As part of my learning, I will try to explain to you how to install KVM on physical computer and server. That is the first stop on the road to install the OpenStack.
- Do not use GUI*
- KVM cluster
- All VM stored on /mapper/red01(Raid5 matrix)
- Two networks: NAT and Bridge to physical interface
- As much automation as possible
Note:
* Usually, I work with servers I have no real access. Also, my connection to the internet is slow, so it is not possible to transfer the graphic interface.
- Ansible
- Linux networking
- libvirt
- CPU: 2x Intel(R) Xeon(R) CPU E5645 @ 2.40GHz
- RAM: 128 GB
- HDD:
- SSD 128GB
- HDD RAID1 1TB
- HDD RAID5 8TB
- HDD RAID5 8TB
I follow my convention of naming. I have a few servers called by dragon names like Farnir, Smaug, etc. That name becomes the environment for all running machines and programs. You should find your standards and follow them. Host name contains following parts:
- Environment name
- Destiny / Use case
- System version (optional)
An exception is the base image, which is base for all next virtual machines.
Examples:
- Domain:
kvm-smaug,kvm-farnir - Host names:
smaug-mysql,
-
Before you start installing the KVM soft, check if your processor has support for virtualization. To check your virtualization capabilities, you need to run the following command:
egrep -c '(svm|vmx)' /proc/cpuinfo -
If you can see 0, that means, your computer does not support virtualization. If you can see any number larger than 0, it's valid. If your computer is not supporting virtualization, try to enable it in BIOS settings. To do it restart your computer, find the "Virtualization (VT-x/AMD-V)" option and turn it on.
-
Now you can verify if your system is correctly set up. To do it run: the
kvm-okcommand. Correct output is:# kvm-ok INFO: /dev/kvm exists KVM acceleration can be used -
Now You can install quemu and all required software. To do it run:
apt-get install qemu-kvm libvirt-bin bridge-utils virt-manager -
The KVM uses virtualization extension like Intel VT for Intel processors or AMD-V for AMD processors. Let's see what modules are enabled for you:
# lsmod | grep kvm kvm_intel 212992 34 kvm 598016 1 kvm_intel irqbypass 16384 20 kvm -
If your module are disabled, please run
modprobe --all kvm -
Verify installation:
# virsh list --all Id Name State ---------------------------------- -
It's done!
/etc/libvirt/
└── qemu.conf
/mapper/red01/kvm
├── disks/
└── images/
/
├── kvm-disks -> /mapper/red01/kvm/disks/
└── kvm-images -> /mapper/red01/kvm/images/
/var/lib/libvirt/
└── networks
└── host-bridge.xml
Open the file: /etc/libvirt/qemu.conf and change the following values:
user = "root"
group = "root"
security_driver = "none"
First two lines set the qemu user and group to root. The last line disables all security for our cluster. I plan to change it, but for learning purposes, I will leave it as it is now.
Then restart the libvirtd service
# service libvirt restart
I use Ubuntu Server 18.04.3 LTS, and I am going to prepare creating bridge network for this system.
Open the /etc/sysctl.conf file and uncomment/add following line:
net.ipv4.ip_forward=1
Then apply changes with the sudo sysctl -p /etc/sysctl.conf command
Then reload procps with the sudo /etc/init.d/procps restart
Type following command: sudo apt-get install bridge-utils
My physical output interface is eno2. I am going to create a bridge for this interface. Make sure if your name is correct. If you do not have any backup connect and you make a mistake, you can lose the connection to your server.
Notes:
- Network: 192.168.0.0/24
- Net mask: 255.255.255.0
- Gateway: 192.168.0.1
- DHCP on the router: enabled
- Open the /etc/netplan/50-cloud-init.yaml and update your physical interface config to similar:
eno2:
addresses: []
dhcp4: false
- Add bridge configuration:
bridges:
br0:
interfaces: [eno2]
dhcp4: false
addresses: [192.168.0.100/24]
gateway4: 192.168.0.1
nameservers:
addresses:
- 192.168.0.1
- 8.8.8.8
What is this configuration mean?
dhcp4: false- With this directive, We disable DHCP for our interfaces. We don't want any IP for the physical interface. We can disable DHCP also for br0 because We assigned the static IP(192.168.0.100).interfaces: [eno2]- Determine which interface do you want to bridge.- Rest of the fields are straightforward
Note: Remember to replace config with your parameters.
