How to extend disk in linux

Increase partition size on which Ubuntu is installed?

I have Ubuntu installed on a drive. The problem is it is running low on disk space. This is installed on vmware. I expanded the virtual drive and booted into ubuntu. But when I opened gparted(sudo gparted), the move/resize option is unavailable. This is the partition Ubuntu is installed on, but I need to resize it. Any ideas? I am comfortable using command line

You can see how they do it in raspi-config for the Raspberry Pi, where expanding the root file system is something you do almost every time you set up a new Raspberry Pi. github.com/RPi-Distro/raspi-config/blob/…

I’ve found a youtube tutorial which can be a good answer for this question, it starts at this position: youtu.be/07hpUigWQyY?t=1538

8 Answers 8

As a matter of fact, you CAN enlarge the root filesystem while Ubuntu is running (I learned this recently myself here) — this sounds incredible but it’s true 🙂

Here’s the list of steps for a simple scenario where you have two partitions, /dev/sda1 is an ext4 partition the OS is booted from and /dev/sdb2 is swap. For this exercise we want to remove the swap partition an extend /dev/sda1 to the whole disk.

1. As always, make sure you have a backup of your data — since we’re going to modify the partition table there’s a chance to lose all your data if you make a typo, for example.
2. Run sudo fdisk /dev/sda
   • use p to list the partitions. Make note of the start cylinder of /dev/sda1
   • use d to delete first the swap partition ( 2 ) and then the /dev/sda1 partition. This is very scary but is actually harmless as the data is not written to the disk until you write the changes to the disk.
   • use n to create a new primary partition. Make sure its start cylinder is exactly the same as the old /dev/sda1 used to have. For the end cylinder agree with the default choice, which is to make the partition to span the whole disk.
   • use a to toggle the bootable flag on the new /dev/sda1
   • review your changes, make a deep breath and use w to write the new partition table to disk. You’ll get a message telling that the kernel couldn’t re-read the partition table because the device is busy, but that’s ok.
3. Reboot with sudo reboot . When the system boots, you’ll have a smaller filesystem living inside a larger partition.
4. The next magic command is resize2fs . Run sudo resize2fs /dev/sda1 — this form will default to making the filesystem to take all available space on the partition.

That’s it, we’ve just resized a partition on which Ubuntu is installed, without booting from an external drive.

@Isaac: Yes, we still have to reboot to make the kernel to re-read the partition table. However, a quick reboot in many scenarios is a much smaller problem than booting from an external medium (think of remote machines, headless machines, VPSes etc.)

Two quick notes: 1. You can use df to show the «resulting» space after this (confirm that it worked), and 2. remember to recreate some swap space!

Thanks so much. Ubuntu never used the space I gave the VirtualBox vdisk. Now I know why. Many thanks!

Thanks for a great post. In step 2, after running ‘n’, it prompts whether to delete a signature. I choose to retain the signature, and approach works as expected.

@Sid: It is very important that the start cylinder is the exactly the same. If the free space is anywhere but immediately after the partition you need to use another method (which would require booting from another drive)

Two Methods

1. Tested in Ubuntu Server 16.04 to 18.10

After extending the volume in VMware ESXi (shutdown vm, backup/export vm, remove all snapshots, then increased number in disk size, take snapshot of vm again before following the next steps so you can revert mistake/error).

The following example is extending volume on root partition on a system that is already using LVM.

