Black Screen after installing fonts and reboot

Ha ha! :grin:
Mine has in my refind (no grub OS).

Oh. check out in your $esp partition, not in OS root partition.
You may need to mount that $esp first, unless it is mounted in fstab already.

Well there is a grub efi in:


Here’s mine… (in the $esp partition)

# It is automatically generated by grub-mkconfig using templates
# from /etc/grub.d and settings from /etc/default/grub

### BEGIN /etc/grub.d/00_header ###
insmod part_gpt
insmod part_msdos

function load_video {
  if [ x$feature_all_video_module = xy ]; then
    insmod all_video
    insmod efi_gop
    insmod efi_uga
    insmod ieee1275_fb
    insmod vbe
    insmod vga
    insmod video_bochs
    insmod video_cirrus

set menu_color_normal=light-gray/black
set menu_color_highlight=green/black

if [ x$feature_default_font_path = xy ] ; then
insmod part_gpt
insmod ext2
set root='hd2,gpt11'
if [ x$feature_platform_search_hint = xy ]; then
  search --no-floppy --fs-uuid --set=root --hint-bios=hd2,gpt11 --hint-efi=hd2,gpt11 --hint-baremetal=ahci2,gpt11  c69d25c9-73a6-4a1e-a586-805d060cc822
  search --no-floppy --fs-uuid --set=root c69d25c9-73a6-4a1e-a586-805d060cc822

if loadfont $font ; then
  set gfxmode=auto
  insmod gfxterm
  set locale_dir=$prefix/locale
  set lang=en_US
  insmod gettext
terminal_input console
terminal_output gfxterm
insmod part_gpt
insmod ext2
set root='hd2,gpt11'
if [ x$feature_platform_search_hint = xy ]; then
  search --no-floppy --fs-uuid --set=root --hint-bios=hd2,gpt11 --hint-efi=hd2,gpt11 --hint-baremetal=ahci2,gpt11  c69d25c9-73a6-4a1e-a586-805d060cc822
  search --no-floppy --fs-uuid --set=root c69d25c9-73a6-4a1e-a586-805d060cc822

insmod png
background_image -m stretch /usr/share/grub/background.png
if [ x$feature_timeout_style = xy ] ; then
  set timeout_style=menu
  set timeout=5
# Fallback normal timeout code in case the timeout_style feature is
# unavailable.
  set timeout=5

### END /etc/grub.d/00_header ###

Typing ahead because I need to be away for a few hours.

See this part

if loadfont $font ; then
  set gfxmode=auto
  insmod gfxterm
  set locale_dir=$prefix/locale
  set lang=en_US
  insmod gettext

Or any part in you grub.cfg with " set gfxmode=auto"
Change this to
set gfxmode=1024x768
and also add one line…
set gfxpayload=1024x768

BTW, is your lang

set lang=en_US
or others?
change to ‘us’ because you cannot have lang due to no grub directory. and therefore no lang directory.

oh, you replied. but if you really do not have any grub.cfg… let us know.
I’ll come back in a few hours.

I believe $esp is already mounted to /boot/efi.

The only grub thing I have is: /boot/efi/EFI/manjaro_grub/grubx64.efi

I just tested this in a VM, and the same thing happens, but the black screen only lasts like 10 seconds, compared to 30 secs in an actual installation.

Steps to reproduce:

  1. Use architect installer
  2. Create partitions (gpt) for / (ext4) & /boot/efi and mount them
  3. Install Kernel 4.14
  4. Install KDE minimal
  5. Install free drivers
  6. Install rEFInd
  7. Generate fstab (device UUID), host name, set locale (en_US.UTF-8), keyboard layout (us), time/clock (america/los angeles, utc), root pass, add user.
  8. Exit and boot into Manjaro
  9. Log in
  10. Install noto-fonts-cjk
  11. Reboot

Edit: Replicated issue installing via Manjaro KDE live disk on a VM, which installs grub not rEFInd.

Edit 2: Replicated issue installing via Manjaro GNOME live disk on a VM w/ grub

Edit 3: Replicated issue installing LXQT minimal via architect on a VM w/ rEFInd.

Edit 4: Kinda replicated issue installing bspwn via architect. Instead of a black screen, you can log in, it’ll show the wallpaper, but it’ll take forever to show the top panel/bar, after it pops up, it takes forever to actually be able to do anything. Also… when booting in after installing noto-fonts-cjk, it always says this the first restart and then it boots in fine:


I am really just doing these tests while playing around with trying out new DEs lol (minus the KDE one)

Edit: Actually… the /boot/efi/EFI/manjaro_grub/grubx64.efi I have was an old grub that I used to use when I used Manjaro with grub. I deleted it.

Okay, understand.
I’ve double checked my /boot/efi/refind/refind.conf because I’ve made many changes to it.
In it was this line I’ve changed or uncommented.

Somewhere around line 144, I’ve uncommented

#resolution 1920x1080
resolution 1024 768
#resolution 1440 900
#resolution 3

Try uncommenting resolution 1024 768 and see if this helps.
Note I have to do same (gfxmode/gfxpayload) in grub too for all uefi grubs
(systemd-boot use text).

