[01:09] having an issue when updating with a certain repo that worked a couple days ago, anyone seen this error or know how to deal with it? W: gpgv:/var/lib/apt/lists/ppa.launchpad.net_fengestad_devel_ubuntu_dists_xenial_InRelease: The repository is insufficiently signed by key B0C05875FDAB46D08017FCE24A526E70416A86EB (weak digest) [01:14] https://juliank.wordpress.com/2016/03/14/dropping-sha-1-support-in-apt/ and https://juliank.wordpress.com/2016/03/15/clarifications-and-updates-on-apt-sha1/ [01:14] its just a warning (for now). tell the repo maintainer to use a proper key to sign the packages === JanC_ is now known as JanC [01:30] k1l_: thank you === SonikkuAmerica is now known as SonikkAmerica === SonikkAmerica is now known as S-USA [03:52] lordievader: when I did a apt-get full-upgrade ubuntu's unity interface broke, there was no bar/menu thing. [04:51] apt-get update ; apt-get dist-upgrade ; apt-get install ubuntu-desktop [04:51] If there are problems, that ought to identify them. [07:04] ok thanks [07:07] you're welcome [09:14] my new desktop xenial screen: http://lotuspsychje.deviantart.com/art/Ubuntu-16-04-64bit-Development-branch-598369242 [09:21] Good morning. [09:31] lotus|xenial: nice [09:32] morgan_: thank you :p [09:32] not seen the unity dash at the bottom of the screen before [09:33] looks better I would say [09:33] morgan_: you can test yourself with gsettings tweak [09:34] I will do when I install 16.04 - i'm running kubuntu/solus at the minute [09:34] I assume there will be an option in the system-settings menu? [09:34] cool [09:34] morgan_: at this moment not yet [09:35] (btw solus is very nice/snappy - lacks packagaes and has bugs though at the min) [09:35] morgan_: still needs the gsettings tweak from terminal, but smooth already [09:37] cool - looking forward to testing Mir now also (now nvidia have a driver..) [09:38] morgan_: 16.04 will still have unity7 by default, and unity8 will be available for install also [09:38] can you install alongside ? [09:38] morgan_: yes, you will be able to choose at login then [09:39] good stuff ! [09:39] morgan_: unity8 by default at 16.10 estimate [09:39] make sense not to have mir, etc on an LTS.. [09:40] who knows on 18.04 :p [09:40] morgan_: alot of changes comming our way mate [09:40] The fact nvidia now have Vulkan/Mir/Wayland support is encouraging for the future... [09:41] morgan_: very [09:52] brb dist-upgrade reboot [09:58] when is the date of the final beta of bubuntu 16.04? [09:58] varaindemian: around 21 april [09:59] varaindemian: check the schedule in topic mate [10:00] lotus|xenial, March 24th [10:00] [10:00] Warning /!\ Final Beta Freeze, Warning /!\ Final Beta [10:00] oh [10:01] lotus|xenial, actually tomorrow? [10:01] varaindemian: final beta freeze yes [10:02] varaindemian: but final official release 21 april [10:03] lotus|xenial, so tomorrow we can expect something new [10:04] varaindemian: every update xenial does, something new happens [10:05] lotus|xenial, they say its the final beta so I expect something more interesting.. [10:05] varaindemian: are you on 16.04 right now? [10:06] nope [10:06] lotus|xenial, I am on arch and I want to change [10:06] varaindemian: every update xenial did, interesting things happened already [10:06] varaindemian: final beta freeze is just the deadline [10:07] new kernels, new package versions,etc [10:07] its already happening [10:08] I see [10:24] hey guys, i installed desktop today but find that watching content over the network with totem movie player is really laggy, any ideas? [10:25] Fudge: upgrade or clean install? [10:25] clean install mate [10:25] Fudge: up to date to latest? [10:25] ive been running server for a while, maybe two months with some issues but desktop is a bit of a mess seemingly [10:26] The following packages have been kept back: usb-modeswitch-data [10:26] 0 to upgrade, 0 to newly install, 0 to remove and 1 not to upgrade. [10:26] basically yeah [10:26] the AU mirror is three hours behind US though [10:26] Fudge: uname -a ? [10:26] Linux vtech 4.4.0-15-generic #31-Ubuntu SMP Fri Mar 18 19:08:31 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux [10:26] good :p [10:27] always wise to check m8 [10:27] Fudge: installed ther ubuntu-restricted-extras ? [10:27] I ticked the box at install but have not installed anything accept cpufreqd and espeakup, espeakup is a console screen reader [10:28] i just dont get it, even