 # fdisk -l /dev/sda Disk /dev/sda: 20 GiB, 21474836480 bytes, 41943040 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: 0xc8b647ff Device Boot Start End Sectors Size Id Type /dev/sda1 * 2048 999423 997376 487M 83 Linux /dev/sda2 1001470 16775167 15773698 7.5G 5 Extended /dev/sda3 16775168 41943039 25167872 12G 8e Linux LVM (***** See new partition /dev/sda3*****) /dev/sda5 1001472 16775167 15773696 7.5G 8e Linux LVM 

 # pvcreate /dev/sda3 Physical volume "/dev/sda3" successfully created 

 # vgdisplay --- Volume group --- VG Name linuxetc1-vg (***** this is vg i'm extending*****) System ID Format lvm2 Metadata Areas 1 Metadata Sequence No 3 VG Access read/write VG Status resizable MAX LV 0 Cur LV 2 Open LV 2 Max PV 0 Cur PV 1 Act PV 1 VG Size 7.52 GiB PE Size 4.00 MiB Total PE 1925 Alloc PE / Size 1925 / 7.52 GiB Free PE / Size 0 / 0 . 

 # vgextend linuxetc1-vg /dev/sda3 Volume group "linuxetc1-vg" successfully extended 

 # lvextend -l+100%FREE /dev/linuxetc1-vg/root Size of logical volume linuxetc1-vg/root changed from 6.52 GiB (1669 extents) to 18.52 GiB (4741 extents). Logical volume root successfully resized. Or if you only wanted to give 5 more GB to root # lvextend -L +5g /dev/linuxetc1-vg/root 

 # df -h Filesystem Size Used Avail Use% Mounted on udev 477M 0 477M 0% /dev tmpfs 100M 4.6M 95M 5% /run /dev/mapper/linuxetc1--vg-root 6.3G 2.5G 3.5G 42% / (***** this line*****) tmpfs 497M 0 497M 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 497M 0 497M 0% /sys/fs/cgroup /dev/sda1 472M 105M 343M 24% /boot tmpfs 100M 0 100M 0% /run/user/1000 

 # resize2fs /dev/mapper/linuxetc1--vg-root resize2fs 1.42.13 (17-May-2015) Filesystem at /dev/mapper/linuxetc1--vg-root is mounted on /; on-line resizing required old_desc_blocks = 1, new_desc_blocks = 2 The filesystem on /dev/mapper/linuxetc1--vg-root is now 4854784 (4k) blocks long. 

 # df -h Filesystem Size Used Avail Use% Mounted on udev 477M 0 477M 0% /dev tmpfs 100M 4.6M 95M 5% /run /dev/mapper/linuxetc1--vg-root 19G 2.5G 15G 15% / (***** this line*****) tmpfs 497M 0 497M 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 497M 0 497M 0% /sys/fs/cgroup /dev/sda1 472M 105M 343M 24% /boot tmpfs 100M 0 100M 0% /run/user/1000 

2. Faster Method Tested in Ubuntu Server 18.04.3 LTS

This method is a little faster than the first. This is due to a new option in the cfdisk menu called «Resize» that saves some steps.

After extending the volume in VMware ESXi (shutdown vm, backup/export vm, remove all snapshots, then increased number in disk size, take snapshot of vm again before following the next steps so you can revert mistake/error).

The following example is extending volume on root partition on a system that is already using LVM.

1. ssh into the system, switch to root and run cfdisk .
2. Choose the partition to extend and select «Resize».
3. Set the «New size».
4. After pressing enter, you’ll see screen with the following note «Partition [someNumber] resized»:
5. Next you’ll need to «Write» (save) your changes:
6. Quit cfdisk . When you exit you may see message «syncing disks».
7. I rebooted my system at this point but it may not be necessary. Now you can pickup from step 19 from the first method (repeated below).
8. (19) Get filesystem name for next step

 # df -h Filesystem Size Used Avail Use% Mounted on udev 477M 0 477M 0% /dev tmpfs 100M 4.6M 95M 5% /run /dev/mapper/linuxetc1--vg-root 6.3G 2.5G 3.5G 42% / (***** this line*****) tmpfs 497M 0 497M 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 497M 0 497M 0% /sys/fs/cgroup /dev/sda1 472M 105M 343M 24% /boot tmpfs 100M 0 100M 0% /run/user/1000 

 # resize2fs /dev/mapper/linuxetc1--vg-root resize2fs 1.42.13 (17-May-2015) Filesystem at /dev/mapper/linuxetc1--vg-root is mounted on /; on-line resizing required old_desc_blocks = 1, new_desc_blocks = 2 The filesystem on /dev/mapper/linuxetc1--vg-root is now 4854784 (4k) blocks long. 

