I am trying to boot Ubuntu on my computer. When I boot Ubuntu, it boots to a black screen. How can I fix this?
If you are trying to install Ubuntu
1. Ensuring your CD/DVD or USB flash drive burned/written correctly and not damaged:
- How to MD5 test Ubuntu ISOs. See also: Where to find the checksums of Ubuntu ISO images? or MD5 hash for ubuntu .iso (14.04 through 17.04)
- Check out these steps to see how to check your CD once you boot into it to make sure it's ready to go.
- Verify the integrity of the burned CD/DVD or written USB flash drive (this also checks that it's accessible by the computer on which you wish to install).
2. Black/purple screen when you try to boot the LiveCD
The Ubuntu installer's startup portion is sometimes incompatible with certain graphics cards. Fixing it and getting to the Ubuntu Desktop to try or install it can often be surprisingly easy fix: the
nomodeset parameter. To see if it works for you:
Boot from the Desktop Live CD and press the ↓ key when you see the below:
Press Enter and select English:
Press F6, use the ← ↑ → ↓ keys to go down to
nomodeset, and press Enter. An x will appear to its left. Then press Esc, and press Enter to "Try Ubuntu without installing."
You can also try
acpi = off and
nomodset also shows up as a black screen.
3. Black screen
"you need to load kernel first" and "can not read file/cd0" errors
when installing to an UEFI capable machine:
Ubuntu's installer 'when attempting to run in UEFI mode) will hang and stop due to different manufacturer's implementations of the UEFI specification and will hang in different ways. To identify if your machine is booting in installer UEFI mode you will see
REF: UEFI Community Ubuntu Documentation Section 2.4
If your machine is CSM capable (which is a full UEFI implementation with an emulated BIOS layer) after selecting any option from the grub list the system will hang at a black screen.
The picture above actually only confirms your DVD/USB booted using UEFI and there will be some means in firmware settings to ensure drives are booted in order to make the UEFI installer run (a solution may possibly be as simple as ensuring SATA is set to AHCI) - check your vendors manual! Also check the UEFI Community Documentation Section 2.3 for more details.
What you need to do first is to disable SECURE BOOT in the firmware settings.
If that does not get the Ubuntu installer running, try disabling anything mentioning UEFI in the firmware settings.
If you cannot find UEFI settings then enable CSM - this will disable the UEFI booting of the installer and then allow a legacy/BIOS install of Ubuntu.
Installing grub-efi afterwards will allow UEFI to be re-enabled. Again refer to UEFI Community Ubuntu Documentation at Section 4
Some machines use a full BIOS with an emulated UEFI layer which may throw errors as described ie "you need to load the kernel first" and "can not read file/cd0"
Not all of these machines implement Secure Boot. Simply selecting UEFI in the BIOS settings will configure UEFI mode on hard drives. There is no solution for these errors and the workaround is to disable UEFI to enable the Ubuntu installer to run in legacy mode; after which boot-repair can be used to install
grub-efi which then allows/needs UEFI switched back on before Ubuntu will boot using UEFI. Once again refer to the UEFI Community Ubuntu Documentation at Section 4
Black/purple screen after you boot Ubuntu for the first time
This usually happens because you have an Nvidia or AMD graphics card, or a laptop with Optimus or switchable/hybrid graphics, and Ubuntu does not have the proprietary drivers installed to allow it to work with these.
The solution is to boot Ubuntu once in
nomodeset mode (your screen may look weird) to bypass the black screen, download and install the drivers, and then reboot to fix it for ever.
Start your computer, and press the Right Shift when booting up, to get the Grub menu. Use the ← ↑ → ↓ keys to navigate/highlight the entry you want (usually the first one).
Press e to edit that entry, which will show you the details:
linuxentry as shown above, use the ← ↑ → ↓ keys to get to it, and then press the End key to get to that line's end (which may be on the next line!).
nomodesetas shown, and press Ctrl+X to boot to where you can successfully install your graphics drivers.
If you are running Ubuntu 12.04 or 12.10, and have a ATI/AMD graphics card, you have to follow the instructions here, otherwise you will run into this problem every time you restart your computer.