pausing my show it takes 2 seconds [10:28] Fudge: sure you got right graphics drivers active? [10:28] onboard video [10:29] Fudge: did you have this glitch at previous kernel also? [10:29] not sure which kernel i was running on trusty, think a vivid one [10:30] linux-image-4.2.0-25-generic [10:30] Fudge: check sudo lshw -C video if driver shows correctly also? [10:30] nah that works fine on this hardware, its a core 2 duo quad core machine [10:30] yeah just checking all options [10:31] product: 82G33/G31 Express Integrated Graphics Controller [10:31] driver= [10:31] VGA compatible controller: Intel Corporation 82G33/G31 Express Integrated Graphics Controller (rev 10) [10:32] how else do i check the driver? [10:32] Fudge: should show at bottom [10:32] ah k [10:32] Fudge: driver=i... [10:33] configuration: driver=i915 latency=0 [10:33] ok looking good [10:33] Fudge: you have any problems playing 1080p over youtube or vlc? [10:34] havnt even openned firefox yet, ill give it a go [10:35] actually i usually only download the lower quality since i cant see the video anyway [10:35] ill get back to you with that one :D [10:36] kk [11:01] lotuspsychje: i had to reset unity settings, see how it goes now [11:01] ok [11:02] the 1080p youtube played fine [11:03] Fudge: so lets assume its a totem issue [11:03] Fudge: can you start totem from terminal please, try that over-network- vid? [11:04] Fudge: did you install ubuntu-restricted-extras ? [11:06] yep sure did [11:09] still laggy when copied file to desktop, might re-install tomorrow see if it happens again [11:11] Fudge: you mean totem plays still laggy after playing file local? [11:11] it doesnt glitch, but if i hit space=pause it takes 3 seconds to resond [11:11] respond [11:15] Fudge: try totem from terminal, might spit out some usefull errors [11:15] nothing i could read mate, at all just that i ran it [11:15] ill re-iunstall after work tomorrow and see if it was just somethign screwy :D [11:16] kk [11:28] I am getting local fonts clumsy http://i.imgur.com/ZAEd4Gy.png [11:29] pavlushka: please dont crosspost. that is just annoying. if its a 16.04 system please only in here [11:29] yes its a 16.04 system [11:30] and ok. [11:39] In 16.04 I cannot write an .iso by right-clicking in it from naytilus anymore... was that program dropped? which UI is used now? [11:40] http://www.ubuntu.com/download/desktop/burn-a-dvd-on-ubuntu ==> I don't have that option anymore [11:41] Was that part of brasero? [12:48] hiyas all [13:52] "Setting up udev (229-3ubuntu1) ... addgroup: The group `input` already exists as a system group. Exiting. /var/lib/dpkg/info/udev.postinst: 109: [: Illegal number: * " [14:17] genii, yes I saw that in my upgrade as well, wonder why it's trying to add a group [14:25] that needs refixing then [14:26] can you guys do "getent group input" please [14:29] TJ-, input:x:107: [14:30] OK, and also "grep 'addgroup.*input' /var/lib/dpkg/info/*.postinst" [14:30] TJ-: Let me get online with that box first, hang on [14:31] !info udev xenial [14:31] udev (source: systemd): /dev/ and hotplug management daemon. In component main, is important. Version 229-2ubuntu1 (xenial), package size 972 kB, installed size 6701 kB (Only available for linux-any) [14:32] genii: how comes your version shows higher? [14:32] TJ-: input:x:106: [14:32] 229-3ubuntu1 is current [14:32] ah bot needs sync? [14:33] genii-testing: ok, the warning message about 'input' group is just that, a reminder, not fatal, but the later test failure is serious [14:34] TJ-, on the 2nd request, /var/lib/dpkg/info/udev.postinst: addgroup --system input [14:34] could you both "pastebinit <( ls -latr /sys/class/net/eno*)" please [14:35] genii-testing: can you create a bug report on this? [14:35] I'm alerting pitti about it, but we ought to have a tracking bug. BluesKaj can then me-too the report [14:35] TJ-: with grep -n ... /var/lib/dpkg/info/udev.postinst:151: addgroup --system input [14:36] Against package udev? [14:36] genii-testing: thanks. that just confirms its a warning. if udev has previously added the 'input' group it lets 'addgroup' fail the 2nd time but you see the warning [14:36] genii-testing: yes [14:36] TJ-, ls -latr /sys/class/net/eno* just gives: /var/lib/dpkg/info/udev.postinst: addgroup --system input [14:37] OK, it's already fixed in the git repo... fix will come into packaging soon [14:37] tj oops wrong line copy , this is the corrct one: ls: cannot access '/sys/class/net/eno*': No such file or directory [14:38] TJ-,^ [14:40] we're sorted, the existing report is bug 1560112 [14:40] bug 1560112 in systemd (Ubuntu) " /var/lib/dpkg/info/udev.postinst: 109: [: Illegal number: *" [Medium,Fix committed] https://launchpad.net/bugs/1560112 [14:40] TJ-: Bug already exists, #1560112 [14:41] Marked it as "affecting me" [14:41] nice find [14:42] BluesKaj: that explains why the test fails. if no entries exist the loop variable will be empty [14:42] the code assumes there will always be entries, when they only apply to VMware guests in the main [14:42] right [14:44] Hello everyone. I am testing MariaDB on Ubuntu 16.04 (daily image). I've set a root password. I've noticed some strange behavior with MariaDB. For one, if I'm logged in as the Linux user root, I can access the MariaDB shell as root with any password (even if I enter a wrong password, I still get in). Also, no regular Linux user can access the root MariaDB shell, even if I type the right password. Is this something specific to Ubuntu's [14:44] implementation of MariaDB? Or maybe a bug? [14:44] jaythelinuxguy: upgrade or clean install? [14:45] lotuspsychje clean [14:45] * BluesKaj wonders when plasma 5.6 is gonna hit the Xenial backports [14:46] BluesKaj: not sure [14:48] ask in #ubunt+2 [14:48] ask in #ubuntu+2 [14:48] (silly joke) [14:49] :p [14:50] !find mariadb-server [14:50] Found: mariadb-server, mariadb-server-10.0, mariadb-server-core-10.0, W:, W:, W:, W:, W:, W:, W: (and 32 others) http://packages.ubuntu.com/search?keywords=mariadb-server&searchon=names&suite=xenial§ion=all [14:51] jaythelinuxguy: wich version are you on mate? [14:51] jaythelinuxguy cat /etc/issue [14:54] jaythelinuxguy: using mysqladmin tool? [14:54] !info mariadb-server [14:54] mariadb-server (source: mariadb-10.0): MariaDB database server (metapackage depending on the latest version). In component universe, is optional. Version 10.0.23-2 (xenial), package size 12 kB, installed size 59 kB [14:58] MonkeyDust: you skared him with your joke [15:00] unless MYSQL_PWD is set in the calling environment, or there is no password set for the mariadb 'root' user, there doesn't seem to be a way in [15:01] /etc/issue: Ubuntu Xenial Xerus (development branch) \n \l [15:02] I didn't use mysqladmin, I used mysql_secure_installation [15:03] jaythelinuxguy: you used that tool to try to access the DB server and got admitted without entering the DB root user's password? [15:03] I used the following as the Linux root user: mysql -u root -p (then I entered a wrong password on purpose) [15:03] I still got in [15:04] jaythelinuxguy: Are you still using the same shell session you called mysql_secure_installation from? [15:04] If I run mysql -u root -p (as a normal Linux user) I can't get into MariaDB at all, even with the right password [15:04] TJ no, I've rebooted several times since [15:04] jaythelinuxguy: ok, so not a leaky environment issue [15:05] I've also set up another fresh 16.04 install just to make sure, and same behavior [15:05] I'm looking at the sourcecode [15:06] jaythelinuxguy: did the password you used with mysql_secure_installation have any shell-special characters in? [15:07] TJ: The first time I did, but then I set the password to 'test123' and had the same issue [15:07] popey in MATE i cannot disable the window animation when reducing a window... already checked 'reduced resources' [15:09] jaythelinuxguy: OK, so nothing special there. It looks as if the shell script mysql_secure_installation.sh may not correctly write the root password into the temporary config file it creates, which would mean the password is empty. Have you, after accessing the DB from the system root user, checked the user and grants tables to see if a password is stored for 'root' ? [15:10] TJ: I have not, I'll check that now [15:11] the shell code has this: "UPDATE mysql.user SET Password=PASSWORD('$esc_pass') WHERE User='root';" [15:13] TJ: It does show a password hash for root, for localhost, 127.0.0.1, and ::1 [15:14] TJ: Also, the password hash changes when I run mysql_secure_installation again [15:15] jaythelinuxguy: right, and when you run the script you see "Password updated successfully!" when it sets the password? [15:15] TJ: Yes, it does say that. [15:15] jaythelinuxguy: ok, so the confirms it works through its set_root_password() function [15:16] so 2 things: 1) the password being hashed