanding I install them in a C drive and whatnot [13:13] yeah. some programs are only made for windows. but you can look if there are linux alternatives. so you get rid of all the windows only stuff. [13:13] oddly enough though, it feels like the 14.04 has some better drivers than my windows. [13:14] I'm fairly sure there isn't a devpro for linux. [13:14] so i would handle that program by program. "i want a program like photoshop. can i run that in wine or is there an linux alternative" [13:15] The ones that don't work tend to be little indi projects of sorts. That and an MMO because the publisher is cheap [13:16] yeah. for games most stuff needs wine. there you can look at play-on-linux. which setups wine as needed. [13:16] but wine will always be only a compatibily layer. [13:16] the mmo basically demands directx. It's flat out not going to run. [13:17] My best bet for that is either a dualboot or a VM [13:17] look at the wine app database for that game. directx games do run in wine [13:18] I've done a google search and everything from the last 3-5 years say "Yeah, not working" [13:18] the game in question is MapleStory. But I probably can live without that game. [13:19] Since I'm basically end game and the only real progress comes in the form of throwing money at the game if I want to actually progress from where I am now. [13:19] or your go dualboot. i still have a dualboot with my gameloader aka windows :) [13:20] I'm dualbooting right now. [13:20] It's a pain in the butt though. Because every time I try to move into the linux base to try and learn how it works, BAM someone pings me and asks me to swap over to play a game with them. [13:21] hehe [13:21] and some of my other games don't have 14.04 support. Emulators and whatnot [13:21] those say to go v15.xx [13:21] Sounds like you know the pain of dualbooting [13:22] xx is running so perfectly right now, im loving it [13:23] I'm saddened by the windows snowball cycle. [13:23] i dont use wine at all. my daily usage is 100% linux compatible. i only play some game from time to time i have to dualboot for. [13:23] games don't get made for linux because no one runs it. No one runs it because their games won't play on it [13:24] goto start [13:24] :( i ditched windows long back fortunately for me [13:25] Man, I installed steam on unbuntu and steam cut my playable games from like 300 down to low 20s. [13:25] ubuntu218: that is not true. steam got a native linux client and ports some games to linux [13:25] It either isn't working, or is because I don't know what I'm doing yet. [13:26] Which is actually pretty likely that it's because I don't know what I'm doing [13:26] ubuntu218: even some games a while ago had native linux clients, like wolfenstein or the enemy territory addon. the whole halflife engine got ported to linux including counter strike. [13:27] a lot of the games I have I picked up for free at some giveaways. those gleams or whatever [13:27] join this group, check our youtube, etc. [13:27] I don't have much in the way of AAA [13:28] ubuntu218: yes, that will still take some time for AAAs [13:29] I still think a lot of my problems stem from me being new and having no idea what I'm doing. [13:29] I actually broke my OS trying to figure it out a few times. :c [13:30] for games look at wine and POL. for the rest look/ask if there is a linux client or an alternative program [13:30] I couldn't figure out how to extend my partition from the left side, so I decided to delete it and remake it. [13:30] POL? [13:30] play on linux [13:30] then after I tried to remake it, grub rescue [13:30] that may be true, but often thats how people learn. Maybe the jump to 16.04 it to much right now for you because it is hard to determine what is bugs and what needs configuration [13:31] and I panicked before I realized I have a laptop to google what to do [13:31] I know I shouldn't go 16.04 until it's at a public full release. [13:31] I'm not nearly good enough at this to take a WIP item and work around it [13:32] i wouldent even look at it like that [13:34] The fact that I'm bad at coding doesn't help. But I think that with a bit more experience, it would just be a better OS for me. [13:34] I like the fact that I could leave it unlocked and I could come back without a child in the house having come in and downloaded who knows what on it [13:34] Someone in the house is known to download iorbits and advance system care programs [13:35] i am in no way skilled but i do like to troubleshoot that has helped me learn quickly. [13:36] for those who don't use windows, advance system care claims to "optimize your settings" as though microsoft wouldn't have largely done that by default [13:36] ubuntu218: in general: learning by doing. just use ubuntu and get to learn it [13:37] That seems right. Just playing with it and trying things until they start to click [13:38] I also look up a lot of help videos. To sort of hold my hand until I can do things a bit [13:39] I found that with command line and the whole linux world reading and hands on (just throw yourself in the problem) worked for me. You need to try and retain as much info as you can while proceeding. Rather than just going through the steps to get from point a to c without remembering or understanding what you did to get there. [13:40] I was able to remember some of the commands [13:40] (not saying you do that just speaking on what i learned) [13:40] sudo apt-get install $program [13:40] and some of the little basics [13:41] yep thats where i started. [13:41] But I think that only works for things that are in the repository [13:41] whenever i use the command line it is so efficient and just x100 [13:42] I can't figure out how to get the GUI to do much besides look at files to find the path for the command line [13:42] so I love when i remeber what I need to do and I can do it all in command line. but when I get stuck on something that I need to go searching for an app or gui im like damn. I wish I was fluent [13:43] hi guys. trying to install kubuntu on virtualbox and cannot proceed with installation [13:43] But using a terminal does make me feel pretty "tv hacker" [13:43] when you say the gui what do you mean? your overall desktop experience? [13:43] i can select language, wireless