Hi All,
I am just carrying this thread over from the now closed Precise thread with hopes of building on it.
Thanks, Ventrical
Commonly Used Sudo Commands for (*)Ubuntu/Unity Crash Recovery
Archived thread 1970289 from Ubuntu Development Version. Markdown source: Ubuntu_Development_Version/thread_1970289_Commonly_Used_Sudo_Commands_for_()UbuntuUnity_Cras.md
**Re: Commonly Used Sudo Code for (*)Ubuntu/Unity Crash Recovery** Here are some common codes of interest to help with Precise recovery in the event of a crash.
How to find your souces.list- this list is used to set repositories and can be done manually. It is also informative to have on hand if you decide to do a transitional upgrade from Oneiric Ocelot to Precise Pangolin, Precise Pangolin to Quantal Quetzel, Quantal Quetzel to Raring Ringtail, Raring Ringtail to Saucy Salamander, Saucy Salamander to Trusty Tahr, Trusty Tahr to Utopic Unicorn, UU to Vivid Vervet, VV to Wily Werewolf, WW to Xenial Xerus, XX to Yakkety Yak, YY to Zesty Zapus, ZZ to Artful Ardvarrk to Bionic Beaver(current development release).
[COLOR=Blue]/etc/apt/sources.list[/COLOR] - check here: [/COLOR]http://ubuntuforums.org/showpost.php...47&postcount=5
[COLOR=Green]Your beta version of Firefox is Broken- here are two possible fixes: [/COLOR]http://ubuntuforums.org/showpost.php...9&postcount=18 http://ubuntuforums.org/showpost.php...0&postcount=19 [COLOR=YellowGreen] [/COLOR] Here is a list of commonly used Ubuntu/Linux terminal Codes (not neccessarily in order and open to interperetation) of PP Crash Recovery Codes -[COLOR=Navy]To be updated:[/COLOR]
- This command is used to auto edit the sources.list file on a previous install and could be considered as a first step to converting to Trusty Tahr. However this and other commands may be moot after Alpha .iso are released. You can still play it safe and test the kernels but breakage may occur nonetheless. ```
sudo sed -i 's/artful/bionic/g' /etc/apt/sources.list
[COLOR=Blue]2.[/COLOR] This command will update your repositories to BB. ```
**sudo apt-get update && sudo apt-get dist-upgrade** [COLOR=Navy]3.[/COLOR]This command is inclusive in #2. but is always good to run after removing or purging stuff. ```
sudo apt-get update
[COLOR=Navy]4.[/COLOR]This command is also included in #2. and upgrades any new files that are set in the repositories. ```
**sudo apt-get dist-upgrade** [COLOR=Navy][COLOR=Blue]4.(a)[/COLOR][/COLOR]Update-manager is a fairly important component of the Ubuntu distribution. As we are supposed to be testing during the development phase, this application would be helped by testing and bug reporting as well.You can always make a judgement call as to whether the changes proposed by update-manager seem safe or not. If in doubt I think the best way to proceed is with aptitude: ```
aptitude update && aptitude safe-upgrade
This alleviates the concern when packages/dependencies may not be fully synced in the repos. They will be held back until dependencies are satisfied.
[COLOR=Navy]5.[/COLOR]This command is an essential command if you are a elementary beta tester as it will upgrade the GRUB bootloader after changes to the system using other commands, are made (like upgrading a kernel). If you have tried to carry out a proceedure and wonder why it had not taken effect on the next boot, it is likely that you did not sudo update-grub. ```
**sudo update-grub** [COLOR=Navy]6.[/COLOR]This command will give you simple information about your system, most commonly used to discover your video adapter. ```
lspci
[COLOR=Navy]7.[/COLOR]This command will display version information of the kernel you are using. ```
**uname -a** [COLOR=Navy]8.[/COLOR]This command will tell you what Version of Ubuntu you are using. It will help validate and document that the prior commands have worked properly. ```
lsb_release -a
[COLOR=Navy]9.[/COLOR]This command will continue to install packages that may have been broken during download or if your download unexpectedly terminated or if you have had a power-failure. ```
**sudo -i dpkg --configure -a** [COLOR=Navy]10.[/COLOR]These two commands can be used separately if you are at the terminal prompt from startup. You can get to the terminal (if you have no desktop) by pressing the keys <Crtl+Alt+F1> It makes booting and restarting quicker and more efficient than if you were in a desktop shell. ```
sudo reboot -sudo poweroff
[COLOR=Navy]11.[/COLOR]One that I have found helpful for fixing broken packages is . ```
sudo apt-get -f install [COLOR=Blue]12.[/COLOR]Will install any packages necessary to fix the broken package if they are available . If some necessary packages are not available you can use this to remove the broken one . ```
sudo apt-get -f remove
[COLOR=Blue]13.[/COLOR]Do not rely on any inaccurate information from the user as to whether he has (or not) added PPAs to his install. We just get the PPAs and read them. Then we parse the PPAs (from ppa-file-name.list to PPA:razz:paname/package format - which is the format required by the command ppa-purge) ```
**if [[ -d /etc/apt/sources.list.d && $(ls -l /etc/apt/sources.list.d/* | wc -l) -gt 0 ]]; then echo -e "\n\nPPA dir exists and is not empty\n\n"; **for PPA_FILE in $(ls /etc/apt/sources.list.d/*); do cat ${PPA_FILE} | grep "deb http" | sed -e 's|deb http:\/\/||' | sed -e 's|.launchpad.net/||' | sed -e 's|/ubuntu||' | sed -e 's|natty||' | sed -e 's|maverick||' | sed -e 's|oneiric||' | sed -e 's|main||' | tee -a $HOME/ppa-sources.list; done; echo "PPA list saved at $HOME/ppa-sources.list"; **else echo -e "\n\nEmpty or inexistent PPA directory\n\n";** fi [COLOR=Blue]14.[/COLOR]Attempts to fix a broken OO Install for NVidia 173 users ```
#!/bin/bash ################################################## # fix_oneiric.sh # Attempts to fix a broken OO Install for NVidia 173 users # [EMAIL="Effenberg0x0@Launchpad.net"]Effenberg0x0@Launchpad.net[/EMAIL] # ################################################## # # Save this file to a known location, such as your Home Folder # Execute it with sudo chmod +x fix-oneiric.sh && sudo bash ./fix-oneiric.sh # ################################################## # Assume the env vars we need may be wrong and get them # from safer sources ROOT_PARTITION=$(mount | grep -i "on / type" | awk '{ print $1}') ROOT_FS=$(mount | grep -i "on / type" | awk '{ print $5}') ROOT_WRITE=$(mount | grep -i "on / type" | awk '{ print $6 }') FIX_USER=$(cat /etc/passwd | grep 1000 | awk 'BEGIN { FS=":" }; {print $1}') ################################################## # Checks for a rw filesystem and remounts if needed if [ $ROOT_WRITE == "(ro)" ]; then mount -f -o remount,rw -t $ROOT_FS $ROOT_PARTITION / fi ################################################## # Make sure GUI sessions are stopped service stop lightdm service kill -s KILL $(pidof lightdm) service stop gdm service kill -s KILL $(pidof gdm) killall -s KILL /usr/bin/X ################################################## # Fixes user home permissions and ownership chown $FIX_USER:$FIX_USER /home/$FIX_USER -R && sudo chmod 750 /home/$FIX_USER -R ################################################## # Fixes Xauthority bug mv /home/$FIX_USER/.Xauthority /home/$FIX_USER/.Xauthority.backup ################################################## # Removes all nvidia remains cd /home/$FIX_USER mkdir nvidia_driver cd nvidia_driver apt-get remove --purge gdm nvidia-173 nvidia-96 nvidia-cg-toolkit nvidia-current-updates nvidia-173-dev nvidia-96-dev nvidia-common nvidia-current-updates-dev nvidia-173-updates nvidia-96-updates nvidia-current nvidia-settings nvidia-173-updates-dev nvidia-96-updates-dev nvidia-current-dev nvidia-settings-updates ################################################## # Downloads nvidia 173 driver wget http://us.download.nvidia.com/XFree86/Linux-x86/173.14.31/NVIDIA-Linux-x86-173.14.31-pkg1.run ################################################## # Exports needed and correct environment variables export DISPLAY=:0.0 export XDG_CURRENT_DESKTOP=Unity export XDG_DATA_DIRS=/usr/share/ubuntu:/usr/share/gnome:/usr/local/share/:/usr/share/ export COMPIZ_CONFIG_PROFILE=ubuntu export GDMSESSION=ubuntu export DESKTOP_SESSION=ubuntu