[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: LVM should not be used by default with FC3

On Thu, 2004-10-07 at 01:36, Nathan Grennan wrote:
>   I have been heard from multiple sources that LVM will be used by
> anaconda when using automatic partitioning. I think this is a very bad
> idea. It breaks the KISS(Keep it simple stupid) principle. I have
> personally used LVM in the past and currently use software raid. Both
> are powerful and wonderful features, but in most cases are overkill,
> make things more complex, and make life hell when something like grub or
> init scripts break.

How exactly would LVM make things complex? Shuffling data around when
you need to have more space than your physical partition can hold is
complex, having your system setup in a way that you can easily(*) extend
filesystems or move them from physical disks on the verge of breaking to
new disks(**) is not.

(*) e.g. with LVM1, when /scratch is to small:

umount /dev/vg00/lv_scratch
e2fsadm.lvm1 -L +5G /dev/vg00/lv_scratch
mount /dev/vg00/lv_scratch

or with LVM2 (we really need to have e2fsadm.lvm2 or better yet a script
that can detect whether t operates on LVM1 or LVM2):

umount /dev/vg00/lv_scratch
lvm lvextend -L +5G /dev/vg00/lv_scratch
resize2fs /dev/vg00/lv_scratch
mount /dev/vg00/lv_scratch

or with FC3 kernels and tools even online:

lvm lvextend -L +5G /dev/vg00/lv_scratch
ext2online /dev/vg00/lv_scratch

(**) e.g. smartd warns you that /dev/sdb which contains the physical
volume /dev/sdb1 is about to fail, so you do simply:

plug in a new disk (say /dev/sdc)
create first partition of type 8e (Linux LVM)
lvm pvcreate /dev/sdc1
lvm vgextend vg00 /dev/sdc1
lvm pvmove /dev/sdb1 /dev/sdc1

This all is simple (albeit not stupid ;-).

     Nils Philippsen    /    Red Hat    /    nphilipp redhat com
"They that can give up essential liberty to obtain a little temporary
 safety deserve neither liberty nor safety."     -- B. Franklin, 1759
 PGP fingerprint:  C4A8 9474 5C4C ADE3 2B8F  656D 47D8 9B65 6951 3011

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]