is detected and i can browse with firefox, but on "prepare" step i have "continue" button greyed out [13:44] I mean if it uses the mouse to launch anything not in the taskbar on the left. === Guest60487 is now known as cowst [13:46] hmm i havent checked out any other enviroments other than the stock unity [13:46] what do you mean on prepare? once you are already installed and loaded? [13:47] I sort of find the number of distros to be a problem with linux. [13:47] Because some things work on some, don't work on others [13:47] ubuntu218: there are way too many "distros" that only change a few programs and a wallpaper. that is right [13:47] definietly a big headache when you are first starting off. i agree with you there. I wanted to tryeverything so quick and kept switching wanst sure what the hell i was doing [13:48] thats why i stock with Ubuntu and Fedora [13:48] At that point, I hardly see how you can call it a new distro if you only change next to nothing [13:48] stuck* [13:48] I've only used ubuntu, because it's apparently a really common distro. [13:48] shockvalu: i load live kubuntu, then select to install it [13:49] in the installer i select language, it detects wireless apparently, then i cannot click on continue when presenting checks for 3rd party SW and updates [13:49] cowst : and at the end of the installation it gets stuck? at preparing and initializing? [13:49] nono [13:50] oh oh [13:50] it does not start installing [13:50] i am still on the 3rd step, right after language and wireless [13:50] try installing by unchecking "automatically install updates during setup" [13:51] i tried every combination of those 2 checks but continue stays gray [13:51] and see if you can move forward. than do the updates after the installation that is usually a problem I have but thats because my network is set up wierd [13:52] i recreated the virtualbox image from scratch leaving all defaults, let's see [13:52] is this a laptop? [13:52] I'm going to reboot into the linux side and then I'll be back [13:52] yes, laptop [13:52] 100% battery and attached to power [13:52] :) [13:52] cowst, like I said in kubuntu, you may not be connected to the internet in the vb , only the router, hence not able to continue [13:52] i have had similar problems but than again not on xx [13:52] but if i can reach google from firefox in live kubuntu i am connected [13:53] so these could be bugs you never know, either way when i ran in to similar situations I would just diable networking for the time being and than install directly not boot to live first [13:54] should there be a prompt during boot? when i start it goes straight to live [13:54] When I boot to live desktop and than install, i feel my laptop doesnt have enough power to run everything so i run in to random problems. [13:54] is there a way to make apt in 16.04 accept weaker signature hashes on apt repositories? [13:55] one should say install and one should say boot to live [13:55] because i know someone who's getting issues with non-SHA-2 hashes on a repository [13:55] teward, ignore the warning if you're using the landing ppa [13:55] And, I'm back. [13:55] bonjour [13:56] Ubuntu has such a fast bootstrap system [13:56] BluesKaj: not for the landing PPA - it's for other repositories [13:56] i'm kind of being forced to proxy data between a user and support and I hate it [13:56] teward, run update agian to make sure it's still there [13:57] shockvalu: ok, i pressed some arrow during boot and it shows the selection menu [13:57] perfect [13:57] not quite [13:57] :D [13:57] there is no install [13:58] teward: its just a warning [13:58] i can start live kubuntu, which has the install procedure [13:58] but there is no install directly [13:58] teward: its not blocking, its just a warning. its just that "apt update && ... "doesnt work since its an errorcode [13:58] i think i saw this disappear in distros recently [13:58] k1l: that breaks Chrome then, it seems [13:59] teward: it breaks everything than the original repos, yeah. but they still work. just run sudo apt update ; sudo apt full-upgrade [13:59] sorry that i have been able to barely help you I just run the stock original with unity [13:59] but i filed a bug on google repos and they said they will fix that for the next chrome release [14:00] So, how can I know if updating my ubuntu number is a good/bad idea? Because I can't downgrade if it doesn't work [14:01] MoonBurst: if in doubt dont use unstable releases :) [14:02] Well, yeah. But like... how can I know if a 14.04 should/shouldn't go into a 15.10 [14:03] what do you mean? [14:03] MoonBurst: its quite pointless to upgrade now to a 15.10. that is too much effort and the 14.04 to 16.04 upgrade will work in july [14:03] Are you asking how do you know if you shoul dstay at 14 move up to 15 or 16? [14:04] upgrading from 14.04 to 15.10 now involves EOL upgrade to 14.10 and 15.04 and then upgrade to 15.10. which is not worth the trouble imho [14:04] Well, since some programs won't work in different versions [14:05] k1l: stupid question, I run my own repository, is there a way in reprepro to define what hash function to use? [14:05] shockvalu: https://dl.dropboxusercontent.com/u/10973298/kubuntu.jpg [14:05] stuck there [14:05] do you recognize that red icon on top? [14:06] i assumed that has something to do with battery, but cannot confirm or know what to do about [14:06] 14.04.04 is LTS (Longer Term Support) its best to stick with the LTS releases when in doubt or to many problems [14:06] teward: hmm, dont know, sorry. might ask in #ubuntu-packaging [14:07] Let me ask you something, other than the versions numbers being higher 15/16 what is the reason from straying away from the LTS release may i ask? [14:07] That's kind of what I thought, sticking to lts because I'm new to Linux as a whole. [14:07] Mostly just that some of my programs don't have a working version for 14.04 and saying they have working support for others [14:08] The dolphin emulator says it works for 15.04 [14:08] But I don't want to upgrade and risk having 12 other things break down [14:09] it seems as if it is checkmarked to say that it is plugged in but showing red as if the battery is almost empty. Thats all i take from it. could be a bug [14:09] i think it is a