Finally my file looks like below:
# This file is generated from information provided by
# the datasource. Changes to it will not persist across an instance.
# To disable cloud-init's network configuration capabilities, write a file
# /etc/cloud/cloud.cfg.d/99-disable-network-config.cfg with the following:
# network: {config: disabled}
network:
version: 2
ethernets:
eno1:
addresses: []
dhcp4: true
eno2:
addresses: []
dhcp4: false
enp4s0f0:
addresses: []
dhcp4: true
optional: true
enp4s0f1:
addresses: []
dhcp4: true
optional: true
enp5s0f0:
addresses: []
dhcp4: true
optional: true
enp5s0f1:
addresses: []
dhcp4: true
optional: true
bridges:
br0:
interfaces: [eno2]
dhcp4: false
addresses: [192.168.0.100/24]
gateway4: 192.168.0.1
nameservers:
addresses:
- 192.168.0.1
- 8.8.8.8
- Now test and apply your configuration
netplan try
netplan apply
- Verify if all works good. You should see 3 interfaces: Physical interface, should be up without any IP
3: eno2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master br0 state UP group default qlen 1000
link/ether 78:2b:cb:33:29:3d brd ff:ff:ff:ff:ff:ff
Bridge interface shoudl be up with static IP: 192.168.0.100:
94: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether 82:43:74:4e:8d:c4 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.100/24 brd 192.168.0.255 scope global br0
valid_lft forever preferred_lft forever
inet6 fe80::8043:74ff:fe4e:8dc4/64 scope link
valid_lft forever preferred_lft forever
And you should be able to see a KVM NAT interface, with some DHCP IP:
111: virbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether 52:54:00:43:8d:69 brd ff:ff:ff:ff:ff:ff
inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
valid_lft forever preferred_lft forever
Today, everybody uses different disks, settings, partitions, file systems, etc.
All storages have pros and cons. Ones provide better control others are faster. For example, an application which is doing some computations and save data on HDD requires a faster drive than DB, which keeps user DB.
KVM pools come against those problems. KVM pool is a mechanism, which helps us manage our drives. For example, We can create one for SSD drives and the second one for HDD drives. Users/administrators that use our infrastructure does not need to know what is under the mask. We can think about that like about a container or like about abstraction layer that isolates physical devices between software.
We have a few types of pool types:
- dir - uses direcectory as a storage for volumes.
- disk - uses physical hard disks as a storage for volumes.
- fs - uses preformated partition to store volumes.
- lvm - uses LVM volume as a storage driver.
- netfs - uses Use network file system(like NFS) to store volumes.
- Create pool on /mapper/red01/kvm/disks . Let's call it
virsh pool-define-as \
--name kvm-disks \
--type dir \
--target /mapper/red01/kvm/disks
virsh pool-build kvm-pool
virsh pool-start kvm-pool
virsh pool-autostart kvm-pool
Above commands define pool with the type of dir, called kvm-disks, which points to the /mapper/red01/kvm/disks directory. The next CLI command builds the pool. For the dir type, it is not necessary, but for another(like a disk), it is required to prepare filesystem. Then we start the pool and enable autostart.
- Verify, your configuration
Run the virsh pool-list command:
Name State Autostart
-------------------------------------------
default active yes
kvm-disks active yes
Run the virsh pool-dumpxml command:
<pool type='dir'>
<name>kvm-disks</name>
<uuid>4ab0ab6e-5006-4dcf-9fdf-d769946783b5</uuid>
<capacity unit='bytes'>7936293720064</capacity>
<allocation unit='bytes'>346643271680</allocation>
<available unit='bytes'>7589650448384</available>
<source>
</source>
<target>
<path>/mapper/red01/kvm/disks</path>
<permissions>
<mode>0755</mode>
<owner>0</owner>
<group>0</group>
</permissions>
</target>
</pool>
Now we can verify if everything is correct. If not, you can run the virsh pool-edit kvm-disks command.