In case you've installed Ubuntu with LUKS encryption / LVM option, it could be that Ubuntu just asks you for your password - and you cannot see it. If you have a black screen, try pressing Alt+← and then Alt+→ to switch your tty, this may bring back the password query and turn backlight back on.
If you have a purple screen (maybe you need to set the
nomodeset-option also?) and you have encrypted your complete Ubuntu installation, try to just type your encryption/LUKS-password after waiting some seconds (or minutes, just to be sure) and continue with a press on Enter. If this is successfull, you should see your Login-screen just a few seconds later.
- Try the Alternate Installer - this is a text based installer that might work better than the liveCD depending on your hardware.
If you do this, you may get a black screen before you even hit grub. A simple Ctrl+Alt+Del gets some users to a usable grub screen
If you have an Nvidia Optimus card you should NOT install nvidia drivers, just use the built in driver, see here:
After selecting boot options you have the opportunity to edit the boot flags manually using your keyboard. Replace
no splashto get an idea of what step your system is failing at. Using that information search the forums or the internet for answers from the community.
If you still can't install Ubuntu then unfortunately you've probably run into a hardware specific bug, please see here: How do I report a bug?
If an update or something else caused your boot problem:
Grub recovery cross links
Reinstalling your machine
Sometimes a kernel upgrade can cause problems, especially if you're using closed drivers, you can confirm if this is the case by booting into your old kernel.
LiveCD recovery crosslinks
- and so on ...
- If you still can't boot Ubuntu then unfortunately you've probably run into some other kind of bug, please see here: How do I report a bug?
If your graphics card is Nvidia, follow these steps:
- In the GRUB menu at startup, press e; then,
- Use the arrow keys to replace
- Then press the Ctrl+x key combination to boot.
If your graphics card is ATI, follow these steps:
- In the GRUB menu at startup, press "e"; then
- Use the arrow keys to replace
- Then press the Ctrl+x key combination to boot.
It appears as we both share the same hardware. At least the same CPU, The same P67 and the same video card (Mine is a GT 440). What you can do is the following:
When booting press the ESC Key or leave the SHIFT key pressed until the GRUB menu appears.
In the GRUB menu select the RECOVERY MODE. It should say something like:
Ubuntu, with Linux 3.2.0-25-generic-pae (recovery mode)
After the recovery mode finishes loading, it should present you with a menu. Select the option about going to the shell with root access (root Option. Last option in the image below)
Now we need to install the Nvidia drivers and update the system. Follow the next lines step by step to do this in the correct order. Also make sure you are connected to the Internet:
A. Updating System
sudo apt-get update
sudo apt-get upgrade
B. Nvidia Drivers
There are 2 ways to install the Proprietary Drivers. the PPA Way or the Default Ubuntu way.
The Default Ubuntu Way involves just typing in the terminal
sudo apt-get install nvidia-current(For the current normal drivers) or
sudo apt-get install nvidia-current-updates(For the latest current drivers). Just pick one.
the PPA way has the Latest bleeding bloody edge drivers. I mention this one since I am testing it in some use cases that relate to problems using TVs and 16:9/16:10 resolutions. To install this one do this:
sudo add-apt-repository ppa:graphics-drivers/ppaand press ENTER to accept a message you will receive. Then do
sudo apt-get update. Lastly do
sudo apt-get install nvidia-375. Note that you can not mix this one with the Ubuntu way. One will overwrite the other one.
Now if you happen to have any problems do the following via the terminal again but this time go to your home folder. In my case it is
/home/cyrex, so I would
cd /home/cyrex. In your case you should change that to your user and the apply the following:
sudo rm -fr ./config/monitors.xml sudo rm -fr .nv* sudo rm -fr /etc/X11/xorg.conf
Then when you boot into Ubuntu change the resolution via the Displays option in the Ubuntu Settings Menu (The cog in the upper right corner)
What we did there was remove the monitors.xml to solve some resolution problems, remove the .nvidia-settings to fix some Nvidia config problems and remove the xorg.conf (Which is not really needed in the latest Ubuntu versions) to remove any badly configured options.
Lastly we can execute the grub option in the recovery menu to fix any bootloader issues.