might not be the password you typed (resulting in a different hash) [15:16] 2) why the heck can the system root user get in without a password [15:17] you need to file a bug against the package [15:18] TJ: I also just created a test user, and that works perfectly as expected, I created a user, gave him grants, and then I can log in as that user. If I enter a wrong password for my test user, I can't get in, so that part works fine. This appears to be limited to the root user [15:18] I'll file a bug [15:18] jaythelinuxguy: another test you can do to get more detailed info [15:19] maybe the #ubuntu-hardened guys need to know this aswell, as it might be security flaw? [15:20] temporarily edit the mysql_secure_installation.sh script itself. Add "set -x" near the top so the script writes debug info to terminal. Secondly, locate the set_root_password function, and the line "esc_pass=..." followed by "do_query ..." and in between them add "echo "Escaped Password: $esc_pass" [15:21] then run the script as "mysql_secure_installation.sh ...args... |& tee /tmp/mariadb.log" [15:21] use that 'test123' as your password and see if the Escaped version is the same [15:22] Thanks TJ [15:23] that call to basic_single_escape() is doing a complex 'sed': "echo "$1" | sed 's/$['"'"'\]$/\\\1/g' " [15:23] that might be messing up [15:32] TJ: Your changes to the mysql_secure_installation script were brilliant [15:32] jaythelinuxguy: did it reveal a problem? [15:33] Still looking through the output [15:34] jaythelinuxguy: I'd grep for "Escape" to begin with :) [15:34] that will show you the password after it was escaped [15:34] Sorry. Yes, it did show the right password [15:34] Oh, that's a shame! That was the only place it could really mess up! [15:37] I filed the bug, and I checked the box that indicates it's a potential security vulnerability. Is there any harm in my making it public since 16.04 is beta anyway? [15:38] jaythelinuxguy: no... what's the bug # ? [15:39] I'll try to reproduce it [15:39] 1561062 [15:40] bug 1561062 [15:40] Error: Launchpad bug 1561062 could not be found [15:40] are you using amd64 architecture? [15:46] Yes, amd64 [15:46] In a virtualbox vm [15:52] I'm testing it in a chroot [16:04] !movelauncher [16:04] To move the Unity launcher to the bottom of the screen in 16.04+, run `gsettings set com.canonical.Unity.Launcher launcher-position Bottom`. To move it to the left, run the same command with Left instead. For older Ubuntu versions, see https://askubuntu.com/questions/33605/can-i-move-the-unity-launcher [16:05] for the xenial users who wanna test :p [16:08] my cpu goes berserk in xenial/unity, was forced to logout and switch to mate [16:09] MonkeyDust: any idea on the culprit? [16:09] MonkeyDust: thats weird...running pretty lightweight on my old amd 3200+ [16:09] doing fine on my Core2Duo as well [16:10] MonkeyDust: what cpu is that? [16:10] currently .9% while idle. [16:10] jaythelinuxguy: same result here. as system UID 0 "mysql -u root" I'm in, with another UID and "mysql -u root -p" I get "ERROR 1698 (28000): Access denied for user 'root'@'localhost'" [16:11] unity now, again: cpu 99-100% and slow performance [16:12] hmmz [16:12] MonkeyDust: clean install? [16:12] jaythelinuxguy: you need to read /usr/share/doc/mariadb-server-10.0/README.Debian.gz [16:12] intel here [16:12] jaythelinuxguy: it's not a bug; it's by design [16:12] lotuspsychje yes, always [16:13] jaythelinuxguy: that is referred to in the package changelog [16:13] anyway, gtg, bbl [16:13] kk [16:13] Thanks TJ I'll read that in a bit, I'll be afk for a few but I'll check that out [16:13] Instead the MariaDB root account is set to be authenticated [16:13] using the unix socket, e.g. any mysqld invocation by root or via sudo will [16:13] let the user see the mysqld prompt. [16:14] that doesn't explain why "mysql -u root -p" fails from another user though [16:22] jaythelinuxguy: looks like I've found the problem with it not setting the root password [16:24] TJ: what did you find? [16:31] It *looks* as if when not the root user because mysql is run from the localhost it still uses the Unix socket, and the non-root user doesn't have privileges, regardless of the user/password on the command line. Can you try setting a grant for root from '%' host ('root'@'%') and then try connecting from the host into the VM over the net with mysql on the host? [16:31] if I am correct that should work [16:31] I was looking for a way to disable unix socket usage but not found an obvious cmd-line option so far [16:32] Do you mean: grant all privileges on *.