export PATH=/home/$USER:/usr/lib/lightdm/lightdm:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games export XDG_SESSION_PATH=/org/freedesktop/DisplayManager/Session0 export XDG_SEAT_PATH=/org/freedesktop/DisplayManager/Seat0 export DEFAULTS_PATH=/usr/share/gconf/ubuntu.default.path ################################################## # Run nvidia installer chmod +x NV* ./NV* ################################################## # Reinstalls all critical DM/DE packages apt-get install --reinstall --fix-broken --allow-unauthenticated lightdm lightdm-gtk-greeter unity unity-2d unity-2d-launcher unity-2d-panel unity unity-2d-places unity-greeter unity-lens-music unity-services unity-2d unity-2d-spread unity-lens-applications unity-place-applications unity-2d-launcher unity-asset-pool unity-lens-files unity-place-files unity-2d-panel unity-common unity-lens-gwibber unity-scope-musicstores compiz compiz-dev compiz-kde compiz-plugins-main-default compizconfig-backend-gconf compiz-fusion-bcop compiz-plugins compiz-plugins-main-dev compizconfig-backend-kconfig compiz-fusion-plugins-extra compiz-plugins-default compizconfig-settings-manager compiz-fusion-plugins-main compiz-plugins-extra compiz-core compiz-gnome compiz-plugins-main ################################################## # Makes sure lightdm was not started service lightdm stop ################################################## # Fixes lightm config bug (if needed) if [ ! -f /etc/lightdm/lightdm.conf ]; then sudo touch /etc/lightdm/lightdm.conf echo -e "[SeatDefaults]\r\nuser-session=ubuntu\r\ngreeter-session=unity-greeter" | tee /etc/lightdm/lightdm.conf ################################################## # Fixes default-display-manager bug (if needed) if [ ! -f /etc/X11/default-display-manager ]; then touch /etc/X11/default-display-manager echo -e "/usr/sbin/lightdm" | tee /etc/X11/default-display-manager ################################################## # Makes sure compiz/compiz-config settings are resetted gconftool-2 --recursive-unset /apps/compiz-1 gconftool-2 --recursive-unset /apps/compizconfig-1 ################################################## # Backup of compiz/configconfig settings mv $FIX_USER/.config/compiz-1/compizconfig $FIX_USER/.config/compiz-1/compizconfig.old mv $FIX_USER/.config/compiz-1 $FIX_USER/.config/compiz.old mv $FIX_USER/.compiz-1 $FIX_USER/.compiz-1.old mv $FIX_USER/.cache/compizconfig-1 $FIX_USER/.cache/compizconfig-1.old ################################################## # Update/upgrade the system and reboot apt-get update && apt-get upgrade reboot now
[COLOR=Blue]15.[/COLOR]Unsolvable weird crash messages at most init procedures, ending in a "black screen" (or a console screen), no DM/DE after many attempts to fix a user setup (video driver, env vars, packages, etc).
A: Check for missing /run, or lack of permissions to write to it. Happens when users upgrade from old releases (that used /var/run and /var/lock instead of /run and /run/lock). Somehow the new install misses the creation of /run and /run/lock.
Check/Fix:```
if [ ! -d /run -a -d /var ]; then sudo ln -s /var/run /run && sudo mkdir -p /run/lock; fi [COLOR=Blue]16.[/COLOR]Unable to run apt-get. "Archive directory /var/cache/apt/archives/partial is missing" Check/Fix: ```
if [ ! -d /var/cache/apt/archives/partial ]; then sudo mkdir -p /var/cache/apt/archives/partial && sudo touch /var/cache/apt/archives/partial/lock && sudo chmod 640 /var/apt/cache/archives/lock && sudo apt-get clean; fi
[COLOR=Blue]17.[/COLOR]System mounted in "Read Only" mode. User is unable to edit config files or install any package, so no one can help him. All requested procedures will fail until it is set rw again. Happens when user has errors=remount-ro in fstab.
Check/Fix: ```
if [ $(sudo mount | grep -i "on / type" | awk '{ print $6 }') == "(ro)" ]; then sudo mount -f -o remount,rw -t $(sudo mount | grep -i "on / type" | awk '{ print $5}') $(sudo mount | grep -i "on / type" | awk '{ print $1}') /; fi [COLOR=Blue]18.[/COLOR]Just tried installing Precise for the first time today, using Alpha 1. I have an nvidia gs7600 card. Installation went fine but on reboot the system repeatedly hung without booting to the Desktop. The 'nomodeset' option did not help. Updating the packages from the recovery mode terminal didn't work either, nor did trying to start lightdm after a terminal log in. What finally allowed booting to the Desktop was removing all nvidia packages and then reinstalling *nvidia-current* (which also installed nvidia-settings). ```
sudo apt-get purge nvidia* sudo apt-get install nvidia-current
* [Last edited by cariboo907]("http://ubuntuforums.org/posthistory.php?p=11556749"); December 22nd, 2011 at 04:09 PM.. Reason: remove hard to read color formatting *
[IMG]http://ubuntuforums.org/images/statusicon/user_online.gif[/IMG] [[IMG]http://ubuntuforums.org/images/buttons/report.gif[/IMG]]("http://ubuntuforums.org/report.php?p=11556749") [[IMG]http://ubuntuforums.org/images/buttons/edit.gif[/IMG]]("http://ubuntuforums.org/editpost.php?do=editpost&p=11556749")Here is a most recent fix for a libxfont1 downgrade from pressureman:)
http://ubuntuforums.org/showpost.php?p=11894101&postcount=106
thanks ventrical , now if cariboo or someone can make this a sticky .
ronacc said: thanks ventrical , now if cariboo or someone can make this a sticky .
Thanks ronacc and Cariboo907 and tekstr1der.
How to fix a lighdm loop.
http://ubuntuforums.org/showpost.php?p=11894959&postcount=123
Followup to pin libxfont1 problem.
http://ubuntuforums.org/showpost.php?p=11895595&postcount=131
For ATI users: fglrx uses xorg.conf, but syntax has changed in the past year (a 10.10 xorg.conf will get you a black screen). So after upgrading, run: Code: sudo rm /etc/X11/xorg.conf sudo aticonfig --initial
How to set repository templates for QQ.
http://ubuntuforums.org/showpost.php?p=11877873&postcount=38 []("http://ubuntuforums.org/showpost.php?p=11878025&postcount=39")
Command to stop the lightdm display manager:
sudo service lightdm stop
sudo service gdm startHere is a fix that was scrounged up by ronacc for a very touchy nvidia-graphics-installer problem and the new 3.4.n.-n kernel.
http://ubuntuforums.org/showpost.php?p=11912598&postcount=33
I had some problems zsyncing the quantal alternate iso. I left a message for guitara but I think I got the wrong link. Anyways .. anyone trying to zsync the quantal-alternate-i386.iso can use this code and URL.
zsync -i ./quantal-alternate-i386.iso http://cdimage.ubuntu.com/daily/current/quantal-alternate-i386.iso.zsync
Thanks..
If you want to update the quantal alternate iso it is better to paste this code to terminal:
zsync -i ./quantal-alternate-i386.iso http://cdimage.ubuntu.com/daily/current/quantal-alternate-i386.iso.zsync
ventrical said: If you want to update the quantal alternate iso it is better to paste this code to terminal:
zsync -i ./quantal-alternate-i386.iso http://cdimage.ubuntu.com/daily/current/quantal-alternate-i386.iso.zsync
The "-i <input file>" is not necessary if the URL you are fetching is for an existing copy of the file. Zsync will automatically assume you are updating an existing download, use that as the seed, and backup the original file to *.old when finished updating.
@pressureman
Thanks:)
*note*
If any are having problems with startup video graphics try hitting <ctrl+F1> at the very start of booting your .ISo and then choose F6 and <nomodeset> . I was just able to resolve a problem with an older nVidia adapter card using this process.
*other note*
My apologies for not adding any extra content to U+1 or here in the forums. Iv'e been just so busy .
regards, ventrical
ventrical said: @pressureman
Thanks:)
*note*
If any are having problems with startup video graphics try hitting <ctrl+F1> at the very start of booting your .ISo and then choose F6 and <nomodeset> . I was just able to resolve a problem with an older nVidia adapter card using this process.
*other note*
My apologies for not adding any extra content to U+1 or here in the forums. Iv'e been just so busy .
regards, ventrical
Just to add a bit of a correction to the above post. Pressing any key will bring up the menu when booting from the iso.