bug at some level, perhaps virtualbox, who knows [14:09] yep [14:09] one thing to blame the installer is definitely no written hint on the problem [14:11] if it would tell me to charge my battery over 50% or plug the power, then i would know battery level recognition is broken at some level :) [14:11] k1l: thanks [14:14] cowst, think thei installer itself can't see the internet thru the VB...try an ethernet connection instead, if possible [14:16] oh, did you download a netinstall image? [14:16] did even catch that [14:18] is it possible to widen the launcher so that there is 2 rows? [14:26] I woud love the ability to be able to tailor the launcher and organize apps with spaces. I really have not played around to much with customizing. Anyone know if this is possible? [14:30] shockvalu, right click on the launcher, there shouldbe a launcher settings option in the dialog [14:31] oh jeez, thanks. brb [14:38] When trying to run an application with optirun I get that: "[ERROR]Cannot access secondary GPU - error: [XORG] (EE)" but nothing more. The Xorg error message seems to be cut off and there is nothing mentioning "secondary" in /var/log/Xorg.* [14:44] Rovanion: tell us the ful story please, 16.04 upgrade or fresh install? grafix chipset? driver version? [14:46] lotuspsychje: Upgrade, Nvidia GF119M [NVS 4200M], nvidia-340-updates (since 361 is known to break bumblebee). [14:47] Rovanion: upgrade from 15.10 or 14.04? [14:47] This is the output I get: http://paste.ubuntu.com/15496502/ . Upgrade from 14.04 [14:47] Rovanion: thats not very recommended mate [14:47] Rovanion: i suggest you try a clean install 16.04 with cable + updates enabled during setup [14:48] lotuspsychje: Becuase users shouldn't upgrade between LTS's? [14:48] Rovanion: if your card is optimus also install nvidia-prime, not bumblebee (is outdated) [14:48] Rovanion: until 16.04 is final, its not recommended [14:49] lotuspsychje: But how are you to beta test that upgrade path then? [14:49] Rovanion: from 15.10 its possible, but not yet from 14.04 [14:50] Rovanion: but again, until final releases i would not reccomend the upgrade way [14:51] But then again: How do you beta test the upgrade path? How do you make sure that there are no kinks when it's released to your average grandpa? [14:52] Rovanion: beta testing should be installed fresh, or use the daily 16.04 [14:53] lotuspsychje: Also, it seems like nvidia-prime can't be activated on a per-app basis. [14:53] lotuspsychje: You aren't answering the question. Just restating the same statement over and over again. [14:55] Rovanion: look mate, your trying things the wrong way... [14:55] Rovanion: if you wanna help test 16.04 please dont upgrade from 14.04 yet === littlebunnyfufu is now known as SonikkuAmerica [14:57] So I take it you're not beta testing the 14.04->16.04 upgrade path before release? [14:57] Rovanion: no [14:58] im on daily 16.04 fresh [14:58] lotuspsychje: Any reason behind this policy? Because it is Canonical policy? [14:59] Rovanion: http://news.softpedia.com/news/ubuntu-16-04-lts-xenial-xerus-final-beta-is-now-available-for-download-502151.shtml [15:00] Rovanion: LTS to LTS jump is perfectly possible on final release [15:01] Rovanion: also sudo do-release-upgrade -d from 15.10 worked, but some issues arrised also [15:01] lotuspsychje: Or at some later point during the beta testing phase, doesn't say anything on that point. [15:03] lotuspsychje: It merely states that at the time of the article there was a bug preventing the upgrade. But I'm upgraded as far as I can see. Some hiccups along the way but nothing unsolvable. [15:03] Rovanion: and you here with issues now... [15:04] uname -a please? [15:04] lotuspsychje: Bumblebee has always meant issues. [15:04] Rovanion: bumblebee is outdated... [15:04] Linux ThinkT420s.rovanion.se 4.4.0-15-generic #31-Ubuntu SMP Fri Mar 18 19:06:37 UTC 2016 i686 i686 i686 GNU/Linux [15:04] !info nvidia-prime [15:04] nvidia-prime (source: nvidia-prime): Tools to enable NVIDIA's Prime. In component main, is optional. Version 0.8.2 (xenial), package size 10 kB, installed size 66 kB (Only available for i386; amd64) [15:06] !info bumblebee [15:06] bumblebee (source: bumblebee): NVIDIA Optimus support for Linux. In component universe, is extra. Version 3.2.1-10 (xenial), package size 47 kB, installed size 180 kB (Only available for linux-any) [15:07] dont use bumblebee :/ [15:56] Can't seem to get nvidia-prime to work either. Nvidia-settings says that prime is not supported. Running driver nvidia-361-updates now. nvidia-prime is installed. Running prime-supported exits with 0 but no output. Starting an FreeGLUT app results in: OpenGL GLX extension not supported by display ':0'. Any ideas? [15:59] hybrid graphics? [15:59] Yup. [15:59] which gpus? [16:00] Purging xserver-xorg and installing it again got me back on my feet. [16:01] BluesKaj: Nvidia NVS 4200M and the graphics chip in my 2nd generation i7. [16:02] Right, and now nvidia-settings got the hang of the setup. [16:07] Right. And I have to restart X.org every time I change profiles with prime? Or can I start a Xephyr running on the discrete card with prime-offload? [16:07] Doesn't seem like prime-offload has a manpage. [16:19] Seems like I'm unable to switch back to the nvidia chip though.. [16:38] Hi! I am running 16.04 (Development Branch). My system doesn't connect to one particular WiFi Network. Is this a 'known issue' or should I 'report as a bug'? === littlebunnyfufu is now known as SusaNeedsABetter === SusaNeedsABetter is now known as SonikkuAmerica [16:39] fossterer, does it connect to any other wifi network ? [16:39] Yes. My home Network works fine [16:39] My University network alone fails [16:41] fossterer, U networks are nototoriuos for blocking anything thats detected as slightly different , make sure you edit your account with the U to accept 16.04 , however that is done. [16:42] BluesKaj, Are you saying that the University Network admins have to be told about this? [16:43] I never had to do it starting from 12.04 to 15.10 [16:45] is this the place for Xenial problems? [16:45] fossterer, do the admins allow development OSs on their networks? [16:45] Bluskaj, There's no such policy on preventing any OS. [16:47] I am getting the following msg: Failed to fetch http://us.archive.ubuntu.com/ubuntu/dists/xenial/InRelease The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 40976EAF437D05B5 NO_PUBKEY 3B4FE6ACC0B21F32 [16:48] any help appreciated, I have redone /etc/apt/sources.list already [16:48] jiohdi, just ignore it, just a warning..being worked on [16:49] I have more than one machine and this one getting that msg is not finding any packages to upgrade while the others do [16:51] jiohdi, some packages are HW dependent so there will be differences [16:52] from dozens of packages on one to zero on the other seems suspicious [16:54] jiohdi, change mirrors then , see what happens [16:55] k [16:55] mine is definitely a much shorter error/warning [16:57] BluesKaj, that did the trick, much thanx [17:02] anyone know a better irc client than xchat-gnome [17:02] 6irssi :) [17:03] jiohdi, hexchat [17:04] jiohdi: xchat-gnome is the most reduced irc-only client you could use. better try hexchat [17:08] believe xchat is no longer supported or being developed further [17:08] futher developed even :-) [17:08] HexChat is pretty good. [17:09] hexchat is the fork with new life :) [17:11] xchat is significantly worse since last I used it [17:11] jiohdi: please dont compare to the -gnome version you used. [17:18] hey. in the latest Kubuntu, when my pc hibernates, i cant seem to turn it back on unless i disconnect the power supply and press power to remove all electricity then replug and turn it on. any ideas? === SuperLag_ is now known as SuperLag [17:22] Oderus, I don't hibernate as such, just suspend session, since it's basically the same. sudo pm-suspend [17:26] okay, i will just turn hibernate off for now [17:31] Oderus, it should be fixed by the official release date...I hope :-) [17:31] BluesKaj: Thank you :) [17:33] BBL [17:34] Hi, I tried installing 16.04 on Bay Trail tablet, and it failed on grub installation [17:34] Anyone got a clue why is that? [17:36] I downloaded amd64 build, and run that with a bootia32.efi file placed in /efi/boot on pendrive [17:38] 32bit efi is pita [17:39] Care to elaborate? I bought this tablet for less than a RPI3 anyways, so I'm treating it carelessly [17:50] Hi! I am running 16.04 (Development Branch). My system doesn't connect to one particular WiFi Network. Is this a 'known issue' or should I 'report as a bug'? [17:50] BluesKaj, I am going to file a bug on this issue. [17:58] some wifi cards are known to make trouble no matter what ubuntu version you use [18:01] whoever told me about hexchat, thanx, much better than xchat... but strangely like xchat once was [18:04] jiohdi: it's a fork as k1l said :) [18:04] jiohdi: its the "new" xchat. [18:05] jiohdi: but the problem was the -gnome version. gnome cuts down programs a lot to meet their rules of what a user should be able to change etc. [18:05] the inability to mark away with ctrl+a and the resize of window when searching annoys me a bit - but other than that it works fine [18:06] so I've been trying to resize my boot partition on a disk with an lvm volume (containing root and swap partitions) running 16.04 and seem to have gotten myself stuck [18:06] I can't seem to move the parition with the lvm volume towards the end of the disk, so I can't enlarge the boot partition [18:07] is there a way I can deactivate the lvm volume group such that I can move the partition with gparted? [18:09] Anyone install the latest Joomla on Ubuntu? I am getting an error that The GD extension for PHP is not available. [18:12] I can inactivate the volume group with vgchange, but it gets reactivated when I refresh the devices with gparted [18:12] Annnd I install php7.0-gd and it works fine. bah [18:14] and "Deactivate" from gparted doesn't seem to do anything :( [18:20] slyrus: yes, gparted causes that during a rescan after every change. Best advice is avoid GUI tools and use the command-line [18:20] Ok, thanks TJ-. [18:21] slyrus: how small is the /boot/ partition? [18:21] Contemplating vgextend/pvmove to move the volume to another disk rather than trying to move the partition by hand. [18:22] I figure I should learn how to do this stuff eventually.... [18:22] slyrus: you could simply add the small partition as a PV to the VG, and then create an LV for /boot/ :) [18:23] oh, that does sound simpler [18:23] slyrus: grub-install will figure that out and include its lvm driver in its core.img so it can activate the VG to get to its modules and menu file [18:23] make sure you you do a grub-install and update-grub before doing a reboot, if you make changes, of course [18:25] slyrus: copy the existing /boot/ into a new LV for boot, then you can repurpose the current /boot/ partition as a PV to recover the space [18:26] If I had read that before hitting return on the pvmove that's what I would have done :) [18:27] well I thought it was so obvious there was no need to mention it! [18:27] heh [18:29] Wll guys WTF [18:30] speaking of stupid questions, can I just add an existing partition as a PV and have an LV automagically created with the existing data? [18:30] I tried to install 16.04 on bay trail tablet, and update-grub returns "failed to get canonical path of '/cow' [18:31] duobix: well, maybe the cow wandered off :) [18:32] duobix: the error sounds like the update-grub is NOT running inside the /target/ chroot (since the copy-on-write /cow is for the installer's in-RAM live environment) [18:32] duobix: are you trying to do something manually, or is that error directly from the ubiquity/debian-installer task? [18:33] I'm trying to manually get a 32bit grub, cause bay trail requires that [18:33] duobix: as in grub-efi-ia32 ? [18:33] 32bit uefi, 64bit cpu, just because [18:34] yessir, exactly that [18:34] duobix: right, so you need to set up the /target/ chroot correctly first to do that [18:34] duobix: can you confirm the /target/ directory contains the fresh 16.04 installation? [18:34] hi, i have a 4 bay system and i installed ubuntu 16.04 on it. it is an UEFI system with GPT tables. when i remove the first drive and reboot, i get dropped to a grub rescue shell. i followed this guide: https://www.thomas-krenn.com/de/wiki/Ubuntu_Software_RAID_mit_redundanten_UEFI_Boot_Eintr%C3%A4gen [18:34] can someone tell me whats wrong? or another tutorial/tips? [18:35] The