 # df -h Filesystem Size Used Avail Use% Mounted on udev 477M 0 477M 0% /dev tmpfs 100M 4.6M 95M 5% /run /dev/mapper/linuxetc1--vg-root 19G 2.5G 15G 15% / (***** this line*****) tmpfs 497M 0 497M 0% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 497M 0 497M 0% /sys/fs/cgroup /dev/sda1 472M 105M 343M 24% /boot tmpfs 100M 0 100M 0% /run/user/1000 

Источник

Resize-Extend a disk partition with unallocated disk space in Linux — CentOS, RHEL, Ubuntu, Debian & more

UPDATE: this post was originally published in 2017 and has noot been updated ever since. For an updated approach that allows to resize/extend any Linux VM using a more secure technique and without the need of a reboot, we strongly recommend to read this post instead.

Expanding disk partitions to use all the available (unallocated) disk space is a common issue among Linux Administrators, expecially when working in a VMware-based Cloud environment: deploying a Linux VM from an existing template will often lead to disk partitions smaller than the disk space allocated during the VM configuration phase.

Consider the following scenario:

This screen can be obtained by running cfdisk from the terminal and shows the following:

• a 524MB boot partition [sda1]
• a 6.8GB drive [sda2], used by the Linux OS and all its installed packages.
• 100GB of unallocated space

It would be great to extend that puny 6.8GB partition and make it become a 106.8GB drive. How can we do that? If you take a look around you’ll see that the web is awfully short of a quick and effective tutorial for this: that’s why I eventually chose to write my own guide: here it is!

Luckily enough, we won’t need anything fancy to perform our task: we’re just going to make good use of fdisk , pvresize , lvdisplay and lvextend , some handy command-line tools shipped with any Linux distribution: that’s great to hear, since it means that this tutorial will work for any Linux distro, including CentOS 5.x, CentOS 6.x, CentOS 7.x, RHEL, Ubuntu, Debian and more!

Step 1: Alter the Partition Table

The first thing we need to do is to modify our partition table to make sda2 end at end of disk: don’t worry, you will not lose your existing data! However, this task will require a reboot in order to write the changes that we’re going to make and also to re-read the updated partition table.

Let’s start with running the following command:

This will make the terminal enter in fdisk mode: once there, type p to print the current partition table: it’s very important to keep note of the numeric values of the START and END columns for the / dev / sda2 partition, as we’re going to need them soon enough. If you want to be sure to not lose them or typing them wrong, just print-screen or paper-print them.

Once done, type d to access the delete mode, and then the number of the partition record that you want to remove (that would be 2 in the above scenario). Again, DO NOT WORRY: you’re not deleting the partition data, just its mapping addresses on the partition table.

Right after that, type n to create a brand-new second partition: choose the same partition mode of the previous one (that would be Primary in the above scenario), then input the START numeric value you’ve recorded earlier — which shoud be the suggested default; also, make sure that the end is at the end of the disk — which should also be the default.

Last but not least, we need to change the partition type from Linux to Linux LVM: to do so, type t to access the change partition type mode, then 2 , then 8e and that’s it.

When done, type p to review your new partition layout. Be sure to triple-check that the start of the new second partition is exactly where the old second partition was: this is very important! In case you did it wrong, type d to delete the wrong partition and start over.

If everything looks right, issue w to write the partition table to disk.

Step 2: Reboot

Right after writing the new partition table to disk, you’ll immediately get a system error message due to the fact that the partition table couldn’t be accessed for read, because the disk is in use. That’s why we need to reboot our system.

Step 3: Expand the LVM Partition

Wait for the system to reboot, then issue the following command:

Источник