I have a desktop that’s my main machine. The unit came with a 1TB HD with Win 7. This part of the system is BIOS booted from an MBR HD. I installed a second 1TB drive for a Linux/BSD multi-boot system which I hope to populate with Opensuse 13.2, Centos, Mint, and FreeBSD. The second HD is configured as a GPT drive via Gparted. I’ve demonstrated that I can select either HD at boot up. I want to have GRUB2 control the selection of the OS from a 1MiB boot partition located on HD2. I’ve read that I should set up /, home, and swap partitions for Opensuse to occupy and then install from the live DVD. Afterward I would install GRUB2 to the boot partition and configure it to select Suse. Then repeat the install / configure routine till all my systems are loaded. My first question is will having multiple partitions with the same volume label cause a problem? Like having a “home” partition for Suse and Mint. Will I have to provide a unique volume label like s_home and m_home or does it matter? Also what about having separate root partitions ////? Ive
read that the fsck file in GRUB2 keeps the partitions aligned with the proper OS by using the GRUB partition labeling system so the volume label is meaningless; is this true? Anyway I’d truly appreciate any knowledgeable input about these concerns and any additional issues I haven’t considered yet before I start this effort. I like to have everything mapped out before I start typing my way into a disaster. TIA
it is not a “must” to add volume labels to your file systems. Most people do not have them. At least not on their system disks. But when you see advantages in having them, they should of course be a unique as possible. Were it only to avoid confusion in your own mind.
I do not understand this at all. What is an “fsck file in GRUB2”?
I only know fsck as a tool (in fact a group of tools, each equiped to handle it’s own type of file system) with which one can check and hopefully repair file systems.
You can not have two volumes with the same name at the same time on a system. That is a name clash Use unique volume names or none at all
fsck is run but not by grub. It is part of start up but that is after grub as done it’s thing
My bad! The file that GRUB2 accesses is the fstab file that is installed and configured by YAST when Opensuse installs. The file fstab contains the mount points for the OS. After much reading last night (I found a complete primer (https://en.opensuse.org SDB: Basics of partitions, filesystems, mount points) which cleared up everything. In fstab the partition physical locations on the drive are identified and you can label them howsoever you wish because that particular fstab entry is unique to and only used by the distribution that created and inspected it when GRUB2 initiated the boot sequence for the distribution. I encourage anyone else to read the primer I identified if they really want or need to understand the mechanisms that make multi-boot systems possible. I have a lot more reading to do because this partitioning requires juggling 3 separate items: GRUB2, the distribution install routine (in opensuse its YAST) and the partitioning tool (in my case GParted).
After reading the above primer I can see why some of the other things I’ve seen written suggested that you partition the disk for only opensuse at first (leaving the rest of the disk space undeclared) and let YAST do its stuff. Then you install GRUB2 to allow it to select the boot entry for opensuse . This may require manually editing the opensuse fstab to accomplish this (I’m not sure yet). Once the system can boot opensuse from GRUB2 as a selection and it actually works then you use GParted to create the next distributions’ partitions from the remaining undeclared space on the drive. And so on.
Like I said I’ve got some more research to do but when I’m done I will write it all up for the next person. I think that the biggest hurdle was re-learning what partitioning means to a Linux / Unix system and forgetting the distortion that is the Microsoft way.
Your still a little off the mark. grub has nothing to do with the fstab. at the grub stage all it has to know is where the kernel is then start the initrd process to load the kernel all is run in a tmp file system until the kernel is up and running and the initrd process continue starting process and mounting partition and moving over to the normal root file system. So grub is long gone when the fstab come ito play.
fstab only says what partitions get mounted at boot.It has nothing at all to do with chaining to other OS’s What is in the menu can be manually adjusted or best to leave yast to do it.
That is in fact not true. Linux uses the same way of partitioning as MS-DOS does. Else multi boot would be impossible.
And Unix in general does not fit in here, because Unix systems have their own way of making partitions (often called slices in Unix) on their own hardware. Only Unix systems on PC hardware might use MS-DOS partitioning.
Read this: https://en.opensuse.org/SDB%3ABasics_of_partitions,_filesystems,_mount_points