problem is, that the 16.04 won't install, because 32bit grub isn't there [18:35] tete_: you've pulled the drive that contains GRUB's /boot/ file-system where the kernels, GRUB modules, live [18:35] oh btw its a mdraid [18:35] And I don't know how to prepare ubuntu for the task [18:35] TJ-, sorry, forgot to mention it is a raid1 for /boot and raid6 for / [18:35] duobix: what stage are you at right now? the installer has completed but you're trying to manually trying to install grub-efi-ia32 ? [18:36] tete_: right, so you've pulled one half of the raid1 mirror for /boot and presumably GRUB core.img therefore cannot activate the degraded raid [18:37] tete_: at the "rescue>" prompt what does "ls" report ? [18:37] Well, I got into live-mode, and tried to install several times, but it failed because of the grub2 not being there. I don't exactly know how exactly to apply 32bit grub 2, so I figured out apt-geting it would be a good idea [18:37] and then applying it. [18:38] duobix: let's be VERY accurate here. "it failed" ... you mean "the system failed to boot after a reboot"? [18:39] No. System failed to install 32 bit grub, and then it stopped the installation [18:39] duobix: ahhh, OK, so at that point you're still in the live installer environment, and the /target/ install has been completed (GRUB is installed last) [18:40] duobix: so, in that state, you CAN set-up the chroot at /target/ enter it and install the version of GRUB you want corretly [18:40] Yes, Indeed [18:41] so the /target/ install is where I put the grub, right? [18:41] duobix: if you're at that stage now then firstly do "mount | target" and ensure the 16.04 install is still mounted at /target/ [18:41] duobix: if it isn't then we have to find it and mount it first [18:41] TJ-, strange, it seems now if it would be in BIOS mode instead of EFI.. just plugged back the drive and booted, will check first whats going on now [18:41] iirc ls showed me my 3 devices [18:42] duobix: you'd expect to see something like "/dev/sda5 on /target" if it is mounted [18:43] tete_: if you've got GRUB's root on RAID1 then you'd expect to see a (md0,msdos1) or similar [18:43] "mount | target" shows me exactly nothing, I suppose I haven't figured out how to interpret "target" [18:44] duobix: that's fine, the installer probably unmounted it. so, first we need to find it :) [18:44] in /dev/ I have no sda folders, strange [18:44] ok [18:44] TJ-, i think something is broken now because of the update i just did before asking here ;) i will follow the tutorial of https://www.thomas-krenn.com/de/wiki/Ubuntu_Software_RAID_mit_redundanten_UEFI_Boot_Eintr%C3%A4gen again and see if i get back to the grub rescue and then ask again [18:44] duobix: so, firstly install a helper program so we can paste command results. "sudo apt-get install pastebinit" [18:45] installed [18:45] tete_: good luck ... keep in mind GRUB itself can't handle writeable RAID, so it in fact had to be installed to the underlying devices in the mirror [18:45] duobix: OK, now do "pastebinit <( sudo blkid )" and tell us the URL [18:46] tete_: as in, if md0 is /dev/sda /dev/sdb then you need to "grub-install /dev/sda && grub-install /dev/sdb" [18:46] http://paste.ubuntu.com/15500537/ [18:46] TJ-, jop, i just got response from grub-install: http://pastebin.com/YNEtCTrA [18:46] hope this is not the reason for my problem (the warnings) [18:47] duobix: what kind of device are you installing on (make/model) ? [18:47] I told you, it's a bay trail tablet xD mmcblk0p3 is where I installed the root directory ( I think) [18:48] duobix: is it a no-name brand? there may be some solutions known for it I was going to search for [18:48] bbl, gotta eat [18:49] Dude, it's a lark ultimate 7i WIN tablet. It's a chinese rebranded polish tablet. [18:49] duobix: OK, so the /dev/sda is the USB ISO installer? [18:50] duobix: and it looks like /dev/mmcblk0p3 is likely the installed Linux file-system. "sudo mount /dev/mmcblk0p3 /target" should therefore mount it [18:50] duobix: after which "ls /target/" should show something similar to "bin boot core dev etc home initrd.img lib lib32 lib64 libx32 lost+found media mnt opt proc root run sbin srv sys tmp usr var vmlinuz" [18:51] mount point target does not exists, it says [18:52] duobix: hmm, i'd have expected it to be kept. However, do "sudo mkdir /target" and then re-run the previous mount command [18:53] anyone know the file location for the splash screen that appears after login in kubuntu xenial? looking to theme [18:56] TJ-, no idea what i did, just followed the guide again and now it booted without the first drive, thanks anyway :) [18:57] tete_: phew :) [19:01] Ok, target shows exactly how you described it [19:03] duobix: Great :) let's get it configured then: "for N in proc sys dev dev/pts etc/resolv.conf; do sudo mount --bind /$N /target/$N; done" [19:05] ok, I did input that [19:05] nothing on console, I suppose it worked? [19:05] duobix: OK, after this next command you will be effectively running that 16.04 system. To get back from it to the current live installer environment you would do "exit". To enter the target now, do "sudo chroot /target" [19:05] Oderus, unfortunately I haven't found any themes that are installable to replace that(IMO) ugly splash screen [19:06] ok, now I'm running as root. [19:07] Oderus, Kubuntu Xenial here too. [19:07] duobix: right, now let's check that 16.04 install has network access: "ping -nc 5 www.ubuntu.com" [19:07] I pinged my own site, seems to work [19:07] no packet loss [19:08] duobix: right. Now we need to mount the tablet's EFI system partition. do "ls /boot/efi/" you should see an empty directory just now [19:08] bbiab, other stuff to do for 20 mins or so [19:08] I confirm, it's empty [19:09] duobix: now let's try to mount the EFI system partition there: "mount -a" and then "ls /boot/efi/" again... should see some directories [19:09] yeah, Boot, bootmgr, BOOTNXT, and EFI [19:10] duobix: great, going very well :) [19:10] duobix: now ensure we have paste ability: "apt-get install pastebinit" [19:11] duobix: once that's confirmed as installed you can show me the result of "pastebinit <( dpkg -l 'grub*' )" which lists all the GRUB packages [19:12] Had to run apt-get update, as pastebinit wasn't