I stand corrected ! :)
Code to remove ppas:
remove-
[COLOR=#000000][COLOR=#0000BB]sudo add[/COLOR][COLOR=#007700]-[/COLOR][COLOR=#0000BB]apt[/COLOR][COLOR=#007700]-[/COLOR][COLOR=#0000BB]repository [/COLOR][COLOR=#007700]--[/COLOR][COLOR=#0000BB]remove ppa[/COLOR][COLOR=#007700]:[/COLOR][COLOR=#0000BB]someppa[/COLOR][COLOR=#007700]/[/COLOR][COLOR=#0000BB]ppa
[/COLOR][/COLOR]purge- [COLOR=#000000][COLOR=#0000BB]sudo apt[/COLOR][COLOR=#007700]-[/COLOR][COLOR=#0000BB]get install ppa[/COLOR][COLOR=#007700]-[/COLOR][COLOR=#0000BB]purge sudo ppa[/COLOR][COLOR=#007700]-[/COLOR][COLOR=#0000BB]purge ppa[/COLOR][COLOR=#007700]:[/COLOR][COLOR=#0000BB]someppa[/COLOR][COLOR=#007700]/[/COLOR][COLOR=#0000BB]ppa [/COLOR][/COLOR]
Code to blacklist 'nouveau'
6) Blacklist nouveau Code: $ sudo nano /etc/modprobe.d/blacklist.conf Add these lines to the end of blacklist.conf: Quote: # Manually added to avoid conflict with nvidia proprietary module blacklist nouveau alias nouveau off
-------------------------7) Check for nvidia in blacklists Make sure you do not have nvidia blacklisted in other files at /etc/modprobe.d/* Code: $ cd /etc/modprobe.d $ grep -iHnr "nvidia" * The commands above should return only this: Quote: blacklist-framebuffer.conf:18:blacklist nvidiafb 8) Create a directory for the driver: It's useful to keep the installer at a known place (for eventuall reinstalls). Create a directory within your user $HOME for the Nvidia installer: Code: $ mkdir -p $HOME/Downloads/nvidia-drivers && cd $HOME/Downloads/nvidia-drivers 9) Get the installer: Download Nvidia installer for your architecture (32-bit or 64-bit). Run "uname -a" if you need to check whether you are using 32-bit or 64-bit.
For 32-bit: Code: wget ftp://download.nvidia.com/XFree86/Linux-x86/304.43/NVIDIA-Linux-x86-304.43.run For 64-bit: Code: wget ftp://download.nvidia.com/XFree86/Linux-x86_64/304.43/NVIDIA-Linux-x86_64-304.43.run OBS: If the direct links above fail, go to http://www.nvidia.com/Download/index.aspx?lang=en-us nd get your drivers there.
10) Check the download: Check whether the file was downloaded/saved to the directory we created: Code: $ ls -lha NVIDIA* Quote: -rw-rw-r-- 1 ahsl ahsl 62M Aug 28 05:04 NVIDIA-Linux-x86_64-304.43.run OBS: The result above is for the 64-bit installer. You might have downloaded the 32-bit version depending on your system specs/arch.
11) Make the installer executable: Code: $ sudo chmod +x NVIDIA* 12) Switch to a Virtual Terminal and stop lightdm/X:
Code: $ sudo service lightdm stop $ sudo killall -s KILL /usr/bin/X 13) Install: Run the Nvidia installer: Code: $ sudo ./NVIDIA* OBS: You should answer YES to ALL questions. The only question I answer NO is the one about DKMS. I have had some problems with it. YMMV.
- Press Ctrl+Alt+F6 and login with your user/pass
14) Restart: Back to the console prompt, restart the PC: Code: $ sudo reboot now 15) (Optional) Sing a happy song during boot: I suggest this one: http://www.youtube.com/watch?v=EcPxszGTCa8
Problems: 1) If you can't see lightdm (stuck in console mode on boot)
Code: $ lsmod | grep -i nv $ lsmod | grep -i nouveau
2) If you can't enter the Ubuntu session (loop back to lightdm):
3) If you have no Dash / Launcher, nothing in Unity:
Code: $ sudo apt-get install nux-utils mesa-utils $ /usr/lib/nux/unity_support_test $ glxgears -info
Code: $ sudo apt-get install compizconfig-settings-manager $ DISPLAY=:0.0 ccsm And activate the Unity plugin. 4) If you have a low performance, tearing, artifacts:
5) Monitors not configured:
- Is nouveau really blacklisted? Is nvidia *not* blacklisted? See steps 6 and 7. Which modules are loaded? Check with:
- Is anything non-standard being passed to the kernel cmdline at boot via Grub? See step 3. You only need special parameters if you know what they do.
- Did the install procedure went through with no errors? You can re-run it just to be sure. Repeat steps 12, 13 and 14 only.
- Are you using an outdated version of xorg packages? Were they pinned / Hold? Check steps 1 and 5.
- Maybe this is a failed install with no GLX support. Try this:
- Press Ctrl+Alt+T. If you don't get a terminal, switch to a VT with Ctrl+Alt+F6.
- Run these commands to check for glx support:
- If you have no support for GLX, you have failed to follow the steps above and should re-do the procedure.
- If you have glx support, the Unity plugin is not loaded. At the terminal / VT run:
- You might have to disable / enable "Sync to VBlank" in compizconfig-settings-manager (ccsm) / OpenGL Plugin to increase performance / avoid tearing depending on how things work for this card.
- You might have to disable / enable "Sync to VBlank" in nvidia-settings / "OpenGL Settings" to increase performance / avoid tearing, depending on how things work for this card.
- You might have to enable "Force fullscreen redraws (buffer swap) on repaint" on ccsm / Workarounds Plugin to fix tearing / repaint problems.
- Maybe the secondary monitor will be automatically activated, maybe it will not. I don't know why this is apparently not consistent.
- Go to Dash / Nvidia-settings and activate / configure the primary/secondary monitor if needed.
OBS:
- Yes, there are typos above.
- I started typing before you posted the second msg to this thread, so 90% of it does not apply to your problem. Yey :)
- You probably can benefit from steps 4 and 5 in the "Problems" section. Unless the procedure in Problems/Step 1 shows you not on nvidia but nouveau or, worse, llvmpipe or something (commands in step 3 can test that). In which case everything above applies :)
Regards, Effenberg
---------------------------------------
(thanks effenberg)
v3.7-rc2-raring
Here is the best and simple way to install kernek RCs,
Quote: Originally Posted by cecilpierce [[IMG]http://ubuntuforums.org/images/rebrand/buttons/viewpost.gif[/IMG]]("http://ubuntuforums.org/showthread.php?p=12310844#post12310844") [I]I would like to try it, what order do I install the 4 .deb files ? When I tried it I got a couple errors about vbox and something else, cant remember, dah![/I]
what i do with new kernel :
- download the required debs (2 headers + 2 images) into an empty folder
- then from that folder: sudo dpkg -i *
Thanks to dino99
If you are having problems with nvidia drivers and the new kernel RC then fooman has the answer. v3.7-rc2-raring
Re: Kernel 3.7-rc1 lost the nvidia drivers when i first booted up (GTX480), no top panel or unity....had to right-click on desktop to open terminal. from there i opened firefox and got the command to install the x-edgers PPA:
sudo add-apt-repository ppa:xorg-edgers/ppa
Code: jackel@jackel-GA-990FXA-UD5:~$ cat /etc/lsb-release DISTRIB_ID=Ubuntu DISTRIB_RELEASE=12.10 DISTRIB_CODENAME=quantal DISTRIB_DESCRIPTION="Ubuntu 12.10" jackel@jackel-GA-990FXA-UD5:~$ uname -r 3.7.0-030700rc2-generic ...working great!! :smile:
Anyone having problems changing their hostnames or want to rename them (if you did not get a chance to do so during install) then here is an excellent link to achieve that.
http://www.webupd8.org/2012/03/how-to-change-hostname-computer-name-in.html
If you download a document file (or any other file for that matter) and conventional ways of searching for that file do not work then you can use the gnome-terminal and enter the code:
dpkg -S name_of_downloaded_or_installed_file_here
This will give you the path and directory name as to where the file is. For example, if you are searching for the mir-doc files - here are the results of that code:
mir-doc: /share/doc/mir-doc/html/inherit_graph_89.png
mir-doc: /share/doc/mir-doc/html/inherit_graph_116.map
mir-doc: /share/doc/mir-doc/html/buffer__initializer_8h__dep__incl.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_null_logger__inherit__graph.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_native_client_platform_factory__inherit__graph.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_mir_binder_rpc_channel__inherit__graph.png
mir-doc: /share/doc/mir-doc/html/surfaces_2surface_8h__incl.map
mir-doc: /share/doc/mir-doc/html/android__client__buffer_8h__dep__incl.png
mir-doc: /share/doc/mir-doc/html/mir__binder__rpc__channel_8h__incl.md5
mir-doc: /share/doc/mir-doc/html/registration__order__focus__sequence_8h__dep__incl.png
mir-doc: /share/doc/mir-doc/html/size_8h__incl.map
mir-doc: /share/doc/mir-doc/html/inherit_graph_104.md5
mir-doc: /share/doc/mir-doc/html/search/all_79.html
mir-doc: /share/doc/mir-doc/html/mir__basic__rpc__channel_8h__incl.png
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_client_context__inherit__graph.md5
mir-doc: /share/doc/mir-doc/html/classmir_1_1surfaces_1_1_graphic_region.html
mir-doc: /share/doc/mir-doc/html/structmir_1_1shell_1_1_surface_creation_parameters-members.html
etc...