On the beginning, I will provide the bash command and explain a bit what it does. To create a new VM, I am going to use the virt-install bash command, which comes from virt tools.
virt-install \
--virt-type=kvm \
--name debian10-base \
--ram 8000 \
--disk pool=kvm-disks,sparse=true,size=20 \
--vcpus 4 \
--os-type linux \
--os-variant debian9 \
--network network=default \
--graphics none \
--console pty,target_type=serial \
--location 'http://ftp.nl.debian.org/debian/dists/buster/main/installer-amd64/' \
--extra-args 'console=ttyS0,115200n8 serial'
Command has a lot of parameters. You can read more here: https://linux.die.net/man/1/virt-install.
virt-type- The hypervisor to install on. We are using KVM so, choice is clear.name- Name of our VMram- Ram allocation for our VMdisk- Configure disk.vcpus- Number of virtual CPUs allocated for VMos-type- Type of virtualized systemos-varian- Variant of selected OS, used for futher optimizazations. For now libvirt does not support Debian 10.network- Network settings*graphics- Settings of graphics interface. We do not need graphics support for now, so disable itconsole- Connect VMs console to pty, with type serial. Serial is configured inextra-argslocation- Location of installation disk. For more disks see below.
You can notice, I used only one network. Moreover, this is the default network. It's a good catch, but It is aware because I want to show you How to attach network interface and set up it in the guest OS.
Ok, let's type above command and install your OS. My installation parameters:
- Language:
English - Location:
Poland - Locales:
United States - en_US.UTF-8 - Keyboard:
American English - Hostname:
debian10-base - Domain name:
kvm-smaug - Username for your account:
kvm
Disk partitions:
- ext4, size: 10GB, mount point: /
- ext4, size: 10GB, mount point: /home
Software to install
- SSH Server
- standard system utilities
Install GRUB in /dev/vda.
- Log in as root (or kvm).
- Type
ip ato see your IP. In my example, I can see my IP address is192.168.122.100. That IP belongs to default networks. Make a note this address.
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: ens2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 52:54:00:f9:5f:ec brd ff:ff:ff:ff:ff:ff
inet 192.168.122.100/24 brd 192.168.122.255 scope global dynamic ens2
valid_lft 3209sec preferred_lft 3209sec
inet6 fe80::5054:ff:fef9:5fec/64 scope link
valid_lft forever preferred_lft forever
- Verify if your SSH server is running. Type
systemctl status sshd
ssh.service - OpenBSD Secure Shell server
Loaded: loaded (/lib/systemd/system/ssh.service; enabled; vendor preset: enab
Active: active (running) since Tue 2019-08-13 14:49:52 CEST; 8min ago
Docs: man:sshd(8)
man:sshd_config(5)
Process: 387 ExecStartPre=/usr/sbin/sshd -t (code=exited, status=0/SUCCESS)
Main PID: 395 (sshd)
Tasks: 1 (limit: 4915)
Memory: 3.7M
└─395 /usr/sbin/sshd -D
Aug 13 14:49:52 debian10-base systemd[1]: Starting OpenBSD Secure Shell server..
Aug 13 14:49:52 debian10-base sshd[395]: Server listening on 0.0.0.0 port 22.
Aug 13 14:49:52 debian10-base sshd[395]: Server listening on :: port 22.
Aug 13 14:49:52 debian10-base systemd[1]: Started OpenBSD Secure Shell server.
-
Reboot VM to see if everything is working as expected. To do that run
systemctl reboot. -
Detach console by typing:
CRTL + ^. It did not work for me, so I closed the terminal. -
Try to connect to your VM from the host computer.
- If everything is working, you did a great job!
