This is mainly for if you've done something silly like trying to clone a Live, running VM image. In this example, the VM initially finds grub and tries to boot but is kicked straight into initramfs rescue mode/busybox right after this.
If you've done this "silly" thing, you could have dataloss but a lot of times just using fsck will fix it as you are guaranteed at best to have some corruption and inconsistencies in the filesystem. My theory is that some files wer........
Many times just doing an update-grub may find Windows and add it to grub, but a lot of times it won't.
Create Windows in a custom grub entry like below:
sudo vi /etc/grub.d/40_custom
menuentry "Windows 10" {
set root='(hd0,0)'
chainloader + 1
}
Change Windows 10 to whatever you want to call it. For example if it was for Windows 11 you'd probably want to ca........
This sometimes happens when trying to install the EFIversion of grub to a device when you are booted into Legacy/MBR mode. It doesn't seem to occur on all machines, but some and seems somewhat BIOS dependent.
grub-install --target=x86_64-efi /dev/sda
Installing for x86_64-efi platform.
grub-install.real: warning: Couldn't find physical volume `(null)'. Some modules may be missing from core image..
grub-install.real: warning: Couldn't find physica........
Step 1.) Upgrade to Debian 11 first
The process to go to Debian 12 is not as smooth as 11, when trying to upgrade from Debian 10. In fact, it doesn't work directly, so you'll first need to follow this guide to update to Debian 11, reboot and come back here if successful.
Step 2.) Update sources.list
Update your /etc/apt/sources.list like this:
deb http://........
Be very careful about what filename you specify in dhcpd.conf if you get an error like this:
NBP filesize is 0 Bytes PXE-E23:Client received TFTP error from server.
If you specify "BOOTx64.efi" then the file had better not be called "BOOTx64.EFI" as it is case sensitive. It's really a case of th........
This seems to have changed for RHEL 8 where a normal dracut to update your initramfs creates a system that only boots for the running kernel. For example if you have Kernel 5 and then chroot into a RHEL 8 variant which uses kernel 4.18, and run dracut, it seems that by default the system will be unbootable.
It is also the case that if you move your RAID array or drives to another server that it will be unbootable, because dracut seems to only include modules needed for the curre........
You can read lots of posts about this issue but there is not much information about why this is the case or how grub determines the root= device name. Some even suggest modifying grub.cfg manually which is a disaster as the next kernel update will cause grub to revert back to the device name.
For most people this won't be an issue but those using template system, automated deployments and working in embedded may run into this issue with custom embedded and created minimal kernel........
You probably didn't do an "update-grub" and grub no longer has any proper menu entries, but before you can fix it let's try to get grub booting anyway.
If you get this lovely black grub screen here's how you can get things booting.
In my case I have a gpt partition with partition 1 and 2. Partion 1 is just my EFI / ESPand partion 2 /dev/sda2 is my root which includes /boot.
You will have to adjust this if you had a separate /boot partition.........
1.) Create your EFI/ESP Partition
If you happen to have some free space on the drive already then this is easy, just create a new partition of at least 100M.
The nice thing about the EFI spec is that it must just be in the first 2.2TB of space so for most users, it means you can simply resize the last partition(downsize it by 100M) and then add an EFI partition at the end.
For example if you had this partition scheme:
/dev/sda1 = /........
I've seen this bizarrely happen on a newly partitioned and custom installed Linux install, particularly if you did not properly unmount before rebooting.
You can find reports of it happening on various
How to fix the error: invalid arch-independent ELF magic.
You need to boot into Live/Rescue mode, chroot into your OS properly and then do a grub-install on each drive that nee........
This article about migrating to a CentOS 7 /8 RAID mdadm array has a lot of info but I wanted to focus specifically on what newer versions of CentOS 7 require to boot mdadm and what changes are necessary on CentOS 7.8+
CentOS 7 / 8 mdadm RAID booting requirements
This assumes you are chrooting into an existing install or using it to get a new deployment ready. However, these steps can........
Just an FYI that the installer ignores your selection of Boot Loader, as it was intended for MBR/Legacy. The installer horribly, even when choosing "Something Else" and manually partitioning and creating an EFIin your install drive, will still install grub to the first EFI partition it finds, even if you are following a guide like this to avoid wiping out the M........
I've encountered this after upgrading some Debian/Ubuntu/Mint based systems for no explicable reason, although there are some bug trackers on Ubuntu that document this: https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1889509
The short end of the solution is that you need to properly reinstall grub.