Command line code on how to tell which video driver you are using:
/usr/lib/nux/unity_support_test -p -f
Sould give a report somthing like this:
ventrical@ventrical-AcerAMD64bitDual:~$ /usr/lib/nux/unity_support_test -p -f
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string: GeForce G210/PCIe/SSE2/3DNOW!
OpenGL version string: 3.3.0 NVIDIA 304.84
Not software rendered: yes
Not blacklisted: yes
GLX fbconfig: yes
GLX texture from pixmap: yes
GL npot or rect textures: yes
GL vertex program: yes
GL fragment program: yes
GL vertex buffer object: yes
GL framebuffer object: yes
GL version is 1.4+: yes
Unity 3D supported: yes
ventrical@ventrical-AcerAMD64bitDual:~$
ventrical said: For ATI users:
fglrx uses xorg.conf, but syntax has changed in the past year (a 10.10 xorg.conf will get you a black screen). So after upgrading, run: Code: sudo rm /etc/X11/xorg.conf sudo aticonfig --initialUser in our LoCo stumbled upon this and (of course since copy&paste is prefered method worldwide...) complained... There should be some divider between these two commands...
sudo mv /etc/X11/xorg.conf /etc/X11/xorg.conf.ventrical
sudo aticonfig --initialWe use not to r(e)m(ove) file but just m(o)ve it, so, in case of any trouble, file can be easily retreived...
zika said: User in our LoCo stumbled upon this and (of course since copy&paste is prefered method worldwide...) complained... There should be some divider between these two commands...
sudo mv /etc/X11/xorg.conf /etc/X11/xorg.conf.ventrical
sudo aticonfig --initialWe use not to r(e)m(ove) file but just m(o)ve it, so, in case of any trouble, file can be easily retreived...
Shouldn't the command look like this?
sudo mv /etc/X11/xorg.conf /etc/X11/xorg.conf.ventrical && sudo aticonfig --initialcariboo907 said: Shouldn't the command look like this?
sudo mv /etc/X11/xorg.conf /etc/X11/xorg.conf.ventrical && sudo aticonfig --initialThe way the were written originally was a typo... That was the point of my post... To correct an obvious typo... Do not shoot the messenger... ;)
I wonder why this thread is under *Ubuntu +1*, and not in Tutorials & Tips. The commands mentioned here so far are not raring-specific, are they?
schragge said: I wonder why this thread is under *Ubuntu +1*, and not in Tutorials & Tips. The commands mentioned here so far are not raring-specific, are they?
They aren't specific to any +1 version, but we seem to get quite a few new users that have no idea how to solve a problem via the command line, so the thread will stay where it is, to help them on their way to fixing a broken installation.
zika said: User in our LoCo stumbled upon this and (of course since copy&paste is prefered method worldwide...) complained... There should be some divider between these two commands...
sudo mv /etc/X11/xorg.conf /etc/X11/xorg.conf.ventrical
sudo aticonfig --initialWe use not to r(e)m(ove) file but just m(o)ve it, so, in case of any trouble, file can be easily retreived...
Yes .. thanks .. I was busy that day , obviously :) My bad.
cariboo907 said: They aren't specific to any +1 version, but we seem to get quite a few new users that have no idea how to solve a problem via the command line, so the thread will stay where it is, to help them on their way to fixing a broken installation.
Thanks. I have been unbelievably busy. Unfortunately the wiki may suffer from non-updated content as well as here on the sticky. I am hoping to have another stretch of spare time come up soon and I have some utility duties on the list. This may not be until after the final release of raring .. but I am still on to getting this updated as soon as I can. :)
Regards,
Ventrical
We are discussing bringing back the Tutorial and Tips sub-forum, as for most the wiki experiment has been a dismal failure. I'm also not sure what is going to happen with the U+1 testing team, as Effenberg0x0 was going through some personal probelms, and has seemed to have disappeared.
cariboo907 said: We are discussing bringing back the Tutorial and Tips sub-forum, as for most the wiki experiment has been a dismal failure. I'm also not sure what is going to happen with the U+1 testing team, as Effenberg0x0 was going through some personal probelms, and has seemed to have disappeared.
I had noticed that users both verteran and new seem to be centered and affixed to Ubuntu-forums. Embedding a message base into the wiki would have been useless duplicity. All the hard, cutting edge action is at ubuntu-forums scratch pad. effenberg said he would be back . He owns U+1 wiki.
And as for U+1 ... is it not obsoleted?? If so , then the wiki was doomed to failure anyways not by my hands.
Cleaned out the U+1 Testers wiki and renamed it Rolling Release in my sig.
Regards,
Ventrical
ventrical said: I had noticed that users both verteran and new seem to be centered and affixed to Ubuntu-forums. Embedding a message base into the wiki would have been useless duplicity. All the hard, cutting edge action is at ubuntu-forums scratch pad.
effenberg said he would be back . He owns U+1 wiki.
And as for U+1 ... is it not obsoleted?? If so , then the wiki was doomed to failure anyways not by my hands.
Hey :) Sorry for disappearing, I had to take some time off and deal with personal stuff... I completely missed this cycle, but I'll be here for the next. BTW, I own nothing, everything is free as in (free beer|speech) :P There's a lot to do in testing-documentation, dev. and doc. of testing methods/best-practices, user-oriented documentation, etc. There has always been, there will always be room for contribution and improvement, whatever the development model is (rolling, bumping, jumping, swimming, etc). And we're not even talking seriously about testing in other platforms (mobile, etc). We'll think of something to do next cycle and, obviously, stuff to bother Nick with.
Regards,
effenberg0x0 said: Hey :) Sorry for disappearing, I had to take some time off and deal with personal stuff... I completely missed this cycle, but I'll be here for the next.
BTW, I own nothing, everything is free as in (free beer|speech) :P There's a lot to do in testing-documentation, dev. and doc. of testing methods/best-practices, user-oriented documentation, etc. There has always been, there will always be room for contribution and improvement, whatever the development model is (rolling, bumping, jumping, swimming, etc). And we're not even talking seriously about testing in other platforms (mobile, etc). We'll think of something to do next cycle and, obviously, stuff to bother Nick with.
Regards,
Good to you you back effenberg! :)
@effenberg
I pretty well butchered the tester wiki and am preparing it for 'Rolling Release' because the Raring links are all but obsoleted now and can't be zsynced (at least from here). When you get time .. lemme know where you would like to go with it and/or any other ideas you have.
Curious to see how "rolling release stable" and "rolling release unstable" will be differentiated in Daily Build.
As I do frequent updates the disk usage as defined by df gets more and more, despite all my efforts to apt-get clean, remove, autoremove, purge, etc. Since the disk usage grows that means to me more and more code that's not being used especially on amd64 so I do a re-install and may get over 10% disk space back. In the days of large disks who cares? I test using multiple stable and unstable partitions, and SSDs of any size are expensive.
Also curious to see how U+1 forum is handled, it's been very useful to me for years. The general "ubuntuforums" is not nearly so useful since the new stuff is very diluted, scattered, and I'm not expert enough in tips and hints specific searching.
BTW, as Canonical concentrates on smartphones and tablets (smartwatches?), pc's and pc releases are becoming less of interest perhaps occasional rolling release reflects their disinterest....
jerrylamos said:
Also curious to see how U+1 forum is handled, it's been very useful to me for years. The general "ubuntuforums" is not nearly so useful since the new stuff is very diluted, scattered, and I'm not expert enough in tips and hints specific searching.
Me too. I think this subforum has created sort of a community and its own culture and changing its purpose / communication might be a decision we will regret later if things, once again, change. We know decisions in FOSS and Ubuntu are sometimes made and later unmade, with or without regret, which is comprehensible: People make them with the information they have in hands and no one can really predict the future with 100% accuracy.
I'd hate to see this community disappear or lose relevance. And I'm not convinced we'll stick to one development model or another. We can all agree that we're testers: We'll continue testing/reporting, talking about features and bugs, with the goal of helping development and improvement. Also, some features can't simply be developed and implemented in a rolling model without some very alpha testing to check viability. Adding some very alpha ideas to rolling development will be too dangerous to be pushed to non-technical users updates. What I'm saying is that, to a certain level, testers are always needed, no matter what the development model is.