On the beginning, let's list all our networks on the KVM node. Execute the following command: virsh net-list.
virsh net-list
Name State Autostart Persistent
----------------------------------------------------------
default active yes yes
We can notice only the default network is available. Let's create a new one then.
cat /var/lib/libvirt/networks/host-bridge.xml
-
Create a new network file in the
/var/lib/libvirt/networks/host-bridge.xmlFile content:<network> <name>host-bridge</name> <forward mode="bridge"/> <bridge name="br0"/> </network> -
Create a network based on the XML file. Execute the following commands:
# virsh net-define /var/lib/libvirt/networks/host-bridge.xml # virsh net-start host-bridge # virsh net-autostart host-bridgeExample:
# virsh net-define /var/lib/libvirt/networks/host-bridge.xml Network host-bridge-1 defined from /var/lib/libvirt/networks/host-bridge.xml # virsh net-list --all Name State Autostart Persistent ---------------------------------------------------------- default active yes yes host-bridge inactive no yes # virsh net-start host-bridge Network host-bridge started # virsh net-list Name State Autostart Persistent ---------------------------------------------------------- default active yes yes host-bridge active no yes # virsh net-autostart host-bridge Network host-bridge marked as autostarted # virsh net-list Name State Autostart Persistent ---------------------------------------------------------- default active yes yes host-bridge active yes yes -
I assume you have got VM called
debian10-base. Let's find it.# virsh list Id Name State ---------------------------------------------------- 41 debian10-base running -
SSH into VM:
virsh console debian10-base -
See current network settings.
# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ens2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:f9:5f:ec brd ff:ff:ff:ff:ff:ff inet 192.168.122.100/24 brd 192.168.122.255 scope global dynamic ens2 valid_lft 3583sec preferred_lft 3583sec inet6 fe80::5054:ff:fef9:5fec/64 scope link valid_lft forever preferred_lft forever -
Type
CTRL + ]to exit from the attached console. -
Generate random MAC address.
# openssl rand -hex 6 | sed 's/\(..\)/\1:/g; s/:$//' 52:54:00:4b:73:5f -
Attach interface with the following command:
# virsh attach-interface \ --domain debian10-base \ --type network \ --source host-bridge \ --model virtio \ --mac 52:54:00:4b:73:5f \ --config \ --live -
Disable netfilter on the bridge. This is required to disable filtering packages by our KVM host for bridged interfaces.
# cat >> /etc/sysctl.conf <<EOF net.bridge.bridge-nf-call-ip6tables = 0 net.bridge.bridge-nf-call-iptables = 0 net.bridge.bridge-nf-call-arptables = 0 EOF # sysctl -p /etc/sysctl.conf -
It's good time to restart our VM
# virsh reboot debian10-base Domain debian10-base is being rebooted -
SSH into VM with
virsh console debian10-base. Then see available networks:# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ens2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:f9:5f:ec brd ff:ff:ff:ff:ff:ff inet 192.168.122.100/24 brd 192.168.122.255 scope global dynamic ens2 valid_lft 3583sec preferred_lft 3583sec inet6 fe80::5054:ff:fef9:5fec/64 scope link valid_lft forever preferred_lft forever 3: ens9: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether 52:54:00:4b:73:5f brd ff:ff:ff:ff:ff:ff -
We can see our MAC addres:
52:54:00:4b:73:5f. Now lets UP the interface# ip link set dev ens9 up -
See if the interface is up
# ip a 3: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:4b:73:5f brd ff:ff:ff:ff:ff:ff inet6 fe80::5054:ff:fe4b:735f/64 scope link valid_lft forever preferred_lft forever -
Obtain new IP from DHCP server with
/usr/sbin/dhclient <interface name># /usr/sbin/dhclient ens9 -
That's it! You should be able to see what is your IP:
# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ens2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:f9:5f:ec brd ff:ff:ff:ff:ff:ff inet 192.168.122.100/24 brd 192.168.122.255 scope global dynamic ens2 valid_lft 2480sec preferred_lft 2480sec inet6 fe80::5054:ff:fef9:5fec/64 scope link valid_lft forever preferred_lft forever 3: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:4b:73:5f brd ff:ff:ff:ff:ff:ff inet 192.168.0.71/24 brd 192.168.0.255 scope global dynamic ens9 valid_lft 692448sec preferred_lft 692448sec inet6 fe80::5054:ff:fe4b:735f/64 scope link valid_lft forever preferred_lft forever
Different people prefer to use a different type of volumes.
Raw format is the most universal and portable. You can export raw disks into another Virtualization format. Another advantage of that format is high performance.