1.) Boot from a LiveCD
2.) Mount your root / filesystem and don't forget to mo........
Usually if you get the grub boot loader and it doesn't show any boot options, it's because grub was not installed correctly and/or the partition that it is supposed to be on has changed or does not exist. It can also happen if you install Linux to one drive, but the boot loader to another by accident, whether EFI or MBR/Legacy mode.
You can normally fix your booting/bootloader/MBR/EFI it by chrooting into your root partition:
#become root
sudo su........
Just a quick note and warning is that if you are testing to see if EFIPXE booting works on a VM, MAKE SURE it actually works. For example Iinitially tested using my Distro's QEMU 2.5+dfsg-5ubuntu10.46 and ovmf BIOS firmware (OVMF supports EFI). However, I found on old versions of QEMU (like 2.5), EFIbooting with GRUB NEVER works so it may appear that you have made a mistake when everything is fine when you boot a physi........
Traditionally kernels were numbered starting from 0 but by default the "new style" of grub boot loading considers each subkernel item to be different so if you have 3 entries for 4.40-148 rather than counting for 1.
To get the expected behavior let's show this example and how we can boot it
We do a grep on menuentry in /boot/grub/grub.cfg to see all of the bootable kernels rather than scrolling through loads of extra entries we don't care about (thou........
This happens to a lot of Nvidia users especially users of newer cards like the RTX series. If for example you are trying to boot and install Linux and you get a black and white grub2 screen instead of a nice graphical welcome installer, you probably suffer from this bug. It is normally followed by the user booting and finding they just have a blank/black screen.
Here is the quick flow of steps to fix it:
If you get a black grub scree........
Most newer distros inexplicably cause your NIC to have what Icall "random" non-standard name conventions because of systemd.
This is a big problem for many people and especially those running servers. Imagine that you have a static IPconfigured for ens33 but then the hard disk is moved to a newer system, the NIC could be anything from ens33 to enp0s1, meaning that manual intervention is required to go and update the NIC config file (eg. /etc/network/interfa........
The reason for doing this is that the installer doesn't seem to work properly for LUKS and the server installer doesn't even support LUKS anymore. When you use the GUI install on Desktop for LUKS it won't boot and will just hang after you enter your password. So the only reliable way is to do it ourselves.
1.) Make a default minimal install of Ubuntu
2.) Have a secondary disk on the server or VM.
3.)........
The cool thing here is that we only need 1 drive to make a RAID 10 or RAID 1 array, we just tell the Linux mdadm utility that the other drive is "missing" and we can then add our original drive to the array after booting into our new RAID array.
Step#1 Install tools we need
yum -y install mdadm rsync
Step #2 Create your partitions on the drive that will be our RAID array
Here I assume it is /dev........
To enable amdgpu we have to set special kernel boot parameters. The easiest way is to make it permanent and apply to all kernels (no messing around with grub.cfg) so we'll edit those defaults in /etc/default/grub by changing the GRUB_CMDLINE_LINUX_DEFAULT parameter. After that don't forget to run "update-grub"to apply it (otherwise amdgpu will never be enabled).
The opensource driver for AMD is really quite impressive these days, as we show it works with ve........
I don't consider a lot of these "extra" kernel modules "nice to have" as they often contain drivers for essential items like your soundcard, your NIC and many other devices that may not work. Sometimes you may find that "sound" or "ethernet" worked before a kernel/OS upgrade and now in the new version they don't. Often it will be because you need to install the "extra" kernel modules.
One other weird thing is that sometimes........
The strange thing is that usually the first install or two will work on any new machine but then it suddenly won't. I had this experience on QEMU 2.13 on a different machine. There is something finicky or buggy about the CUCM installer even when choosing the same virtual hardware specs.
qemu-kvm command:
/usr/libexec/qemu-kvm -version
QEMU PC emulator version 0.12.1 (qemu-kvm-0.12.1.2-2.506.el6_10.1), Copyright (c) 2003-2008 Fabrice Bellard
........
chroot which stands for change root allows you to virtually operate in another operating system even though you haven't booted it. It is commonly used to deploy new distros, applications and to fix a broken Linux/Unix install or prep a new system image without having to physically boot the drive or disk.
So in this example let's say we have a drive that has a Linux OS installed on /dev/sdb1 and we have mounted this partition on /mnt/sdb1
The key point is to edit the &quo........
I don't recall having this issue in the distant past but nowadays at least Debian seems to be very picky about this.