So maybe we should keep keep the subforum as "U+1" and operate as usual.
However, if for any reason we are required to change the name of the subforum, I'd advocate for "Ubuntu Testers". This name would finally acknowledge who we are and what we have been doing here for some years now... It would also present us and our activities properly to the larger Ubuntu Community. As a bonus, it would make our activities (as well as the risks involved in it) clearer to newbies.
I am confident any changes to our subforum name and activities won't be taken by forum admins without a proper discussion. And a proper discussion regarding this subforum is only acceptable if it involves its users - we should be able to express our opinions if any change is to be made. Also, we have many Ubuntu Members and Forum Admins and Moderators using this subforum. These people definitely understand what we do here and I believe they are likely to make thoroughly considered decisions.
(This is IMHO anyway).
Regards, Effenberg
effenberg0x0 said:
So maybe we should keep keep the subforum as "U+1" and operate as usual.
Funny you should mention that, as we were talking about that very thing last week. I plan on putting up a poll in the next couple of days, to see how the tester all feel about whether we should keep closing the sub-forum and changing to the name of the next version, or simply naming it U+1, and continue on as we have.
cariboo907 said: Funny you should mention that, as we were talking about that very thing last week. I plan on putting up a poll in the next couple of days, to see how the tester all feel about whether we should keep closing the sub-forum and changing to the name of the next version, or simply naming it U+1, and continue on as we have.
Hey Cariboo, I never really liked the idea of closing and restarting each cycle. Many bugs persist between releases, some concepts, workarounds, knowledge apply to all releases, etc. I'm very much in favor of not closing it anymore. Hopefully that's what the poll results will indicate.
Question: Let's say people decide to stop closing it. Would we keep the closed ones in the archive, as they are now, or would we merge them back? (is it even possible?) I mean, if we decide that it makes sense to not restart the sub-forum anymore, maybe it would make no sense to have the sub-forum content of the previous releases archived.
Regards, Effenberg
I won't say no to merging the old archives into the U+1, but at this time, I really doubt it, I'll have to bring it up with the forum council first. The way we've always done it, was started before the present forum council, and as such we have so far just followed tradition. We are making some changes that we hope will undo some of the restrictions previous councils have put in place, but we won't be making wholesale changes at any time soon.
cariboo907 said: Funny you should mention that, as we were talking about that very thing last week. I plan on putting up a poll in the next couple of days, to see how the tester all feel about whether we should keep closing the sub-forum and changing to the name of the next version, or simply naming it U+1, and continue on as we have.
I think this would be a great idea because merging the older threads would in fact be like a rolling model. Other testers would not be then so discouraged to "bump" an older thread up for comparative analisys to the current cycle. There will always be a symbiosis with previous cycles and keeping the U+1 thread a 'rolling' thread may be a good model to experiment with and may even set a template for the eventual rolling distro models to come. If you want to defragment the *knowledge base* then this would be the way to do it.IMO
We've got 8 development release sub-forums archived, and even though there are some underlying bugs that are still bothersome, much of the information in those sub-forums is nowhere near relevant today. The one big problem I see, is that the archived threads aren't indexed by Google at the moment, and if they are folded back into the U+! sub-forum, we could have quite a few more necromanced threads, than we are seeing already, recently we seem to be seeing 10- 15 threads a day brought back from the dead, that have to be dealt with. We don't want to put so much work on the moderation team with the things they deal with day to day that it isn't fun for them to do what they do any more, so we'll really have to think long and hard on this proposal.
Yes, that would be a tremendous load, however, suppose that , for experiemental purposes, it is proposed to the Forum Council to open the most previous archive (Quantal Quetzal) and put it in it's own forum under Specialized Support under a suggested heading "U+1 Ammendments-Quantal". I am sure that a small rule can be implied that the forums topics are to be topical for research and ammendments only in relationship to the current U+1 cycle. So there would always be two back. When '13.10' cycle is ready to roll it would be two back - to raring and quantal.
ie;
Ubuntu Specialised Support Ubuntu +1 (SS) Ubuntu +1 Ammendments Raring Ubuntu + 1 Ammendments Quantal
It's like being in a library. You have current research information in front of you but you have to examine an archive. The archive is locked in another room so you have to take an extra step to get there. After examining the archive material it is found that there is something significant to current or something current that is significant to editing an ammendment into the archive. Since the archive is locked , it cannot be edited. Of course the information can be pointed to in a link, quoted and corrected in the current volume, copy and pasted .. etc.. but would it not be much more efficient to make the ammendment as an addendum to the topic matter of any particular subject in question?
I have noticed that people are always referring to " something similar in the last cycle" ... um.. it's sort of like loosing your keys if you get what I mean. This way the knowledge base may be utilized more often and it could save some downtime.
Regards, Ventrical
Maybe U+1 sub-forum (current or archive) content shouldn't be indexed by Google at all, as it may provide nontechnical users with dangerous info... And that is easily doable via robots.txt, Google's Webmaster Tools, etc.
That would free us to merge the archives to current, reduce the risk to nontechnical users created by Google hits to U+1 content and, given our small traffic in comparison to General Help, it would hardly impact UbuntuForums popularity in Google searches.
Also, I believe we could merge the archives while keeping its threads closed, to avoid necromancy.
It's something to consider anyway.
I belong to another site where it is policy that posts for "like issues" are contained in a single sub-forum. We now have single threads approaching 2,300 pages and 34,000 posts long. Difficult to read, easy to become confused by outdated issues and almost impossible to follow individual points.
PJs Ronin said: I belong to another site where it is policy that posts for "like issues" are contained in a single sub-forum. We now have single threads approaching 2,300 pages and 34,000 posts long. Difficult to read, easy to become confused by outdated issues and almost impossible to follow individual points.
Sort of like the conky thread in the Cafe, 2175 pages and 21,748 replies, and well over 6.4 million views.
I don't think we'll see any of that type of thread here, but I do worry about threads that have old or bad information being brought back to life, until we get the auto-close plugin enabled.
Anyone know what happened to :
sudo -i dpkg --configure -a
I even tried it in Precise and it comes up with an error.
My follow-up question is;
Does there actually have to be broken packages for this to work?
Regards, Ventrical
ventrical said: Anyone know what happened to :
sudo -i dpkg --configure -a
I even tried it in Precise and it comes up with an error.
My follow-up question is;
Does there actually have to be broken packages for this to work?
Regards, Ventrical What error is produced?
ventrical said: Anyone know what happened to :
sudo -i dpkg --configure -a
I even tried it in Precise and it comes up with an error.
My follow-up question is;
Does there actually have to be broken packages for this to work?
Regards, Ventrical Doesn't the -i mean install. Is it just ``` sudo dpkg --configure -a
stinkeye said: Doesn't the -i mean install.
Is it just ``` sudo dpkg --configure -a
Let's see what is the error...zika said: „-i“ is used as a switch for „sudo“...
Let's see what is the error...
Exactly .. it is needed for root permissions. Back over a year ago I was doing an install/update and there was a lightning strike here. The power went out in the middle of while those packakes were being unpacked. I used that command and it worked just perfectly. Now, I guess the only way I can prove this is to do an update and emulate a power outage.
Anyways , I was just trying to get some confirmation as there is a discussion about this in another thread.
stinkeye said: Doesn't the -i mean install.
Is it just ``` sudo dpkg --configure -a
The -i is for root, However .. I tired it as you had posted and it just did nothing - no error at all. I think that command is specifically for broken packages. I ised it with the (-i) over a year ago when there was a power outage in my area and that command successfully fixed all the broken packages.ventrical said: The -i is for root, However .. I tired it as you had posted and it just did nothing - no error at all. I think that command is specifically for broken packages. I ised it with the (-i) over a year ago when there was a power outage in my area and that command successfully fixed all the broken packages.Man files are not so far away:
--configure package...|-a|--pending
Configure a package which has been unpacked but not yet config‐
ured. If -a or --pending is given instead of package, all
unpacked but unconfigured packages are configured.
To reconfigure a package which has already been configured, try
the dpkg-reconfigure(8) command instead.
Configuring consists of the following steps:
1. Unpack the conffiles, and at the same time back up the old
conffiles, so that they can be restored if something goes wrong.
2. Run postinst script, if provided by the package.
The:
sudo -i dpkg --configure -a
worked perfectly from terminal (not GUI terminal) [alt +F1 terminal]. I had dowloaded 75MB of updates and pulled the plug just as synaptic was starting to unpack the updates. Of course all packages were then updated with that command.
zika said: Man files are not so far away:
--configure package...|-a|--pending
Configure a package which has been unpacked but not yet config‐
ured. If -a or --pending is given instead of package, all
unpacked but unconfigured packages are configured.