The qcow format provides several like features: snapshots, compressions, backups, etc.
At the moment, it's not important what kind of format you choose.
Like You remember We use the "kvm-pool" pool as a volumes storage.
-
Let's check our volumes pool.
# virsh pool-dumpxml kvm-disks <pool type='dir'> <name>kvm-disks</name> <uuid>4ab0ab6e-5006-4dcf-9fdf-d769946783b5</uuid> <capacity unit='bytes'>7936293720064</capacity> <allocation unit='bytes'>348723073024</allocation> <available unit='bytes'>7587570647040</available> <source> </source> <target> <path>/mapper/red01/kvm/disks</path> <permissions> <mode>0755</mode> <owner>0</owner> <group>0</group> </permissions> </target> </pool> -
Go to the folder where pool is defined:
/mapper/red01/kvm/disks -
Let's create new
10Gdisk:# cd /mapper/red01/kvm/disks # qemu-img create -f raw test-disk-10G 10G Formatting 'test-disk-10G', fmt=raw size=10737418240You can also use dd command:
# dd if=/dev/zero of=test-disk-2-10G bs=1M count=10240 status=progress 10078912512 bytes (10 GB, 9.4 GiB) copied, 9 s, 1.1 GB/s 10240+0 records in 10240+0 records out 10737418240 bytes (11 GB, 10 GiB) copied, 9.54978 s, 1.1 GB/s# sudo ls -lh total 15G -rw------- 1 root root 21G Sep 24 12:30 debian10-base.qcow2 -rw------- 1 root root 8.1G Aug 13 09:55 debian9.qcow2 -rw-r--r-- 1 root root 10G Sep 24 12:27 test-disk-10G -rw-r--r-- 1 root root 10G Sep 24 12:30 test-disk-2-10G -
I am going to attach this image to the
debian10-baseVM. Let's log into that instance.# virsh console debian10-base Connected to domain debian10-base Escape character is ^] root@debian10-base:~# -
Find currently attached disks
root@debian10-base:~# ls -als /dev | grep -e vd.$ 0 brw-rw---- 1 root disk 254, 0 Sep 10 12:39 vdaHere You can see that currently attached is disk /dev/vda
-
We can attach our disk as vdb device. Let's do this! Log out from the guest VM, then attach created volume.
# virsh attach-disk debian10-base \ --source /mapper/red01/kvm/disks/test-disk-10G \ --target vdb \ --persistent Disk attached successfully -
Login to the guest VM and verify.
root@debian10-base:~# fdisk -l | grep '^Disk /dev/vd[a-z]' Disk /dev/vda: 20 GiB, 21474836480 bytes, 41943040 sectors Disk /dev/vdb: 10 GiB, 10737418240 bytes, 20971520 sectors -
Create a new partition
root@debian10-base:~# fdisk /dev/vdb Welcome to fdisk (util-linux 2.33.1). Changes will remain in memory only, until you decide to write them. Be careful before using the write command. Device does not contain a recognized partition table. Created a new DOS disklabel with disk identifier 0xa0b8bd13. Command (m for help): n Partition type p primary (0 primary, 0 extended, 4 free) e extended (container for logical partitions) Select (default p): p Partition number (1-4, default 1): First sector (2048-20971519, default 2048): Last sector, +/-sectors or +/-size{K,M,G,T,P} (2048-20971519, default 20971519): Created a new partition 1 of type 'Linux' and of size 10 GiB. Command (m for help): w The partition table has been altered. Calling ioctl() to re-read partition table. Syncing disks. -
Verify partitions
root@debian10-base:~# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT vda 254:0 0 20G 0 disk ├─vda1 254:1 0 9.3G 0 part / └─vda2 254:2 0 10.7G 0 part /home vdb 254:16 0 10G 0 disk └─vdb1 254:17 0 10G 0 part -
Format created partition