found, ok, now the link [19:13] duobix: ahhh, good one, usually that's been done by the installer [19:13] http://paste.ubuntu.com/15500996 [19:14] I think I'll need grub-efi-ia32 [19:15] ahhh; that shows is the Legacy BIOS GRUB packages is currently installed "ii grub-pc" [19:15] correct [19:15] duobix: so do "apt-get purge grub-pc" then "apt-get install grub-efi-ia32" [19:17] ok, It finishe [19:17] *finished [19:18] duobix: assuming the install doesn't fail then you may need to follow up with "update-grub" to be sure, and at that point it should be ready. If so, do "exit" to return to the installer live environment, then "sudo shutdown -r now" to reboot now [19:18] duobix: so "update-grub" [19:18] duobix: then "exit" and "sudo shutdown -r now" and test :) [19:19] shouldn't I remove the pendrive? [19:19] ohh [19:19] It seems like it's booting? [19:19] once the shutdown has completed but the default boot entry should be Ubuntu [19:20] it seems it booted [19:20] it depends if the system firmware puts removable devices in front of the default boot entry or not [19:20] yay, it's alive. GIANT THANK YOU [19:20] duobix: Fantastic :) that was easy :D [19:21] And btw thanks for explanations along the way [19:21] you're welcome; it's good to learn what you're doing as you go [19:22] now I gotta figure out why wifi is not working on this thing, I kinda have no clue what interface that could be on [19:22] is there something like lsusb for pcie devices? [19:23] duobix: start off with "lspci -nn" for PCI devices, or "lsusb" for USB. also, "iwconfig" will list any existing wifi interfaces [19:23] duobix: you can also list all network hardware with "lshw -C network" [19:26] Well I used ifconfig to get interfaces before, and saw no wireless device in there... also usb ain't showing a thing, PCI also [19:26] duobix: can you do "pastebinit <( lspci -nn; lsusb; lshw -C network )" so I can review the output? [19:26] yeah, give me a sec === alkisg is now known as alkisg_away [19:28] http://paste.ubuntu.com/15501223/ [19:28] also do mind that the one usb device listed as wifi is my usb stick, And I kind of wanted the internal one to work [19:28] but I got no idea what it could be xD [19:28] So its on the USB interface, at bus location 001:003 [19:29] ahhh, OK, so that's not it [19:29] ok, let's dig deeper: "dmesg | pastebinit" [19:30] http://paste.ubuntu.com/15501249 === alkisg_away is now known as alkisg [19:31] line 748... shows a kernel bug caused by a problem with the baytrail GPIO (general purpose Input Output) controller [19:32] I kinda suspected this; because the kernel cannot activate the GPIO controller a bunch of devices cannot be accessed [19:33] aaand as this is a chinese-almost-no name device, this could have no driver whatsoever, right? [19:34] No, the driver is there (called pinctrl for "pin control") but its not configured correctly for the tablet [19:34] Ohh, I see [19:35] Is there a recomended RTFM for me about that? [19:35] we may be able to get it to configure correctly however. At line 444 there's another kernel bug that indicates Linux is possibly not being told the correct config info from the ACPI interface of the tablet [19:36] line 285 indicates a bug in the tablet's firmware which could have knock-on effects like this [19:37] duobix: can you confirm this file exists: "ls /sys/firmware/acpi/tables/DSDT" [19:38] the file exists === alkisg is now known as alkisg_away [19:39] duobix: OK, let's grab some info from it: "pastebinit <( sudo strings /sys/firmware/acpi/tables/DSDT | grep -i windows )" [19:39] Tried to cat the output, looks like 1. binary 2. Garbage 3. Binary Garbage [19:40] *output=what was in dat file [19:40] duobix: what I'm looking for is the textual names of the windows versions the ACPI recognises; if there are some, we can get Linux to pretend to be one [19:40] Ok, sec [19:41] http://paste.ubuntu.com/15501392/ [19:42] windows 2013 - thaaat's kind of funny xD [19:43] right, so what we'll try is adding an option to the kernel command-line that gets it to report itself as a Windows version [19:44] Ok xD [19:44] duobix: " sudo sed -i 's/^\(GRUB_CMDLINE_LINUX=.*\)"$/\1 \\"acpi_osi=Windows 2013\\""/' /etc/default/grub " [19:44] duobix: then check that was added with "grep acpi_osi /etc/default/grub" and you should see something ending: ... \"acpi_osi=Windows 2013\"" [19:47] Yes, I see GRUB_CMDLINE_LINUX=" \"acpi_osi=Windows 2013\"" [19:48] duobix: OK, do a reboot. once it's restarted confirm that setting is in the kernel command-line with "cat /proc/cmdline" [19:48] duobix: then let me see the "dmesg | pastebinit" again so I can tell if it has helped, or made things worse :) [19:49] duobix: Arggh!! hang on, I forgot the important but [19:49] duobix: you need to do "sudo update-grub" to add that setting to GRUB's /boot/grub/grub.cfg :D :D [19:49] duobix: sorry - tired here [19:50] Yeah, restarted and it returned to normal xD So ok, one more time [19:50] good thing bash got the comands remembered. [19:51] yes, in the $HOME/.bash_history [19:52] duobix: Are you the first person to put Ubuntu/Linux on that tablet? I can't find any other indications of attempts [19:53] I think I'm one of the very first, some tried, but not everything worked [19:53] on my machine touch and sound doesn't work also [19:53] which will also be hanging off the GPIO too [19:53] so, we fix this, you should get everything to show up [19:54] ok, ran update-grub. Reboot? [19:54] the kernel trace at line 444 does show in the stack trace that the kernel is reading the ACPI instructions, which tell it where the GPIO ports are, it tries to make a GPIO request, and that times out and fails. [19:55] duobix: yes [19:55] duobix: I'm *hoping* that by having Linux pretend to be Windows, ACPI will give it better information about the GPIO side [19:57] http://paste.ubuntu.com/15501597 [20:01] unfortunately, same failures [20:02] now, try adding " pci=nocrs" to that line in /etc/default/grub we edited last time [20:02] do you have a handy text editor to do that (use gksudo if using GUI) [20:03] I use mostly nano on linux, and vi-like editor on my WinPC (i need that for work) [20:03] add to the line GRUB_CMDLINE_LINUX="... pci-nocrs" and then do the "sudo update-grub" and the reboot again, and repeat