I had this problem last night. All of a sudden my system wouldn't boot up anymore. BIOS check would finish, then it would just hang there on a black screen with the cursor flashing. Left it there for several hours just in case. When that didn't work, I unplugged all my USB devices and all of a sudden it booted up fine again. I haven't narrowed it down exactly, but in my case it was either my USB hub or the iPod plugged into that USB hub that was causing it to hang.
Not saying this is necessarily the problem you're having, but hope your boot problem is as easy to fix as unplugging some USB devices...!
If you are using the Windows Installer (Wubi)
Wubi overrides are identical to normal installs except the first time you reboot after running the installer in Windows.
To complicate things, since Ubuntu 11.10 there are two distinct methods to install with Wubi. The first way is using the Desktop ISO, which applies to all sub-flavours (Kubuntu, Lubuntu, Mythbuntu etc.) and also if you downloaded the Ubuntu ISO yourself.
The second method is using a pre-installed disk-image, if you run
wubi.exe standalone and choose to install Ubuntu.
You'll see this:
Completing the Ubuntu installation For more installation options, press ESC now 5...4...3...2...1
Press Esc and then you see this:
Normal mode Safe graphic mode ACPI workarounds Verbose mode Demo mode
Ignore Safe graphic mode as it applied to Ubuntu in 2008 and does nothing for the modern nvidia/radeon issue. Place your cursor on Normal mode and press E. Then edit the entry and insert
nomodeset as shown here (look for it between
quiet splash in the middle; note there may be some other differences but don't change anything else - just add
linux /ubuntu/install/boot/vmlinuz debian-installer/custom-installation=/ubuntu/install/custom-installation iso-scan/filename=/ubuntu/install/installation.iso automatic-ubiquity noprompt nomodeset quiet splash boot=casper ro debian-installer/locale=en_US.UTF-8 console-setup/layoutcode=us console-setup/variantcode= -- rootflags-syncio initrd /ubuntu/install/boot/initrd.lz
Now press Ctrl+X to boot.
NOTE: This only applies to the Installation; the next time you boot you have to override it again, and for this it will be the same as for a normal install (answered above). Make sure you hold Shift to make the Grub menu show though.
When you run
wubi.exe standalone and install Ubuntu (not a sub-flavour), it downloads a pre-installed, compressed disk image with a default Ubuntu install, and then decompresses this to the size of the virtual disk. There is no
grub.cfg setup yet so it uses the file
\ubuntu\install\wubildr-disk.cfg for the first boot which you can edit and add
loopback loop0 /ubuntu/disks/root.disk set root=(loop0) search --set=diskroot -f -n /ubuntu/disks/root.disk probe --set=diskuuid -u $diskroot linux /vmlinuz root=UUID=$diskuuid loop=/ubuntu/disks/root.disk preseed/file=/ubuntu/install/preseed.cfg wubi-diskimage ro quiet splash nomodeset initrd /initrd.img boot
Note - if you've come to this thread after booting for the first time, it's possible that the
grub.cfg has already been created (even if it froze up). In this case, editing the
\ubuntu\install\wubildr-disk.cfg file will do nothing - it always checks for
/boot/grub/grub.cfg inside the virtual disk first. So you should follow the instructions for the normal install above.
Note also that the Grub Menu is suppressed by default on Wubi installs (even though there are two operating systems - because you boot Ubuntu from Windows, and therefore adding a Windows entry from Ubuntu's Grub Menu makes no sense) so you have to press and hold the Shift key after selecting Ubuntu in order to display the Grub Menu. On Windows 8, it reboots after you elect to boot into Wubi, in which case, you should hold the Shift key after the BIOS posts.
(This is only for Windows 8 with BIOS - Wubi doesn't work with UEFI).
I also had this problem, or a similar one. It turned out that, for some reason, Ubuntu had started with the screen brightness on its lowest setting. If I went into a very dark room, I could see the screen just well enough to go to the "Brightness and Lock" control panel and turn the brightness up to where it should be.
how to fix :
Start ubuntu, login, now in the blackscreen go to the console Ctrl+Alt+F1 and type
nautiluscannot be opened in terminal type
- now without exiting or stoping the
unitycomand return to the ubuntu interface in my pc is Ctrl+Alt+F7
- open terminal in nautilus using Ctrl+Alt+T and type
nautilusstill open, in terminal type
- search for graphic drivers, my problem was the default drivers of X.org. I have an ati video card and I installed the drivers of fglrx-updates if you have nvidia install the drivers of nvidia or if you have ati driver
- after login u will see nautilus running well and the wallpaper the only thing not working is unity. open terminal and type:
dconf reset -f /org/compiz/and
I hope this fix your problem.
