blob: b6c767806cad51e838c1c43574625e972ea0d54e (
plain) (
tree)
|
|
<!DOCTYPE html>
<html dir="ltr" lang="en">
<head>
<meta charset='utf-8'>
<title>LVM</title>
</head>
<body>
<a href="index.html">Tools Index</a>
<h1>LVM</h1>
<p>Read <a href="https://raid.wiki.kernel.org/index.php/RAID_setup">Raid Setup</a>,
the only thing you will need outside system is:
"Patience, Pizza, and your favorite caffeinated beverage.".
<a href="https://wiki.archlinux.org/index.php/Software_RAID_and_LVM">Arch Wiki</a>
article about Sofware RAID and LVM.</p>
<p>LVM or Logic Volume Manager bring one more layer, read
<a href="http://www.tuxradar.com/content/lvm-made-easy">Lvm made easy</a>.
Partitions under lvm are easy to be resized, moved and there is
a tool to help encrypt. There is more freedom to name physical
disk names exp; production, development, backups...</p>
<p>Basic idea behind RAID is to deal with independent disks
as an array of drives. Raid 0 uses two or more disks as one,
with performance gains without fault-tolerance. From raid 1
to 6 they offer diferent fault tolerance mechanisms.</p>
<p>Until now "from install" there is only one partition,
it is good idea to have a system with diferent partitions for each
propos. If is a "fresh install";</p>
<pre>
# cd /iso/crux/opt/
# pkgadd lvm2#2.02.107-1.pkg.tar.xz
#
</pre>
<h2 id="lvmpart">1. LVM Partition</h2>
<p>There is no need to create a partition with fdisk or parted
if all device will be used for lvm, just <a href="#pv">pvcreate</a>
against the device (pvcreate /dev/sda).</p>
<p>Create a LVM partition with parted;</p>
<pre>
parted --script ${DEV} \
unit mib \
mkpart primary 1000 4000 \
set 1 lvm on
</pre>
<h2 id="pv">2. Create physical volume</h2>
<pre>
# pvcreate /dev/sdb3
Physical volume "/dev/sdb3" successfully created
</pre>
<h2 id="vg">3. Create volume group</h2>
<pre>
# vgcreate vg_system /dev/sdb3
Volume group "vg_system" successfully created
# vgcreate homevg /dev/sdb4
Volume group "homevg" successfully created
#
</pre>
<h3>3.1. Search Volume Groups</h3>
<pre>
# vgscan
Reading all physical volumes. This may take a while...
Found volume group "homevg" using metadata type lvm2
Found volume group "vg_system" using metadata type lvm2
#
</pre>
<h2 id="lv">4. Create logical volume</h2>
<pre>
# lvcreate -L 15G -n distfileslv vg_system
Logical volume "distfileslv" created.
# lvcreate -L 8G -n packageslv vg_system
Logical volume "packageslv" created.
# lvcreate -L 4G -n swaplv vg_system
Logical volume "swaplv" created.
# lvcreate -L 80G -n homelv homevg
Logical volume "homelv" created.
#
</pre>
<pre>
# mkfs.ext4 /dev/vg_system/distfileslv
# mkfs.ext4 /dev/vg_system/packageslv
# mkswap /dev/vg_system/swaplv
# mkfs.ext4 /dev/homevg/homelv
</pre>
<h3>4.1. Activate Deactivate</h3>
<p>Deactivate logical volumes;</p>
<pre>
# lvchange -a -n /dev/vg_system/packageslv
# lvchange -a -n /dev/vg_system/distfileslv
# swapoff /dev/vg_system/sawplv
# lvchange -a -n /dev/vg_system/swaplv
</pre>
<p>Deactivate volume group;</p>
<pre>
# vgchange -a n vg_system
0 logical volume(s) in volume group "vg_system" now active
#
</pre>
<p>Activate volume group;</p>
<pre>
# vgchange -a y vg_system
3 logical volume(s) in volume group "vg_system" now active
#
</pre>
<h2 id="fsck">5. Maintenance</h2>
<h3 id="resize">Resize</h3>
<p>First umount all lvm partitions;</p>
<pre>
# pvs
</pre>
<pre>
# pvresize /dev/sdb
</pre>
<pre>
# vgs
</pre>
<pre>
# lvresize --resizefs --size +25GB /dev/mapper/vg_system-lv_ports
</pre>
<pre>
# vgs
</pre>
<h2 id="encrypt">7. Encryption</h2>
<a href="index.html">Tools Index</a>
<p>
This is part of the c9-doc Manual.
Copyright (C) 2016
c9 team.
See the file <a href="../fdl-1.3-standalone.html">Gnu Free Documentation License</a>
for copying conditions.</p>
</body>
</html>
|