the "dmesg | pastebinit" [20:03] OK, you know your stuff then, I don't need to tell you how :) [20:04] "pci-nocrs" will tell the kernel to ignore some faulty/missing info in the ACPI tables, and let the kernel figure it out itself. That may help (a bit) [20:04] Typos! [20:04] it should be "pci=nocrs" !! [20:05] so, how should I add it, ....2013\"" "pci=nocrs" ? [20:06] or within the quotes ? [20:08] Ok, I went with the first option. [20:09] I tried updating grub, but I got: 12: /etc/default/grub: pci=nocrs: not found [20:10] sorry, was reading some kernel info about this. pci=nocrs should be inside the enclosing double quotation marks, so the line should end ...2013\" pci=nocrs" [20:11] the reason for those earlier embedded \" is due to the space in the 'Windows 2013' string. Without the enclosing double quotes on the kernel command-line it would read it as two options: 'acpi_osi=Windows' and '2013' [20:11] ok :) [20:12] and as you're editing a shell script that uses " itself we have to escape them with \" to include them! [20:12] I figured as much, that's why I asked [20:12] OK .. many folks don't realise the importance [20:13] Yet I'm always amazed by some shit-solutions on Windows that give me cancer [20:13] So that's why I also asked, as I know from experiance that cancer can be given easily xD [20:14] Yay, grub updated! Reboot. [20:15] http://paste.ubuntu.com/15501841/ [20:18] no better; best to remove that option since it doesn't help [20:19] so in summary; there are several bugs in the tablet's ACPI implemention, and the GPIO configuration is bad in some way [20:20] You could try using kernel 4.5.0 - there may be improvments [20:20] Yeah, I might do that [20:21] But really I don't know how to get the newer kernel [20:21] BTW there's a funny thing about this particular tablet [20:22] ubuntu kernel teams builds the mainline kernels into packages: for 4.5.0 it's at http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.5-wily/ [20:22] It's bios has a kind of overlay where mouse works [20:23] So I suppose I just download one of them and dpkg them? [20:23] GUI interface in the UEFI setup? yes, that is often done for 'user friendlyness' [20:24] no, GUI is there only with like 6 icons, the rest is normal bios/other programs [20:24] xD [20:24] I noticed the tablet is currently booting a .signed kernel, indicating Secure Boot is enabled. Those mainline builds don't get signed so you'd need to ensure you can disable Secure Boot and still have the tablet boot [20:24] I actually did disable secure boot [20:24] ahhh, but the signed packages are installed, OK, that's good [20:25] the tablet is using the amd64 build isn't it? [20:25] yes it is [20:25] The Atom CPUs are 64bit [20:25] but the uefi is 32bit, and windows are shipped 32bit [20:25] then you'd want linux-image-4.5.0-040500-generic_4.5.0-040500.201603140130_amd64.deb and linux-headers-4.5.0-040500-generic_4.5.0-040500.201603140130_amd64.deb [20:26] oh, and linux-headers-4.5.0-040500_4.5.0-040500.201603140130_all.deb [20:27] best way is to download them all to a sub-dir, and then do "sudo dpkg -Ri path/to/subdir" and it will install them all at once [20:27] I have to leave for a while now; hopefully you can make progress on that tablet [20:27] Thanks for help anyways, much appreciated [20:28] oh wow, it hanged now xD [20:53] duobix: when it's running again can you do "lsmod | grep gpio" - let's find out which driver is loaded [20:53] The new kernel is there, so let's find out [20:54] gpio_keys 20480 0 [20:54] rfkill_gpio 16384 0 [20:54] ok, and now "lsmod | grep pnctrl" [20:56] nothing. [20:59] OK, pinctrl_baytrail is built-in to the kernel image [21:00] I'm working up the kernel bug stack trace from line 452 to line 446 looking at what each function does/expects [21:00] from http://paste.ubuntu.com/15501841/ [21:00] http://paste.ubuntu.com/15502217/ [21:00] here you go, from 4.5 [21:08] what *may* be promising here is that the stack trace seems to indicate the failure is due to the code in the ACPI DSDT of the tablet, and Linux has a mechanism for loading a replacement DSDT (which means potential you can dissasemble the current DSDT, identify and fix the bug, and use the fixed version) [21:11] Mhm, so, where do I begin? [21:16] Ok, I'm reading https://wiki.archlinux.org/index.php/DSDT#Find_a_fixed_DSDT [21:17] here's what to do to gather what we need: [21:18] duobix: "mkdir /tmp/acpi && pushd /tmp/acpi" then "sudo acpidump -b" (dumps the binary tables into the directory) then to dissasemble the DSDT do "iasl -d dsdt.dat" and you can send me the disassembly with "pastebinit dsdt.dsl" [21:19] k, gimme a sec, gotta ssh into that tablet [21:19] OK. here's a link to some intense ACPI analyse work I did back in 2007. http://tjworld.net/snc/sourcecode.html and http://tjworld.net/snc/ for the overview of that actual project [21:21] I don't have acpidump, hmm [21:22] "sudo apt-get install acpica-tools" [21:23] http://paste.ubuntu.com/15502634/ [21:24] right... give me a few minutes to read it [21:25] K chill, I gotta ready your acpi analyse work xD [21:27] as an aside, if you search your dsdt.dsl for "Windows 2013" you'll see the method that tries to match the OS name and the fact it sets a different value to OSYS depending. You'll also notice the later versions of Windows set higher values, indicating more functions provided. You'll also notice at the start of the method the lowest value is set as a default "OSYS = 0x07D0" which is what happens when the OS [21:27] is Linux. [21:28] If you search for OSYS in the rest of the file you'll see conditional code that only does some functions if that value is greater than that default value, indicating functionality that is not enabled with lower values of OSYS [21:36] "OperationRegion (GPOP" will get you to the GPIO definitions [21:37] yeah I see [21:39] ho wow, that's a lot of gpio. [21:40] if you do this you can get a feel, overview, of all the devices represented and by the indents their parent>child relationships. "grep 'Device (' dsdt.dsl " [21:42] yes, I see this kind-of tree structure [21:42] that shows the Device (SDHB) has the WLAN controllers under it [21:42] :D [21:43] so, SDIO controller B has WLAN, WLA2 and RTLW under it. ASCII text there shows it has a