I realize this is an old question, but it's also pretty general without any details about the specific hardware involved. That said, you can't file a bug or go about fixing things until you figure out some more details.
I thought I'd take a stab at this since I faced the issue and recovered from it pretty recently. I'll probably run through here again later and throw in some more info and simplify the steps, but the answer list is already pretty big, so I'll go easy on the screenshots.
Recovery mode is your friend, but you don't always need a single-user root session to solve things. In fact, you might just be able to do a normal console login by selecting "resume" without considering any of the other options on the recovery menu. The nice thing about a normal console session over the single-user root mode is that you can get multiple terminals running at once--Switch between them or open up new ones with Alt+F1, Alt+F2, etc. There's a good chance that it's a video driver issue which is preventing you from going into the graphical login, and it might just be a result of some upgrade you did before rebooting the computer.
You might go a couple of years at a time without experiencing similar issues, but it's a good idea to know your hardware and to be prepared to use the terminal. Basically there are two video drivers to worry about: the kernel driver and the xorg driver. Xorg is a video server that uses the x11 protocol to display things in full color with depth and all kinds of crazy effects--It's an abstraction layer between applications like the desktop environment or windowing managers and the kernel driver. The kernel driver is yet another abstraction layer, but it's a bit closer to communicating with the actual hardware.
It's the kernel's job (in this case, Linux) to pass messages between applications and the hardware. The drivers can either be compiled into the kernel or added in a more ad hoc way through kernel modules. Probably you're using modules unless you configured and compiled your own custom kernel. The kernel driver as a module gets loaded shortly after you boot up, which allows for easier upgrades when you power down to swap out a card. The good news is that there are some more or less standard tools that you can run from the command line to give you more information about those kinds of drivers, the actual hardware and whether they're loading: lspci, dmidecode and dmesg, to name a few. There are man pages (e.g.,
$man dmidecode) and many howtos on those kinds of tools, so I won't go into too much detail here for now.
Then there are the xorg drivers. To list what's available in the repositories, you might type
apt-cache search xserver-xorg-video | less to give you a list of all possible drivers. Piping it to
less with the '|' symbol which you can probably type by tapping the slash key while holding down shift (to be clear on what symbol this is), gives you the option to scroll back and forth through the list of drivers (with the arrow keys). To get more info on a specific driver, you might type
apt-cache show xserver-xorg-video-vesa (to pick one at random). To install one, you could type
apt-get install xserver-xorg-video-vesa and hope for the best. As of I don't know how many versions ago Xorg will try to load one of the installed drivers for you automatically, but under certain conditions you might have a configuration file lingering around in
xorg.conf. So take a look and see if there's one there:
If you upgraded an Xorg driver without directly upgrading Xorg itself, there's a chance that reverting to the old driver via
apt-get install will not automatically pull in the version of Xorg that it's compatible with--It should but apt doesn't always do what it should. Minimally, you'll need a matching version of xserver-xorg-core. Don't bother with uninstalling the upgraded xorg replacement though, just enter the command
apt-get install xserver-xorg-core to revert back and uninstall the newer version automatically. This advice applies mostly to transitional renamed packages which provide virtual packages to replace ones that are still being maintained in the same branch of the package tree. Virtual packages are sometimes a mess and can do funny things with any of a number of dependencies which are getting swapped around in the upgrade/downgrade process, but concentrate on getting back to the GUI first.
Now that I've given an overview of some directions to start with troubleshooting, let's get back to the console screen that you hopefully pulled off without a hitch from choosing "resume" at the recovery menu. It's a pain to be stuck without a mouse at the console when you've got a lot of copying and pasting to do, so prepare yourself with some gpm for mouse support and some other tools: links/links2 or w3m (web browsers), vim (text editor), dpkg, apt, less (vim style keys and searching like man), and grep. I'm probably leaving a few out.