* to 'root'@'%'; [16:32] I did that and it still won't work, let me double-check [16:33] Yeah still can't log in [16:34] Oh sorry I didn't notice the "over the net" part [16:34] one sec [16:34] That doesn't work either [16:36] tried using "--protocol TCP" locally and that doesn't either [16:38] checked with tcpdump and "--protocol TCP" does have it talk over the 'lo' network interface, rather than the socket [16:39] also confirmed the hashed stored password matches by doing "select password('test123'); " [16:44] do you see anything in the /var/log/auth.log or /var/log/mysql/error.log ? [17:12] Nothing in those logs [17:17] I am getting local fonts clumsy http://i.imgur.com/ZAEd4Gy.png [17:19] I found something weird. I removed the root@localhost unix_plugin prequirement and it still fails over TCP [17:20] I then altered /etc/hosts to change 'localhost' to another name so it wouldn't reverse-lookup 127.0.0.1 or ::1 to that, but it still does, so it may never be trying to use the entries in mysql.user for root@127.0.0.1 or root@::1 [17:34] hmmm, according to the docs to use the unix_socket plugin it should have a plugin file named auth_socket.so in the plugins directory, and I don't see such [17:38] debian packaging adds 3 patches to make it built-in, one of which generates the default user table with the root/localhost entries, and that unix_socket in the plugin column [17:44] TJ: Very interesting, thank you for all your help. I do agree your findings are peculiar [17:49] I wonder if it being built-in is over-riding the network connection [18:23] jaythelinuxguy: just done some testing with a new user. If I 'create user ...' and then 'grant all on *.* to ... with grant option;' I can log-in correctly over TCP [18:24] TJ: That's the behavior I'm seeing as well [18:24] jaythelinuxguy: but as soon as I change the grant to be '...identified by unix_socket...' it no longer works. You cannot have multiple GRANT table entries for the same user@hostname [18:25] so you cannot add an additional GRANT for the same user@host which doesn't use unix_socket. [18:25] I'd describe that as a very bad GOTCHYA! bug - totally unexpected [18:25] Wow Ubuntu's implementation of MariaDB is getting stranger by the minute [18:26] this is standard with unix_socket; these patches were added in Debian and that's why they're in Ubuntu [18:38] it seems as this affects mysql 5.7 too - We've made the bug public and I've emailed the Debian maintainer === ibm is now known as Guest50932 [18:54] Thanks TJ for all your help, I'll watch the bug report as things develop [19:00] jaythelinuxguy: I've added some notes about the failure scenario just now [19:02] I thought I'd be sneaky and create another entry for root@localhost without the unix_socket, but the user table primary key is User|Host, grrr [19:07] jaythelinuxguy: found a sneaky workaround :D === alkisg is now known as alkisg_away === ubott2 is now known as ubottu === flocculant is now known as Guest75478 === Guest75478 is now known as flocculant === ibm is now known as Guest65405 === alkisg_away is now known as alkisg [21:08] after updating ubuntu it set grub to ro and not rw for my mounted swap file why is that? took a while to figure it out and fix it! [21:12] huh? [21:13] Madhumper69: err how is grub related to swap files? [21:16] linux /boot/vmlinuz-3.13.0-24-generic root=UUID=AAC884AC1F144321 loop=/ubuntu/disks/root.disk ro quiet splash $vt_handoff had to change ro to rw [21:17] in /etc/grub.d/forgetnameof file [21:17] loop is not a swap file [21:17] im running 16.04lts i copy pasted from forum [21:18] resolved my issue anyways [21:18] Madhumper69 #ubuntu+1 === Guest0 is now known as House_wrk [21:21] Madhumper69: the root file-system *should* be mounted 'ro' to being with; the init system will remount rw === alkisg is now known as alkisg_away [22:02] well by changing it to rw i can boot into gui otherwise i was getting mass errors and couldnt do anything [22:02] lots dbus errors [22:03] bare with me i am like 2months old to ubuntu and linux === _7 is now known as SonikkuAmerica [22:58] TJ-: BTW earlier when I did ubuntu-bug udev to report that issue, made the machine segfault. Just noticed it now as I'm shutting down for the night [22:59] genii: ouch! [23:00] Hockey watching time :)