RTL8723 [21:43] Realtek Wireless, [21:44] it's commented as RTLW somewhere? [21:44] and right now we need GPIO to work to gain access to the SDIO [21:45] where Device (SDHB ...) method is, in its first few lines it has "Name (_DEP, Package (0x02) // _DEP: Dependencies" within which "PEPD" and "GPO2" so we need GPO2 working [21:45] duobix: yes, the RTL8723 device name is used in the ACPI and there's comments alongside for that "Device (RTLW ..." [21:47] for "Device (GPO2)" "Method (_CRS ...)" shows the resources required for it. I'm going to compare that against the kernel's dmesg [21:50] duobix: can you archive up all those ACPI .dat files and put the archive where I can download it? I need to grab info out of some of the other tables [21:51] I will asap, give me a second [21:52] Thanks [21:52] i notice in the kernel's dmesg "byt_gpio INT33FC:00: pin 100 forcibly re-configured as GPIO" - INT33FC is the GPO0 "ValleyView General Purpose Input/Output (GPIO) controller" [21:53] rockgovsky.com/acpi.tar.gz [21:58] thanks. [22:03] it looks like some the IO/memory address ranges required by these devices aren't declared by the firmware earlier; there should be entries in the BIOS-e820 table (reported at beginning of dmesg) covering them [22:06] So the device isn't telling what's where. [22:10] A lot of things about this tablet seem familiar, the partitions did sound like some kind of an android device [22:10] the stack trace tells me the ACPI code in the DSDT is being executed by the kernel (the kernel interprets and runs the byte-code in the DSDT, rather like a java Virtual machine) [22:11] and it seems the acpi makes calls to initialise the GPIO controller with incorrect/incomplete data/addresses or whatever, so it causes a time-out and that bug [22:11] you mean the mmcblk0 ? [22:12] yeah, and all the rest, It sounds very much like an android device to me [22:12] that's just because its using a multi media block device controller (MMC) and that's the standard naming convention for such devices [22:12] remember that android *is* Linux so devices will be identical in that respect [22:13] From what I remember another tablet I got was a bit different in that matter, but I gotta check it out later [22:13] Dell Venue 8 Pro [22:15] So, If the ACPI code got reduced of unnecessary stuff that might trigger that timeout, then I might have some luck, am I right? [22:18] no, it's all needed. [22:18] I *think* the root of the problem is line 269 "[ 0.248878] \_SB_.PCI0 (33DB4D5B-1FF7-401C-9657-7441C03DD766): _OSC invalid UUID" [22:20] according to dsdt.dsl line 1488 where that is defined as GUID and then used in the Method _OSC [22:20] the dmesg shows the arguments are 1, 0x1e, 0 which will be ARG1, ARG2, ARG3 in _OSC [22:25] sorry, that's not right. The method takes 4 arguments: Method (_OSC, 4, ...) [22:25] so ARG0..ARG3 of which only ARG0 ARG1 and ARG3 are used in _OSC [22:26] I'm trying to figure out what the kernel reports the invalid UUID [22:27] that's strange that they didn't use one argument that told *something* about system capabilities [22:28] right. when an ACPI method is executed that message is printed when the return code is OSC_INVALID_UUID_ERROR which is 4 [22:29] which is bit 2 in the result code [22:30] which looks supiciously like "CDW1 |= 0x04" in the Else clause of the _OSC method [22:31] so, the "If (((Arg0 == GUID) && NEXP))" fails. We can be pretty sure that ARG0 was the GUID since linux reports that UUID, so that leaves the && NEXP test. which means that NEXP must be evaluated to TRUE, no-zero. So now to trace back and find out where and how that is set [22:33] OK, and NEXP stands for "Native PCI Express" [22:34] there's memory reserved for it earlier in the DSDT but nothing in the DSDT that gives it a value, so that suggests it defaults to 0 and something else would usually set it. [22:37] That sounds like a random ecounter. [22:37] Right, I think I understand now. _OSC will do native PCI express bus management in firmware when NEXP != 0, otherwise the OS is expected to do that [22:38] so _OSC correctly bails out since NEXP==0 but it reports an error code that says UUID was wrong, when the If test actually has 2 things that can be wrong. In this case the UUID is fine, but NEXP is the test that failed. So, in summary, we can ignore that and move on [22:43] right, so that's back to the stack trace at line 427 of the 4.5.0 dmesg. [22:45] so lets see if we can enable some ACPI debug messages to get some idea of what is going on at that point. if we can pinpoint which ACPI instruction is failing that should help [22:47] duobix: can you check if debugging is enabled: "grep ACPI.*DEBUG /boot/config-$(uname -r)" [22:51] Well [22:52] CONFIG_ACPI_EC_DEBUGFS=m adn CONFIG_THINKPAD_ACPI_DEBUGFACILITIES=y [22:52] *and [22:53] those are enabled, rest is commented with #, i guess? [22:53] yes, that doesn't help, grr [22:53] http://paste.ubuntu.com/15503453/ [22:57] I'll build you a kernel with it enabled. hang on. Can you "pastebinit /boot/config-$(uname -r)" [22:58] http://paste.ubuntu.com/15503489/ [23:25] So with this kernel we'll know what is wrong? [23:32] it'll allow us to enable acpi debugging of various kinds, which should allow us to understand what data is being processed when it fails [23:33] this'll give you an overview: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/acpi/debug.txt [23:33] acpi_debug= can be set on the kernel's command-line to enable various parts [23:35] ** (appstreamcli:3037): CRITICAL **: Error while moving old database out of the way. is the message I get when running apt-get update on live session [23:36] don't know if this is because of read only issues or what (I booted with toram) Is there any way to get apt-get to write somewhere else? [23:37] anyone know the file location for the splash screen that appears after login in kubuntu xenial? looking to theme it myself, but need file location [23:42] In settings > appearance there is an option for splash screen but i don't know the name of the program that causes it to appear (whatever ksplash replacement is?) and no idea where the files are :) [23:44] ok, gotta go afk for a while [23:46] http://osdir.com/ml/ubuntu-bugs/2016-03/msg19319.html [23:46] ok looks like a xenial beta 2 issue