Some particularly useful commands for dpkg are
dpkg -L to show files for packages that are already installed and
dpkg -l | less to show all packages which are currently installed (piped to
less).Sometimes gpm is a little finicky about letting you select things, so you can restart it with
/etc/init.d/gpm restart but you might have problems with clicking on links in a page before you restart w3m or the browser
links. w3m is a little easier to scroll around and generally better for authenticated sessions (e.g., logging into forums for help). It takes a bit of getting used to hitting the Esc key to click on links (the hyperlinks) though, and the learning curve is a bit steeper than with the browser known as
Unless you've got an Nvidia card or something with proprietary driver support for linux that you want to try, I'd shy away from kernel drivers before trying things with xorg--Try troubleshooting the xorg drivers first because it can be a lot easier than customizing a kernel for hardware (depending on the brand). The thing is that you might wind up following a series of links that lead you in the wrong direction, with chip makers sending you to the card makers and card makers giving you no support. As for trying out different kernels for different "vanilla" versions of the driver, stick with kernel versions that aren't far off from your current one (given by
uname -r) unless you're really interested in testing. There's a pretty good chance that the latest mainstream kernel won't even boot up on your system, so why bother if you're stuck with a half-way broken setup? Keep focused on doing the bare minimum that it takes to get back up again so you're not falling behind on too much work. You can type things up in emacs, vim or pico/nano or check your email in mutt or pine, but eventually you'll want to come back to the 21st century.
If Ubuntu 10.04 booted, but not 10.10 or newer versions booted
Chances are your computer's ACPI is not supported. Ubuntu 10.04 supported drivers for the ACPI, but dropped supported for that since 10.10.
To try that, change the BootOptions to
nolapic and see if the Ubuntu Live CD boots (info).
If it worked, you have 2 options to make the LiveCD boot:
Disable the "New Card Interface" (which I saw the method in UbuntuForums)
Go to BIOS > Security> I/O interface Security> "New interface card". Set it to Locked.
Make sure you have a Windows Recovery CD available if you have Windows, because Windows may show up as an error.
Or, disabling the ACPI, which is not recommended because it may disable some crucial features on you computer (like fans). This is not recommended unless you have tried the first option and you know what you are doing!
Restart your computer.
Press the key indicated during the startup messages (usually F2, Esc, or F1) to enter BIOS.
Click on the "Power Settings" menu.
Highlight the ACPI entry, press Enter, select "Disabled" and press Enter again.
Press Esc and Y to exit and save.
Again, make sure you have a Windows Recovery CD available if you have Windows, because Windows may show up as an error.
Different PC manufacturers have different BIOSes, so read your computer's manual if your computer's BIOS doesn't show up.
I had the same problem.
I just fixed it. (kind of) My solution (so you can boot back into your GUI) (don't know if it really was this or if it where something together):
open TTY (CTRL + ALT + F1) to type the commands.
make sure the Radeon open-source drivers are installed:
delete all the fglrx drivers:
I first tried: https://wiki.ubuntu.com/X/Troubleshooting/VideoDriverDetection#Problem:_Need_to_purge_-fglrx
and later tried this:
sudo apt-get remove fglrx*
after that, I could reboot and log in.
OPTIONAL: if you have a login loop (type password and is accepted, it will re-ask for your password):
(make a backup of the file)
mv ~/.Xauthority ~/.Xauthority.old
and retry to log in.
any questions, just ask.
If you have more than one video port on your graphics card (or more than one graphics card), then plugging the monitor into a different port may fix the problem.
I've had an issue in the past with dual-DVI graphic cards, where it won't boot using one of the ports (secondary maybe?), but works fine on the other port.
I upgraded from 10.10 to 11.04 to 12.04 LTS, and when 12.04 loaded for the first time, I had a blank screen after the GRUB menu. It actually booted fine - I could SSH into the system as usual, but the video at the actual console didn't work. Booting into recovery mode worked fine, too. Here was the solution.