I used dd to copy one hard drive to another and tried booting it. Everything seemed fine with grub working but each time it would drop to the busybox shell. There is no particular error so this is misleading.
Normally the first things you would check are to make sure your fstab is correct (that the UUID is correct)and that you've updated grub.&........
This guide will work for most modern Linux versions like Centos RHEL 7, Debian, Mint, Ubuntu etc...
In Centos 7 the days of editing the "kernel"line and adding "single"are gone. On top of that sometimes after a new install passwords do not work, maybe you forgot your password or for some other reason you need to break in or fix your system? It could also be because you can't mount your root / or some other /etc/fstab error and many other err........
When using the .sh script the rendering doesn't work after an upgrade of related packages to kdenlive.
kdenlive (kdenlive:amd64 (4:17.04.1+git201705191233~ubuntu16.04.1)) with affine or composite transitions was fine but is now broken during the time of transition it is just a white screen.
The previous version was fine:
kdenlive:amd64 (4:17.04.1+git201705191233~ubuntu16.04.1
But now I can't find it or install it:........
Done on Centos 7.3 very important as clearly based on older guides it was a lot easier and more simpler! Hint do not use grub2-install!
If you have trouble booting after this check this CentOS mdadm RAID booting/fixing guide.
One huge caveat if you are an oldschool user or sysadmin who has avoided UEFIbooting
The nor........
1.) Replicate the number of partitions in your new drives.
gdisk /dev/sda
gdisk /dev/sdb
I created 3 partitions of the same same size.
partition #1: +1G (/boot)
partition #2: +60G (swap)
partition #3: rest of it (/)
#note if you are using GPT/gdisk you need to create separate a partition at least 1MB in size (in my case I would a 4th partition and mark it type ef02).........
mdadm won't boot in Ubuntu/Mint/Debian anymore.
You just get the following in a loop:
mdadm: CREATE group disk not found
Incrementally started RAID arrays.
Incrementally starting RAID arrays...
mdadm: CREATE group disk not found
Incrementally started RAID arrays.
Incrementally starting RAID arrays...
mdadm: CREATE group disk not found
Incrementally started RAID arrays.
Incrementally starting RAID arrays...
mdadm: CREATE group dis........
vzkernel-2.6.32-042stab116.2.x86_64.rpm&n........
This happened while an mdadm array was syncing, all access from writing a new blank file to opening a small .txt file was very slow:
[222117.312078] kjournald starting. Commit interval 5 seconds
[222117.685060] EXT3-fs (md0): using internal journal
[222117.685096] EXT3-fs (md0): mounted filesystem with ordered data mode
[222122.376847] kjournald starting. Commit interval 5 seconds
[222122.602825] EXT3-fs (md2): using internal jour........
It freezes on"grub> setup (hd1)
Checking if "/boot/grub/stage1" exists..."
Basically it looks like the partiion table is corrupt or the data/files are not there.........
grub> root (hd0,0)
root (hd0,0)
Filesystem type is ext2fs, partition type 0xfd
grub> setup (hd0)
setup (hd0)
But if you do:
root (hd1,0)
setup (hd1)
it does work, I think hd0/sda had a GPT partition that was not removed properly (what I did was just dd bs=512 count=1 the partition table from another drive since the partition table should be identical).
Checking if "/boot/grub/........
In this example we have 2 drives in a RAID array and /dev/sdb is the one that failed. /dev/sda1 is also the /boot partition which we tell grub to install on /dev/sdb eg install root (hd0,0) /dev/sda1 on the new drive /dev/sdb (hd1)
First copy the partition table from /dev/sda to /dev/sdb
dd if=/dev/sda of=/dev/sdb bs=512 count=1
Run partprobe to detect the new partition table
partprobe........
Here is the scenario you or a client have a remote machine that was installed as a standard/default minimal Centos 6.x machine on a single disk with LVM for whatever reason. Often many people do not know how to install it to a RAID array so it is common to have this problem and why reinstall if you don't need to? In some cases on a remote system you can't easily reinstall without physical or KVM access.
So in this case you add a second physical or disk or already ha........
Iwas surprised to see that Linux Mint at the latest 17.2 version still has NO mdadm installer option, and worse the installer will not be able to create a proper booting environment even when you do install it.