root@debian10-base:~# mkfs.ext4 /dev/vdb1 mke2fs 1.44.5 (15-Dec-2018) Creating filesystem with 2621184 4k blocks and 655360 inodes Filesystem UUID: 659eed8b-c913-45ad-a2a6-79c499e338a2 Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632 Allocating group tables: done Writing inode tables: done Creating journal (16384 blocks): done Writing superblocks and filesystem accounting information: done -
Mount partition to the
/var/lib/mysqlroot@debian10-base:~#mount /dev/vdb1 /var/lib/mysql root@debian10-base:~# mount | grep vdb1 /dev/vdb1 on /var/lib/mysql type ext4 (rw,relatime) -
Add a fstab entry
root@debian10-base:~# echo '/dev/vdb1 /var/lib/mysql ext4 defaults 0 0' > /etc/fstab root@debian10-base:~# tail -n 1 /etc/fstab /dev/vdb1 /var/lib/mysql ext4 defaults 0 0
-
Before We start let's check current file system size for particular guest
root@debian10-base:~# df -h Filesystem Size Used Avail Use% Mounted on udev 3.8G 0 3.8G 0% /dev tmpfs 779M 77M 703M 10% /run /dev/vda1 9.2G 1.1G 7.7G 12% / tmpfs 3.9G 0 3.9G 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup /dev/vda2 11G 41M 9.9G 1% /home tmpfs 779M 0 779M 0% /run/user/0 /dev/vdb1 9.8G 37M 9.3G 1% /var/lib/mysql -
Shutdown the guest VM
root@debian10-base:~# shutdown -P now -
On the host machine resize the disk file
# sudo qemu-img resize -f raw ./test-disk-10G +25G Image resized. -
Start guest VM
# virsh start debian10-base Domain debian10-base started -
Login into guest VM, and verify filesystem
# virsh console debian10-base root@debian10-base:~# df -h Filesystem Size Used Avail Use% Mounted on udev 3.8G 0 3.8G 0% /dev tmpfs 779M 8.4M 771M 2% /run /dev/vda1 9.2G 1.1G 7.7G 12% / tmpfs 3.9G 0 3.9G 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup /dev/vdb1 9.8G 37M 9.3G 1% /var/lib/mysql tmpfs 779M 0 779M 0% /run/user/0 root@debian10-base:~# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT vda 254:0 0 20G 0 disk ├─vda1 254:1 0 9.3G 0 part / └─vda2 254:2 0 10.7G 0 part vdb 254:16 0 35G 0 disk └─vdb1 254:17 0 10G 0 part /var/lib/mysqlAs You can see disk size is 35G, but used is only 10. Let's change it.
-
Modify partition
root@debian10-base:~# fdisk /dev/vdb1 Command (m for help): p Disk /dev/vdb: 35 GiB, 37580963840 bytes, 73400320 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0xa0b8bd13 Device Boot Start End Sectors Size Id Type /dev/vdb1 2048 20971519 20969472 10G 83 Linux Command (m for help): d Selected partition 1 Partition 1 has been deleted. Command (m for help): p Disk /dev/vdb: 35 GiB, 37580963840 bytes, 73400320 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0xa0b8bd13 Command (m for help): n Partition type p primary (0 primary, 0 extended, 4 free) e extended (container for logical partitions) Select (default p): p Partition number (1-4, default 1): First sector (2048-73400319, default 2048): Last sector, +/-sectors or +/-size{K,M,G,T,P} (2048-73400319, default 73400319): Created a new partition 1 of type 'Linux' and of size 35 GiB. Partition #1 contains a ext4 signature. Do you want to remove the signature? [Y]es/[N]o: N Command (m for help): w The partition table has been altered. Syncing disks.Commands:
- p - Print partition table
- d - Delete partition
- n - Create new partition
- w - Write changes
-
Resize current mount point
root@debian10-base:~# resize2fs /dev/vdb1 -
Reboot your guest VM, then verify results
root@debian10-base:~# reboot root@debian10-base:~# df -h Filesystem Size Used Avail Use% Mounted on udev 3.8G 0 3.8G 0% /dev tmpfs 779M 8.4M 771M 2% /run /dev/vda1 9.2G 1.1G 7.7G 12% / tmpfs 3.9G 0 3.9G 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 3.9G 0 3.9G 0% /sys/fs/cgroup /dev/vdb1 35G 48M 33G 1% /var/lib/mysql tmpfs 779M 0 779M 0% /run/user/0