To reconfigure a package which has already been configured, try
the dpkg-reconfigure(8) command instead.
Configuring consists of the following steps:
1. Unpack the conffiles, and at the same time back up the old
conffiles, so that they can be restored if something goes wrong.
2. Run postinst script, if provided by the package.
My bad ..
It's
sudo -i dpkg --configure -a
not
sudo dpkg -i --configure -a
thanks
ventrical said: My bad ..
It's
sudo -i dpkg --configure -a
not
sudo dpkg -i --configure -a
thanks It was like that in Your original post and we (re)solved that in prevoius posts...
zika said: It was like that in Your original post and we (re)solved that in prevoius posts...
No zika...
Please see the correction here:
http://ubuntuforums.org/showthread.php?t=2136435&p=12607444#post12607444
which was pasted from my post here:
http://ubuntuforums.org/showthread.php?t=1970289&p=11893873#post11893873
my bad :)
Sorry, but what is the purpose of -i here? I mean why do you need to cd to */root* and execute */root/.profile*? Wouldn't just
sudo dpkg --configure -asudo -i dpkg --configure -asudo su -lc 'dpkg --configure -a'
sudo bash -lc 'cd /root; dpkg --configure -a'
schragge said: Sorry, but what is the purpose of -i here? I mean why do you need to cd to */root* and execute */root/.profile*? Wouldn't just
sudo dpkg --configure -asudo -i dpkg --configure -asudo su -lc 'dpkg --configure -a'
sudo bash -lc 'cd /root; dpkg --configure -a'
ventrical said: No zika...
Please see the correction here:
http://ubuntuforums.org/showthread.php?t=2136435&p=12607444#post12607444
which was pasted from my post here:
http://ubuntuforums.org/showthread.php?t=1970289&p=11893873#post11893873
my bad :)You've got me... I'm unable to follow this circulus vitiosus...
zika said: No obvious (to me) reason... You're just completely right... As far as I know...
I can still see the purpose of it with *apt-get* if you are behind a proxy and */root/.profile* contains
export http_proxy=http://example.com:3128@zika
Don't worry about it.
@ cariboo907
Could you please edit this post.
http://ubuntuforums.org/showthread.php?t=1970289&p=11893873#post11893873
It is #9. on the list where it's states 'sudo dpkg -i --configure -a' It should be 'sudo -i dpkg --configure -a
Thank you.
@all
Most of the commands in this thread are copied and pasted Looks like one slipped through the net.
schragge said: Sorry, but what is the purpose of -i here? I mean why do you need to cd to */root* and execute */root/.profile*? Wouldn't just
sudo dpkg --configure -asudo -i dpkg --configure -asudo su -lc 'dpkg --configure -a'
sudo bash -lc 'cd /root; dpkg --configure -a'
I just copy and paste them as they come. When I first used this command I was told that I needed to be root (which is the -i). I had tried to use just the dpkg with out the sudo -i but it did not work at that time.
ventrical said: I just copy and paste them as they come. When I first used this command I was told that I needed to be root (which is the -i). I had tried to use just the dpkg with out the sudo -i but it did not work at that time.That must have been before 8.04... Since then it worked for me like a clock...
To temporarily gain root privileges, just *sudo* would be enough:
sudo dpkg --configure -aThe *-i* option is only needed if you also want to execute */root/.profile* to set up environment for root user. But if the only way to become root is through *sudo* as it is on Ubuntu then this could (and IMHO should) be done by editing */etc/sudoers*.
schragge said: To temporarily gain root privileges, just *sudo* would be enough:
sudo dpkg --configure -aThe *-i* option is only needed if you also want to execute */root/.profile* to set up environment for root user. But if the only way to become root is through *sudo* as it is on Ubuntu then this could (and IMHO should) be done by editing */etc/sudoers*.
Yes .. zika had pointed that out to me some time ago and I made the correction at the wiki but I cannot make the correction here as I do not have moderator privlidges. That's my whole point.
zika said: That must have been before 8.04... Since then it worked for me like a clock...
Actually it was during the Oneric Ocelot cycle. I'll see if I still have that install and check for logs/history.
ventrical said: Actually it was during the Oneric Ocelot cycle. I'll see if I still have that install and check for logs/history.
I do not have such install but a brief search gives: https://help.ubuntu.com/community/KarmicUpgrades/Kubuntu/8.04 In accordance with my loyal old notebook... One of the first entries... ;)
zika said: I do not have such install but a brief search gives:
https://help.ubuntu.com/community/KarmicUpgrades/Kubuntu/8.04 In accordance with my loyal old notebook... One of the first entries... ;)
Just beautiful :) Now I have the bug to search down the thread where I got that sudo -i from:)
@cariboo907
Thanks .. ummm but one thing .. you forgot the -a
sudo -i dpkg --configure -a
:)
.. and if you think it best to drop the '-i' then please do so.
Thanks again.
;)
I added the a
Not getting involved in a circular argument about -i :)
I might
sudo -ithen do
dpkg --configure -aIf I had a whole bunch of other stuff to do - but I'd only use
sudo dpkg --configure -a if that was all I was doing ;)
Thanks very much Elfy.
If anyone is having problems with gksu and password authentication then please check out this link for the fix.
http://ubuntuforums.org/showthread.php?t=2138386&p=12616119#post12616119
+1
Make sure that the sources.list text has this added info.
http://ubuntuforums.org/showthread.php?t=2138987&p=12619169&viewfull=1#post12619169
Just some notes on the btrfs file system.
Btrfs can be downloaded from the saucy repositories.
sudo apt-get install apt-btrfs-snapshot
*note* You have to have your file system formatted to btrfs when you are doing an install from Ubiquity. apt-btrfs-snapshot can only be run from the #root terminal. If you get no desktop after an upgrade and cannot access a terminal you can get to a terminal by holding down the <shift> key (or any key) which will bring up the GRub menu.
From the GRuB menu you can choose <Recovery Mode> option, then <root>. Once you have dropped to root# you can enter:
mount -o remount,rw /
then , if you type in <exit> it will bring you back to the recovery GUI with the option to choose an @apt-snapshot.
The other way is to enter all the information at the cli and then reboot from there.
root_name#> apt-btrfs-snapshot set-default @apt-snapshot_name_of_snapshot
This is a great recovery tool to learn how to use , especially if you are experimenting with 'mir' server ppa or other cutting edge Ubuntu tools.
How to tell your current graphics adapter card and driver:
sudo lshw -C display
Other desktop enviroments downloaded&installed from the repos or Ubuntu Software Center can and will overwrite data to a file:
/etc/lightdm/lighdm.conf and/or /etc/lightdm/lightdm.conf.d/50-unity-greeter.conf
The new lightdm.conf file will look like this after installing KDE Plasma Desktop:
[SeatDefaults]
user-session=kde-plasma
greeter-session=lightdm-kde-greeter
To restore your original greeter you have to edit : /etc/lightdm/lightdm.conf in this way:
[SeatDefaults]
user-session=unity-greeter
type=unity
regards..
How to reduce disk space usage using the command line.
You can reduce disk space usage, by cleaning out /var/cache/apt/archives using apt-get clean.
Before:
df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 9.5G 6.9G 2.2G 77% /
none 4.0K 0 4.0K 0% /sys/fs/cgroup
udev 992M 4.0K 992M 1% /dev
tmpfs 201M 2.7M 198M 2% /run
none 5.0M 0 5.0M 0% /run/lock
none 1001M 780K 1001M 1% /run/shm
none 100M 32K 100M 1% /run/user
/dev/sdb3 218G 83G 124G 40% /home
After:
df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 9.5G 6.3G 2.7G 71% /
none 4.0K 0 4.0K 0% /sys/fs/cgroup
udev 992M 4.0K 992M 1% /dev
tmpfs 201M 2.7M 198M 2% /run
none 5.0M 0 5.0M 0% /run/lock
none 1001M 780K 1001M 1% /run/shm
none 100M 32K 100M 1% /run/user
/dev/sdb3 218G 83G 124G 40% /home
I just cleaned the archive directory out last week, so there isn't a great difference, but I did gain about 500MiB.
as per ..
http://ubuntuforums.org/showthread.php?t=2174436&page=7&p=12797824#post12797824
ventrical said: How to reduce disk space usgae using the command line.
You can reduce disk space usage, by cleaning out /var/cache/apt/archives using apt-get clean.