Here’s my screen shot

ps: don’t know why, upload completed, but no shots. :relaxed:

As I said, this happens when using only grub too. I don’t think it’s a rEFInd or grub issue. As grub is just for booting into the system. It happens after booting in (that’s why you can see the cursor).

Yes, it happens with grub too. That’s why I said I’ve to change grub resolution too.
And it happens after pressing ‘boot’ in grub menu or in refind menu but before the OS boots up.
And there’s many post on this, here’s one recently, on grub though.

You can always try it, we won’t know if you don’t try it out, won’t we?

another one

Luckily, it has the same effects if I just sudo pacman -Rns noto-fonts-cjk, reboot, sudo pacman -S noto-fonts-cjk, reboot.


Uncommented the resolution 1024 768 line, reboot, installed noto-fonts-cjk, then reboot again. Same effects.

KDE grub



reboot, installed noto-fonts-cjk, then reboot again. Same effects.

I think it’s mainly black screening for a while because it has to load so many fonts for the first time at first reboot. The noto-fonts-cjk package is pretty big.

Ok, noted. thanks for trying. Appreciated.
But note I do also have noto-fonts-cjk (in some OS’s) as well as additional wqy fonts (chinese).
Since I’m here, here’s the screen shots.!



oops, still fail…

That grub looks NICE!!!

I actually have never themed my rEFInd (or grub in the past when I used it). I should try it sometime.

1 Like

Just for the record, it needs sudo.
I remember always in xorg.0.log the notice about fonts dir (or something) not found, but won’t do so long timeout.
You could check your xorg log to see any interesting info.

Ended up theming lol

Good to know, though I still don’t have the grub.cfg anyway haha.

1 Like

@realmain, can you provide the output of your terminal command ?

dmesg | grep microcode

I noticed your screen shot shows vmlinuz-4.14.xxxxxxx instead of grubx64.efi.
Is there another entry with grubx64.efi and you’ve chosen to hide it? If yes, it can be
untagged at that ‘recycled’ symbol at bottom.

If there is, can you try to boot this grubx64.efi ?
Note you said you have that grubx64.efi.


I installed Manjaro via Architect, and chose rEFInd, so I don’t believe it installs grub on top of that. rEFInd searches for my kernels installed, and allows me to boot into it.

I’ve clicked on that yesterday, it shows that nothing is hidden.

I found out that the grubx64.efi I had was from an old Manjaro installation that I have never removed. When I select it, it goes into grub rescue. I went in and manually deleted it.

I can run that at home later tonight after work.

Edit: Ended up not going home last night, so… I’ll do it tonight lol

Okay, thanks, I’ll wait for your dmesg output. Going to sleep now.

~ >>> dmesg | grep microcode                                                                                                                  
[    0.735447] microcode: CPU0: patch_level=0x08001129
[    0.735451] microcode: CPU1: patch_level=0x08001129
[    0.735668] microcode: CPU2: patch_level=0x08001129
[    0.735680] microcode: CPU3: patch_level=0x08001129
[    0.735701] microcode: CPU4: patch_level=0x08001129
[    0.735713] microcode: CPU5: patch_level=0x08001129
[    0.735735] microcode: CPU6: patch_level=0x08001129
[    0.735748] microcode: CPU7: patch_level=0x08001129
[    0.735770] microcode: CPU8: patch_level=0x08001129
[    0.735783] microcode: CPU9: patch_level=0x08001129
[    0.735805] microcode: CPU10: patch_level=0x08001129
[    0.735818] microcode: CPU11: patch_level=0x08001129
[    0.735840] microcode: CPU12: patch_level=0x08001129
[    0.735852] microcode: CPU13: patch_level=0x08001129
[    0.735866] microcode: CPU14: patch_level=0x08001129
[    0.735878] microcode: CPU15: patch_level=0x08001129
[    0.735917] microcode: Microcode Update Driver: v2.2.

Is this the complete output?

Also, you may want to check the difference when you booted up with grub.
I know you don’t have grub, but you can boot up using the manjaro install media as follows…

  1. Take the manjaro livecd and start it up in uefi.
  2. Do not boot up to live OS, but press ‘c’ at the menu and we’ll get to the grub prompt (grub>).
  3. At grub prompt, check output of
    grub> echo $grub_platform
    If output is “pc”, reboot the livecd media, this time making sure it is booted up in uefi.
    If output is “efi”, continue below.
  4. grub> insmod fat
    grub> insmod ext2
    grub> search -f /intel-ucode.img --set=root
    grub> search -f /etc/mkinitcpio.conf --set=pqr
    grub> probe -u $pqr --set=xyz
    grub> linux /vmlinuz-4.14-x86_64 root=UUID=$xyz rw
    grub> initrd /intel-ucode.img /initramfs-4.14-x86_64.img
    grub> boot
  5. When booted, at terminal
    dmesg | grep microcode
  6. Compare the difference between this and the one booted from rEFInd.

You may want to reconsider how you boot up considering the concerns on Spectre and Meltdown.
Of course, if we have the complete output, we won’t have to reboot in grub. But that is entirely up to you if you want to guard against possible security issues or just plainly have a proper cpu firmware.



I use an AMD CPU, so this does not exist.

Ok, lucky you!! :grin:


This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.

Forum kindly sponsored by Bytemark