How to setup mdadm in Linux mint LiveCD
sudo su
apt-get install mdadm
# partition as you need and then create your mdadm devices
# create your SWAP md0
mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sda1 /d........
root (hd2,1)
Filesystem type unknown, partition type 0x83
grub> root (hd2,2)
root (hd2,2)
Filesystem type is ext2fs, partition type 0x83
grub> setup (hd2)
setup (hd2)
Checking if "/boot/grub/stage1" exists... no
Checking if "/grub/stage1" exists... no
#weird thing about grub is that the drive you enter is considered hd0
For example when booted fu........
I messed up the bootloader by accident on a standard Centos 6.3 install because I turned the /dev/vda1 boot partition into an mdadm raid 1. This was all done correctly aside from one point Ididn't realize was an issue metadata=00.90 is the only thing that will allow you to boot (otherwise grub won't work and you won't boot).
So the next step is rescue mode from a CD right? The problem you will find is that grub does not detect your hard drives, this is Ibelieve is be........
#count=10000 makes an image of 10000MB make sure your image is at least the same as your existing
dd if=/dev/zero of=yourimage.img bs=1M count=10000
# losetup -fv newimage.raw
# fdisk -cu /dev/loop0
# kpartx -a /dev/loop0
# dd if= of=/dev/mapper/loop0p1
# e2fsck -f /dev/mapper/loop0p1
# resize2fs /dev/mapper/loop0p1
# a lot of guides tell you to edit /etc/fst........
./configure
./configure: line 91: cd: /lib/modules/2.6.32-042stab084.25/build: No such file or directory
Error: kernel version not found.
Please make sure your kernel is configured.
dr-xr-xr-x. 4 root root 4096 Feb 21 06:13 ..
lrwxrwxrwx 1 root root 45 Feb 21 06:13 build -> ../../../usr/src/kernels/2.6.32-042stab084.25
drwxr-xr-x 2 root root 4096 Feb 12 20........
mount -o bind /proc /sda2/proc
mount -o bind /dev/ /sda2/dev
mount -o bind /sys /sda2/sys
chroot /sda2
mint / # mount -o bind /proc /sda2/proc
mint / # mount -o bind /dev/ /sda2/dev
mint / # mount -o bind /sys /sda2/sys
mint / # chroot /sda2
mint / # cd ~
mint ~ # ls
Desktop
mint ~ # cd /
mint / # ls
bin Desktop dev-temp home&nb........
This is a great way to upgrade your RAID array or move it/copy it to a new set of hard drives.
Eg. you have a current RAID 1 array on older/slower drives.
Just add at least 1 of the new drives to the array, update grub/install it and then boot into it. Then you have a transparent data migration that is fully synchronized.
mdadm --grow /dev/md126 --raid-devices 3
md127 : active raid1 sdc1........
mdadm --manage /dev/md3 --add /dev/sda1
cat /proc/mdstat
Personalities : [raid1] [linear] [multipath] [raid0] [raid6] [raid5] [raid4] [raid10]
md0 : inactive sdd2[1] sdd1[2](S)
31270272 blocks
md3 : active raid1 sda1[2] sdb1[1] sdc1[3](F)
943730240 blocks [2/1] [_U]
[>....................]........
cat | grub --device-map=/dev/null
Now pay close to the attention of the beginning.
Type: "device (hd0) VPS.img" this is telling what hd0 will be to GRUB and we're telling it the disk image file "VPS.img" in the current directory is hd0, you can specify alternate paths and image names of course.
GNU GRUB version 0.97 (640K lower / 3072K upper memory)
[ Minimal BASH-li........
This booting error is because the Xen PV guest image uses the Xen kernel, this is not compatible with anything but a host running a Xen kernel.
I did a kpartx -av virtual.img and then it created some partitions that showed up in fdisk.
I mounted it and did a chroot into it and removed the xen kernel and installed a normal kernel but Xen still shows the same kernel in Grub (only the Xen one).
This is strange but it seems like this Xen PV guest has some sort of hidden or........
One of my test Centos 5 containers was on a partition that filled up and it threw all sorts of errors and stopped responding but now I can't boot it again anymore.
All the console shows is the Linux Penguin on the top left corner and the xm console says "usbcore: registered new driver hub" and has halted there.
Centos 5 Xen container stuck/frozen won't boot on "usbcore: registered new driver hub"
Another great way of troubleshooting is booting fro........
ifup eth0
SIOCSIFFLAGS: Cannot allocate memory
Failed to bring up eth0
This is on a Centos 5.5 machine with OpenVZ kernel 2.6.18-238.12.1.el5.028stab091.1PAE, I updated 3 systems with the same hardware to the same kernel and for some reason the network didn't come back.