Hardware = Macbook, early 2008, Macbook4,1. (Black - Early 2008/Penryn)
After much trial and error, the solution was the editing and updating GRUB as shown below. The relevant edits were
- GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset"
sudo nano /etc/defaults/grub
# If you change this file, run 'update-grub' afterwards to update # /boot/grub/grub.cfg. # For full documentation of the options in this file, see: # info -f grub -n 'Simple configuration' GRUB_DEFAULT=0 #GRUB_HIDDEN_TIMEOUT=0 GRUB_HIDDEN_TIMEOUT_QUIET=true GRUB_TIMEOUT=2 GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian` GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset" GRUB_CMDLINE_LINUX="nopat" # Uncomment to enable BadRAM filtering, modify to suit your needs # This works with Linux (no patch required) and with any kernel that obtains # the memory map information from GRUB (GNU Mach, kernel of FreeBSD ...) #GRUB_BADRAM="0x01234567,0xfefefefe,0x89abcdef,0xefefefef" # Uncomment to disable graphical terminal (grub-pc only) #GRUB_TERMINAL=console # The resolution used on graphical terminal # note that you can use only modes which your graphic card supports via VBE # you can see them in real GRUB with the command `vbeinfo' GRUB_GFXMODE=1024x768 # Uncomment if you don't want GRUB to pass "root=UUID=xxx" parameter to Linux #GRUB_DISABLE_LINUX_UUID=true # Uncomment to disable generation of recovery mode menu entries #GRUB_DISABLE_RECOVERY="true" # Uncomment to get a beep at grub start #GRUB_INIT_TUNE="480 440 1"
Turns out I had to create a LiveCD and completely reinstall GRUB. After I did this, things are working fine again.
Documentation for re-installing grub is here : help.ubuntu.com.
This one helped at least a bit: http://ubuntuforums.org/showpost.php?p=9965194&postcount=8
gksu gedit /etc/initramfs-tools/conf.d/splash.
- In the text editor, add
FRAMEBUFFER=yto the file.
- Save the file and quit the text editor.
- Run (in a Terminal):
sudo update-initramfs -u
When the grub boot menu comes up, press e to edit the correct boot line. Where it says
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash" change this to
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset".
If this works, then you can install proprietary graphics drivers which should get you going again.
If you have trouble with the above, you can change the above line permanently in the grub configuration file.
Boot into recovery mode (selection from the grub boot loader) and edit the file
/etc/default/grub as per the above instructions.
You can edit this file by typing:
You'll have to update the grub menu:
then reboot by typing
I had this issue with Ubuntu 12.04 64bit version. I install all goes well, I enable the Nvidia drivers reboot and nothing but a black screen. I re-installed several times with the same results. I then remembered that I had this problem with previous versions of Ubuntu. I downloaded and installed the 32bit version then installed the Nvidia drivers and I did not have this issue. It only happens when I use the 64bit version with the Nvidia drivers.
On my notebook I had funny problem. I thought I had black screen two times and I had to shut down it with button. Shortly before I tried again and I barely saw some dark letters in the center of the screen so I pressed the button which brightens up the screen and now it works :)
Also, I've used http://sourceforge.net/projects/unetbootin/, maybe it helped.
UBUNTU 12.04 LTS install Problem, stuck/crash at loading screen. (Nvidia Graphics Cards)
- Remove Graphics Card from your machine.
- Install Ubuntu (You shouldn't get any errors when running generic driver.)
After Ubuntu is installed then put graphics card back into machine while the tower still powered off. Then swap your VGA/HDMI/DVI cables to the graphics card. Turn on computer and select the recovery option from the GRUB menu then boot normally.
Go to Nvidia website and grab the linux driver it will come in .run format
Install Driver using the following steps.
a. Move the driver to the desktop and rename it something simple and easy to remember.
c. Run command
sudo /etc/init.d/lightdm stop
d. Run command
cd ~/Desktop DRIVERNAME.run(What you renamed it.)
e. Run command
chmod +x DRIVERNAME.run
f. Run command
g. Follow instructions and continue
TIP If you still get hung with "It appears you are running "X" server then change the command on Step 5c to as follows:
sudo /etc/init.d/gdm stop
Ctrl+Alt+t (Only use steps 7 & 8 if you can't access the restart button or see your screen.)
Run the command
sudo shutdown 0 -hif the other command fails.)
Boot computer and enjoy :) I suggest printing these instructions.