Before:
df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 9.5G 6.9G 2.2G 77% /
none 4.0K 0 4.0K 0% /sys/fs/cgroup
udev 992M 4.0K 992M 1% /dev
tmpfs 201M 2.7M 198M 2% /run
none 5.0M 0 5.0M 0% /run/lock
none 1001M 780K 1001M 1% /run/shm
none 100M 32K 100M 1% /run/user
/dev/sdb3 218G 83G 124G 40% /home
After:
df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sdb1 9.5G 6.3G 2.7G 71% /
none 4.0K 0 4.0K 0% /sys/fs/cgroup
udev 992M 4.0K 992M 1% /dev
tmpfs 201M 2.7M 198M 2% /run
none 5.0M 0 5.0M 0% /run/lock
none 1001M 780K 1001M 1% /run/shm
none 100M 32K 100M 1% /run/user
/dev/sdb3 218G 83G 124G 40% /home
I just cleaned the archive directory out last week, so there isn't a great difference, but I did gain about 500MiB.
as per ..
http://ubuntuforums.org/showthread.php?t=2174436&page=7&p=12797824#post12797824 It depends very much on options You choose for apt... I'm not sure what are defaults but that is simple enough to check...Most of the time You don have even to think about clean/autoclean:``` :~$ ll /var/cache/apt/archives/total 116 drwxr-xr-x 4 root root 57344 Sep 24 21:25 ./ drwxr-xr-x 3 root root 4096 Sep 24 21:25 ../ drwxr-xr-x 2 root root 49152 Sep 24 21:01 apt-fast/ -rw-r----- 1 root root 0 Jun 24 22:57 lock drwxr-xr-x 2 root root 4096 Sep 24 21:20 partial/ :~$ ll /var/cache/apt/archives/apt-fast/ total 108 drwxr-xr-x 2 root root 49152 Sep 24 21:01 ./ drwxr-xr-x 4 root root 57344 Sep 24 21:25 ../ :~$ ll /var/cache/apt/archives/partial/ total 60 drwxr-xr-x 2 root root 4096 Sep 24 21:20 ./ drwxr-xr-x 4 root root 57344 Sep 24 21:25 ../
No sense keeping something You're goint to erase anyway...Thanks for the addendum.
;)
ventrical said: If you download a document file (or any other file for that matter) and conventional ways of searching for that file do not work then you can use the gnome-terminal and enter the code:
dpkg -S name_of_downloaded_or_installed_file_here
This will give you the path and directory name as to where the file is. For example, if you are searching for the mir-doc files - here are the results of that code:
mir-doc: /share/doc/mir-doc/html/inherit_graph_89.png
mir-doc: /share/doc/mir-doc/html/inherit_graph_116.map
mir-doc: /share/doc/mir-doc/html/buffer__initializer_8h__dep__incl.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_null_logger__inherit__graph.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_native_client_platform_factory__inherit__graph.map
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_mir_binder_rpc_channel__inherit__graph.png
mir-doc: /share/doc/mir-doc/html/surfaces_2surface_8h__incl.map
mir-doc: /share/doc/mir-doc/html/android__client__buffer_8h__dep__incl.png
mir-doc: /share/doc/mir-doc/html/mir__binder__rpc__channel_8h__incl.md5
mir-doc: /share/doc/mir-doc/html/registration__order__focus__sequence_8h__dep__incl.png
mir-doc: /share/doc/mir-doc/html/size_8h__incl.map
mir-doc: /share/doc/mir-doc/html/inherit_graph_104.md5
mir-doc: /share/doc/mir-doc/html/search/all_79.html
mir-doc: /share/doc/mir-doc/html/mir__basic__rpc__channel_8h__incl.png
mir-doc: /share/doc/mir-doc/html/classmir_1_1client_1_1_client_context__inherit__graph.md5
mir-doc: /share/doc/mir-doc/html/classmir_1_1surfaces_1_1_graphic_region.html
mir-doc: /share/doc/mir-doc/html/structmir_1_1shell_1_1_surface_creation_parameters-members.html
etc...Often method proposed gives less than complete answer to a question/query posed... There is (installed by default) better tool. Just use``` locate string_You_want_to_find
To refresh database that is being queryied by command given above simply use```
sudo updatedbExamples that I could think of are giving lengthy output so I do not want to burden Forum by quoteing them, You could try it for Yourself``` [COLOR=#000000]dpkg -S locate locate locate[/COLOR]
„Your“ method has one advantage: it gives the name of the package that brought a file to Your machine, but that is also the source of its disadvantage I'm writing about...
It simply does not fullfill promised: I've bol-ed that in qoute of Your message...[/COLOR]Thanks, something solved my login problem for Ubuntu 13.04. Mal
*note* This post deals with an experiment I was trying randomly in hopes of finding a more expeditious way to 'roll' or upgrade to the next development release (which is currently Trusty Tahr - 14.04)
The most commonly used standard and conventional method is to sudo sed...
sudo sed -i 's/saucy/trusty/g' /etc/apt/sources.list
and then
sudo apt-get update && sudo apt-get dist-upgrade
then
sudo apt-get update sudo apt-get dist-upgrade
with the caveat of editing the Ubuntu.info file at /usr/share/python-apt/templates/ubuntu.info
Some members, including myself, commented at how rather slow this method was.
I then decided to rename the 'saucy-desktop-i386.iso' to 'trusty-desktop-i386.iso' and then zsync the file to see if it would work.
zsync -i ./trusty-desktop-i386.iso http://cdimage.ubuntu.com/daily-live/current/trusty-desktop-i386.iso.zsync
and for some reason ,as which with Ubuntu never ceases to amaze me, the zsync worked, installed to USB using SCD and successfully installed to hdd with the Ubuntu.info file being updated afterwards:
sudo apt-get update sudo apt-get upgrade
Although my experiment was not very complicated it proved out that the development release can be appended to the last .iso release (or for that matter and previous iso release) and this makes for a very economic and expeditious way of updating/upgrading saving a tremendous amount of downtime and bandwidth.
This method of renaming is experimental. Although it may appear to be rock solid from my standpoint it may not be rock solid from the varying veiws or observations of others -so, as always, as is with development , this may bork your system.
regards..
How to get rid of :KVM disabled by BIOS: message in Trusty.
http://askubuntu.com/questions/263179/get-rid-of-kvm-disabled-by-bios
EDIT:
Special Note about Mir.
Development of Mir will not happen until April 2016 cycle (16.04). If you install Mir components or Unity8 there is a good probability that you will bork your system.
http://en.wikipedia.org/wiki/Mir_%28software%29
Old reminder note about Mir.
[URL="https://wiki.ubuntu.com/Mir/Installing"]https://wiki.ubuntu.com/Mir/Installing [/URL] There is a serious carfunkle here. After reading this link: (The link is no longer current or updated).<I took it upon myself to edit the page with pertinent updates>
http://ubuntuforums.org/showthread.php?t=2205637
it shows that many of the lighdm.conf files are no longer current with the data included in the wiki:
If you want to temporarily switch back to using X.org on your system, run following commands: $> sudo vi /etc/lightdm/lightdm.conf.d/10-unity-system-compositor.conf
and add a comment to the second line: [SeatDefaults] #type=unity
This has now been changed to represent to correct directory. I could find no current Mir or unity-system compositor web-sites with the new instructions.
New info on lighdm.conf files and directory changes and some how to's.
Edit: Note that Mir and/or Unity8 will most likely break your system. Development of Mir has ceased until 16.04 cycle so install Mir or Unty8 at your own risk.
http://en.wikipedia.org/wiki/Mir_%28software%29
New method to install mir.
sudo apt-get install unity-system-compositor
sudo apt-get install ubuntu-desktop-mir
The new location for the new compositor.conf file is:
/usr/share/lightdm/lightdm.conf.d/10-unity-system-compositor.conf
or here
Here is a really streamlined and reliable way to make a bootable USB disk with an (*)ubuntu image.
- Download the ISO
- Insert your USB stick in the knowledge that it’s going to get wiped
- Open the “Disks” application
- Choose your USB stick and click on the cog icon on the righthand side
- Choose “Restore Disk Image”
- Browse to and select the ISO you downloaded in #1
- Click “Start restoring”
- Wait
- Boot and select “Try Ubuntu….”
- Done *
How to boot with SystemD
courtesy of elfin post :)
to boot with systemd in utopic is simple
boot, edit the boot line - find the '*linux* line and add init=/lib/systemd/systemd then F10 will boot with systemd instead of upstart
when you're sure all is ok you can edit grub and add it there
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash init=/lib/systemd/systemd "
don't forget to update grub after changing that
Two more ways to go over the fence and fully/completely adopt SystemD: 1.``` sudo apt-get install systemd-sysv
2.```
sudo apt-get purge upstartUse with caution (changes are reversible in almost all the cases, AMD mostly, NVidia might be tricky) and only if You're really determined. I was and still am and have no regrets. To the contrary. Read appropriate thread to get more details about problems with NVidia etc.
zika said: Two more ways to go over the fence and fully/completely adopt SystemD:
1.``` sudo apt-get install systemd-sysv
2.```
sudo apt-get purge upstartUse with caution (changes are reversible in almost all the cases, AMD mostly, NVidia might be tricky) and only if You're really determined. I was and still am and have no regrets.