This only happened once the first time I booted into the new kernel and hasn't happened again but this is a serious issue for those running systems from remote locations such as a datacente........
Warning: Unable to open /dev/fd0 read-write (Read-only file system). /dev/fd0 has been opened read-only.
That's a very annoying error, it's simply because Centos for some reason thinks it's wise to load the "floppy" kernel module, who has a floppy drive? I haven't seen or used a floppy for over 12 years!
It's more than just annoying, if you probe the drives attached to your system, eg. with grub or partprobe, it keeps trying to locate a flopp........
GNU GRUB version 0.97 (640K lower / 3072K upper memory)
[ Minimal BASH-like line editing is supported. For the first word, TAB
lists possible command completions. Anywhere else TAB lists the possible
completions of a device/filename.]
grub> root (hd1,0)
Filesystem type is ext2fs, partition type 0xfd
grub> setup........
I wanted to Import/Use a .vmdk hard disk image file from VMWare. Generally you can just "point" VirtualBox to it and use it and it will work but I found an exception.
One of my Centos 4.4 x64 images wouldn't boot. I had two copies, an older one and the newer one. The older one booted as normal (once I changed the VirtualBox driver to IDE from SATA). The newer one stopped at the "GRUB loading" message no matter what I tried.
As far as........
Moving to RAID was a pain.
What you have to do is the following from an existing install:
Install mdadm
Create your mdadm RAID 1 array on your spare hard drive.
Start it with the missing disk.
rsync the entire contents of your current / to the md partition.
Here's a good way of doing it:
rsync -Pha --exclude=/proc/* --exclude=/sys/* --exclude=/mnt/* /. /mnt/md2........
This is a "pro" feature I guess so you don't have to see the ugly GRUB screen but if you're having boot problems or trying to update/customize your system it can be frustrating.You have to do some Googling to find out that you have to hold shift to see the GRUB2 menu.
You have to hold shift before GRUB2 loads and you will see it........
had trouble trying to revert Ubuntu 10.04 LTS from grub2, won't boot mdraid and did not even install mdadm during the installation!
I have tried moving back to GRUB 0.97
backed up original /boot and then copied /boot from an old Debian install. Modified device.map and menu.lst and put the appropriate kernels and initrd for Ubuntu back in /boot
I ran grub:
root (hd0,1)
grub> setup (hd0)
Checking if "/boot/grub/stage1........
Why would you want to downgrade the superblock? Old mdadm verisons like mdadm 2.5.6 only use the 0.90 superblock/metadata and new versions use 1,1.0,1.1 and 1.2 superblocks by default.
There are some annoying caveats with this, first of all the new superblocks (later than 0.90) CANNOT be read by GRUB, so you won't even be able to install GRUB. Even worse, old versions of mdadm CANNOT automatically detect arrays even if they were created with a new version of mdadm with th........
I successfully created a single RAID 1 partition which includes /boot inside it and my root directory through the Debian installer. It said GRUB installed successfully but when I try booting the OS it seems GRUB can't read anything.
When trying to boot from GRUB
GRUB Loading stage 1.5.
GRUB loading, please wait...
Error 2
I get "Error 2" when trying to boot Debian. I also notice from a LiveCD that........
GRUB Boot FreeBSD[code:1:a7050277b7]title FreeBSD 6.0
root (hd0,2,a)
kernel /boot/loader[/code:1:a7050277b7]
Just insert the above code into GRUB's [b:a7050277b7]menu.lst[/b:a7050277b7] and make sure you set the root path correctly.
[quote:a7050277b7]Let's quickly explain what the root (hd0,2,a) means:
*hd0 stands for the primary master drive
*2 stands for partition 3
*a stands for the........
Nice General Linux RAID 1 GuideFull examples/tutorials that should work for any Linux system using GRUB or LILO as the boot loader.
This is the only tutorial I've seen that clearly shows how you can convert an existing non-RAID system to software RAID1 remotely, without ever having to be at the computer. This is important for people who co-locate or rent dedicated servers that they may not have physical access to in a timely manner.
https://alioth.debia........
You can see the problem below, GRUB recognizes my hard drive and sees the partitions but cannot access them. This is from the GRUB boot disc I'm using.
What happened is that I had some power issues causing this system's power to be interrupted several times and basically an on/off on/off situation.
The system won't boot, I just get a flashing cursor and no message or error from GRUB. Obviously the problem is that my MBR seems corrupt or some other issue.
My........