PS I am not sure if you can place graphics card into PC after installation this is the way it worked for me and I am passing it on. I might suggest trying to install the graphics card on step 1 with machine off first as it is much safer this way.
In case that the black screen is only intermittent (and that there might be a blinking cursor), lightgdm having a race condition and not being able to start properly could be the issue.
At least that was the case for me. See here for a solution: http://www.webupd8.org/2013/01/ubuntu-lightdm-black-screen-when-using.html (see also this bug report).
The gist of it: Use gdm and not lightgdm (i.e.
sudo apt-get install gdm, and choose
gdm as default login manager when asked).
Let me quickly describe the symptoms I had: At first, because also the graphics was having a problem, when this problem occurred, I would only see a blank screen, and no chance to switch to the other terminals by pressing Ctrl+Alt+F1-6 (the screen simply stayed completely black, or rather, a very dark purple or something).
This I fixed by adding the "nomodeset" kernel option, as stated e.g. in this answer.
But after that, I still intermittently couldn't boot up properly; now it would stop with a blinking cursor. And this, as the above link tells in more detail, is caused by lightdm having a race condition - which manifests itself mainly if the boot-up is very quick, e.g. from an SSD (as it is the case for me).
Hope this helps someone.
On my new laptop, this problem has turned up again (I guess it has something to do with the dual graphics cards, integrated intel and an nvidia 1050 card, between which I switch with prime-select). This time around, gdm wouldn't work properly at all in my XUbuntu installation, so I turned to another solution: As restarting lightdm from the command line after startup fixed the problem, I added a systemd service which waits 2 seconds and then restarts lightdm. Ping me if you need details on this.
I tried those methods as well -- no joy. Here's what did worked for me.
Here's the part that was useful:
Getting 1366x768 resolution
echo insmod 915resolution echo 915resolution 58 1366 768 32
chmod +x /etc/grub.d/01_915resolution
/etc/default/grub, assign value
1366x768x32 to variables
reboot, now you have 1366x768 resolution.
Actually, I already had this resolution. But the added commands got rid of the black screen on boot-up and the need to switch video modes.
This is specific to where the install fails and your installation behaves badly.
- You need access to the Recovery menu, if you don't then look for other options.
- Internet access with a DHCP-enabled network
- GUTS! Since this is a delicate process.
- Once you are in the recovery menu, select Activate Networking.
- Now select drop to a root shell.
mount -a. (In my case that bad was the installation that I was forced to do this)
- Verify that your network is up and running:
ping -c 2 18.104.22.168If this fails run
ping -c google.comif this fails run
- Now update your repositories:
- Install debsums:
apt-get install debsums
debsums -s. It will give you a list of packages that have problems. Take note of each.
- Now reinstall the packages that has problems:
apt-get --reinstall install packages.
- Update your grub just in case.
- Reboot and good luck.
After upgrading from 12.10 to 13.04, the login screen is black because brightness is set to lowest level (Intel Integrated Graphics)
I've noticed that it would be a brightness problem cause I've listened to the Ubuntu's default drum sound when booting for the first time after upgrading. Before I find this solution, I had to increase the brightness level to see anything on the screen.
For me, the solution came from this bug report at https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/1173059, by changing /etc/default/grub as root this way:
Make a backup file, so you'll be able to restore it, if this solution doesn't work:
sudo cp /etc/default/grub /etc/default/grub.original
To open the file with sudo you can use this in the Terminal, for example:
sudo gedit /etc/default/grub
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash acpi_osi=Linux acpi_backlight=vendor"
You'll need to run update-grub2 to apply the change:
That's it. After rebooting, it worked flawlessly for me (that is, my login screen has a normal brightness level).
Use the latest version of Ubuntu
Other answers here are very good, but one point that deserves emphasis is that, as a general rule, you should use the latest version of Ubuntu, or at least the latest long-term support (LTS) version. Ubuntu, like all OSes, relies on drivers for video (and other) hardware, and these drivers often lag hardware introduction. That is, if your computer was introduced as a new model in 2016, it may include hardware with no support in Ubuntu releases prior to that year. Even when hardware is supported, there may be bugs that might be fixed in a more recent release.