Thanks zika.
So it's not as easy as editing grub back to it's default and re-installing upstart?
I'm just asking ?
Regards..
ventrical said: Thanks zika.
So it's not as easy as editing grub back to it's default and re-installing upstart? I'm just asking ? Regards..> zika said:
*Two more ways* to go over the fence and fully/completely adopt SystemD:
1.``` sudo apt-get install systemd-sysv
2.```
sudo apt-get purge upstartUse with caution (*changes are reversible* in almost all the cases, AMD mostly, NVidia might be tricky) and only if You're really determined. I was and still am and have no regrets. To the contrary. Read appropriate thread to get more details about problems with NVidia etc. In these „two more ways“ there is not any change in /etc/default/grub needed as I've written in a thread mentioned. Change back, as I wrote is a „simple“ (mind the caveat I explained) purge of systemd-sysv or install of upstart since devs have done a good job and those two simply do not coexist together but do ask for other one when removed.
ventrical said: The only caveat is that I had to edit out```
init=/lib/systemd/systemd
zika said: As I wrote in message You've quoted, because You've put it there. Rarely clean job that dev(s) have done.
I asked elfy to move it there. That it would be more topical there in systemd experience.
But I'll put the link here for reference.
http://ubuntuforums.org/showthread.php?t=2224798&page=14&p=13146459#post13146459
Thank you elfy.
Thank you zika.
Upgrading to a new development cycle: Notes:
The rules remain the same but the speed of information is dynamic and can sometimes often collide. This 15.05 Vivid Vervet Cycle came very quickly, as expected by some, as a continuance of the anticipated rolling release philology.
Also, there is a certain order or 'flowchart' to follow when executing this proceedure. There are several diffirent methods that seem to work well for some and not so well for others. Mistakes will be made and systems will break - but for the most part the rolling releases have been designed to present us testers with somewhat of a fucntioning desktop experience (or a least a working X terminal).
There are basically two main files that have to be edited and updated. The first file is sources.list which can be found here :
/etc/apt/sources.list
and the Ubuntu.info file which can be found here:
/usr/share/python-apt/templates/Ubuntu.info
These two files are very different from each other but are essential for the proper update/upgrade to the new development cycle.
When the command:
sudo sed -i 's/utopic/vivid/g' /etc/apt/sources.list
is executed followed by:
sudo apt-get update && sudo apt-get dist-upgrade
and
sudo apt-get update
sudo apt-get upgrade
it will convert the sources.list text to the current repositories for (vivid) but it will also convert the templates in Ubuntu.info to /devel/
This may seem to be very confusing. Back a few cycles ago we were able to use /devel/ verebose to change our sources.list and get updates/upgrades but now it does not seem to work so , in this cycle, it is more prudent to use the first part of the release cycle name. Again , in this case that is /vivid/.
Cariboo907 has detailed the proceedure to insert the new template for the /vivid/ cycle here at this link:
https://wiki.ubuntu.com/U%2B1/common-problems#Software-Properties-Gtk_doesn.27t_work
Make sure to append this info at the top of the Ubuntu.info file.
There is a changelog URL. It can either be left in or commented out. Some say it is useful (and probably is) but it still reported errors on one of my installs. Surf the pertinent echos in Ubuntu Development Release and decide for yourselves. Try different things. It's free :)
Also some will have the proposed repositories turned on. Other would beg to differ to have them disabled. It is most surely if you have them enabled that you will bork your desktop and network , unless, of course , you are an advanced user.
Regards..
If you are having problems with apport reporting random bugs you can always turn it off.
Buckyball reported this: http://ubuntuforums.org/showthread.php?t=2251463&p=13160868#post13160868
and zika reported this: http://ubuntuforums.org/showthread.php?t=2251463&p=13160964#post13160964
Regards..
Startup Disk Creator is not working correctly, loaded with bugs and will bust a persistent ver 3.0 USB ISO when it is updated. However, that USB will work on ver 2.0 hardware but not version 3.0 hardware. Then , after sitting over night in off state it will not boot up at all and is not recognized by ver 3.0 hardware. It will make it appear that you ver 3.0 USB has a hardware brick, which , in fact, is no the case. If you have a similar bust then remove your partition from USB flash _drive, create a new partition and mark it ext4 and then try below.
Here is a method already posted that will fix that.
http://ubuntuforums.org/showthread.php?t=1970289&page=10&p=13145410#post13145410
We seem to be straying away for command line solutions here, so, another way to create a bootable live usb drive, is to use Disks->Restore disk image, there's no persistence, but at least it's more reliable than than the current boot disk creation utility.
How to get popular vlc dvd player running with the sudo commands at this link.
http://ubuntuforums.org/showthread.php?t=2266654&p=13234289#post13234289
This is the best current sudo command to get snappy working in a kvm:
dale@ventrical-MS-7798:~/Downloads$ kvm -m 512 -redir :8090::80 -redir :8022::22 ubuntu-15.04-snappy-amd64+generic.img
Precise 12.04 end of life April 25, 2017 --upgrade your Precise systems--
sudo do-release-upgrade -m server
jbicha mentioned to put this in a sticky post. Not sure if this is the right place as there does not seem to be a fix for it.
It has to do with unattended - upgrades and apt-update-service taking a long time in bootup sequence. The bugs are being worked on.
Here is one link:
https://ubuntuforums.org/showthread.php?t=2346439&p=13582859#post13582859
and here is the bug posted by harry:
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1686470
It looks like it is being worked on.
..so stand by..
regards..
Here is a sudo helper for upgrading from 17.04 to 17.10 from wgarcia .
I just updgraded to Ubuntu 17.10 from 17.04, and of course after restart I got gdm3 and only the option to start the Gnome desktop. To go back to Unity I had to issue Code: sudo dkpg-reconfigure lightdm to set up lightdm again as manager, and there I had the option to start Unity, which was still configured exactly as I left it in 17.04. The only bug I got so far is:
https://bugs.launchpad.net/ubuntu/+s...s/+bug/1703046
So now I'm in the ship of testing Unity along Gnome in 17.10.
There has been lots of problems with nvidia/gdm during 17.10 cycle, freezing on plymouth . etc.. Here is a workaround that worked for me:
We have been having lots of problems with RC and beta etc on PCs with nVidia.
Here is what I did as workaround/fix. I downgraded my nVidia card (changed it) to see if there was hardware problem.
1.Start the PC that has just had the daily/current hard installed.
2.(a) Wait for Grub screen. 2.(b) Choose <advanced options> THEN 2.(c) Choose Recovery option. 3.Wait till all verbose code runs. 4 .Click "resume". Don't do anything!
6.Enter password and it should load up ubuntu on xorg. (it did on mine because old nVidia does not support wayland)
- Hold down the shift key while booting.
- Wait for reboot. It may boot into weird gdm resolution with no gdm Cog-Wheel.
- There might be fuzzy resolution. Don't worry about it.
- Restart from panel.
- Go through normal boot and it should boot up to gdm screen with proper resolution.
- Log on and you will have proper resolution of gnome desktop.
You will notice that it loaded with the old vesa driver which is just too awesome!:smile:
my old nVidia card on gnome3 with xorg.
Regards..
Bump:
For 18.04 cycle.
- This command is used to auto edit the sources.list file on a previous install and could be considered as a first step to converting to Trusty Tahr. However this and other commands may be moot after Alpha .iso are released. You can still play it safe and test the kernels but breakage may occur nonetheless. ```
sudo sed -i 's/artful/bionic/g' /etc/apt/sources.list
[COLOR=Blue]2.[/COLOR] This command will update your repositories to BB. ```
**sudo apt-get update && sudo apt-get dist-upgrade** [COLOR=Navy]3.[/COLOR]This command is inclusive in #2. but is always good to run after removing or purging stuff. ```
sudo apt-get update
[COLOR=Navy]4.[/COLOR]This command is also included in #2. and upgrades any new files that are set in the repositories. ```
**sudo apt-get dist-upgrade** Regards..
P-I-H shares with us that this is an easier way to upgrade to new develoment cycle, avoiding software-properties-gtk crash.
sudo sed -i 's/artful/bionic/g' /etc/apt/sources.list
sudo sed -i 's/artful/bionic/g' /usr/share/python-apt/templates/Ubuntu.info
sudo apt-get update
sudo apt-get upgrade
sudo reboot
Regards..