Ubuntu release numbers are dates -- 16.04 was released in April (04) of 2016; 16.10 was released in October (10) of 2016; 17.04 was released in April (04) of 2017; and so on. New versions of Ubuntu appear every six months.
The April releases in even-numbered years are long-term support (LTS) versions. Like other releases, these see kernel and X Window System version updates. Starting about three months after the initial LTS release, and every six months thereafter for about two years, new point releases appear for LTS releases. These point releases incorporate the kernel and X updates from the previous release. Thus, 16.04.1 uses a (slightly updated) kernel and X from 16.04; 16.04.2 borrows these components from 16.10; 16.04.3 will use these components from 17.04; and so on. A total of five point releases are made available for LTS releases, after which point you'll need to upgrade to a newer Ubuntu release if you need new major versions. (There may be bug fixes within a kernel or X series, but not updates to new major versions.)
Thus, if you have particularly new hardware, you may need to use the very latest release, even if it's not an LTS release; or if you insist on an LTS release, you may need to use the latest point release in its sequence. For very new hardware, though, you may need to wait or track down bleeding-edge drivers that may be difficult to install.
Note that non-LTS releases are supported for just nine months; but LTS releases are supported for five years. Thus, it's generally best to stick with the latest LTS point-release, if possible; but you may need to use the latest non-LTS release, or even a beta version of an upcoming release, to get the very latest kernel and X Window System.
If your hardware is older, you may be able to use an older Ubuntu version; but as I've already noted, the newest version may incorporate bug fixes that can help resolve problems, even on older hardware.
situation 1) Impossible to solve at uefi-black-screen - uefi-bios has to be visible so you could then use an live-usb-stick properly. The accumulator has a chip too, this way the ransomware attacks via this chip...
a) pull out accumulator after machine is switched off (20 seconds pressing slightly the power-button off).
b) pull off the power-supply cable both ends.
c) after waiting only do use the power-supply cable
d) switch on and quickly two fingers changing pressing 'ESC' and 'F2'
e) disable recovery mode of uefi-settings
f) disable secure-boot
g) get sure that usb-booting order allows usb-live-stick at 1st place !!! save and reboot from bios (leave accumulator away)
h) re-install ubuntu or repair the disk with ubuntu-recovery-modes, but I would prefer complete new installation alongside previous ubuntu-installation and delete Windows. i) you need to delete Windows, because the chip of the accumulator wants to make a revival of ransomware for windows as previous first-boot-installation.
situation 2) Black screen with Windows on booting first hard-disk, then it is deadly sure ransomware:
... I had black screen today in the morning too, but my password was too long ... so the ransomware had no luck, despite of fact that I dont have dual-boot installation with windows here. There was only an buffer-overrun visible at my machine. I rebooted and the buffer-overrun (or underrun? - hybrid device here ...) was gone.
In case Jorge Castro has a dual-boot installation with Windows (no-matter-which-one) then I simply would press quick changing with two fingers the keys 'esc' and 'F2' to get back into uefi-bios before the machine boots into black screen after you have pressed the switch-off-button slightly for 20 seconds.
a) switch-off the machine with slightly pressing 20 seconds the power- supply-button
b) wait a while ... in case of notebook pull the plugin-cable of power- supply and stick it in again, then take out the accumulator and put it in again.
c) wait a while ... then switch-on your machine
d) now quick changing with two fingers between key 'F2' and 'ESC'
e) uefi-bios should appear again ... then this is not finish !
f like foolproof) decide the boot-mode whether to enable or disable secure-boot (http://www.eightforums.com/tutorials/17058-secure-boot-enable-disable-uefi.html). This decides whether you get back into windows to launch really http://malwarebytes.org - program.
g like gentleman) this is not gentleman-style but I saved your day.
When malwarebytes - program finished its job and avast too finished its job, use then a freeware registry-cleaner (http://heise.de)
situation 3.) This way ubuntu-recovery should work again after step 1.) and step 2.)
a) free the memory
b) repair the file-system
c) go into root - terminal
sudo apt-get install --reinstall grub* --fix-missing
sudo apt-get update
sudo passwd (new password like method here : http://nosy-x.com/password-001.html in paranoid method a suiting quantum-password)
happy coding and dont forget to eat a meal ...