VPS upgrade: more RAM, CPU and disk (and what to do inside the VPS)

In the Change plan tab you can add memory to your VPS (cores and disk grow with it) or more IPv4 addresses, and you pay only the prorated part of the price. We'll show what you need to do after paying so the VPS actually uses the new resources.

What you can increase#

Open My services, click your VPS and go to the Change plan tab. You set the new RAM and the number of IPv4 addresses, and the Now and New table shows what will change.

Both VPS Performance and VPS Budget are priced by RAM (Performance 82 CZK, Budget 61 CZK per GB a month); cores and disk are included. How much an upgrade costs is described in the Renewal and service upgrade guide.

You can only go up. A VPS cannot be downsized — the disk cannot shrink. If you need less, order a new VPS, move your data and cancel the current service on the Cancel service tab with the Cancel at the end of the period button. It keeps running until the end of the paid period.

Ordering and paying the difference#

1. Set the new RAM or number of IPv4 addresses (or use Quick pick: packages).
2. Check the summary: the Monthly row in the New column is the new monthly price; Pay now is the prorated part of the difference for the remaining days of the current period (“X of Y days until …”).
3. Click Order and pay and pay the proforma invoice (for payment methods see Payment methods and fees).
4. After payment the upgrade runs automatically. If the tab shows “A previous paid upgrade is still being completed”, wait a few minutes.

If you don't finish the payment, you'll see You have an unpaid upgrade order with the Finish payment and Cancel order buttons. You can order another upgrade only after this order is settled. More about payments in the Renewal and service upgrade guide.

Restart after a RAM and CPU upgrade#

After payment, the Console and Change plan tabs show Upgrade waiting for a restart: “More RAM and cores apply after a VPS restart.”

1. Save and stop game servers and applications that save data when restarting. System services (a database, for example) shut down on their own during a clean restart, but the Restart button waits for the system to shut down for only about 20 seconds; after that it hard-restarts the VPS.
2. Click Go to console.
3. Click Restart and confirm the restart.
4. Once the VPS is up, check the resources (see Verification).

The gentlest option: instead of the Restart button, run the poweroff command inside the VPS. The system shuts down with no time limit. When the Console tab shows the Stopped status, click Start. The new memory and cores are loaded on this start.

Extending the disk inside the VPS#

After a disk upgrade, the VPS's “physical” disk is bigger, but the partition and file system on it still have the original size. The extra space has to be added to the system partition.

The easiest way: restart the VPS with the Restart button. The VPS system usually extends the partition and file system by itself at boot. Then check the size with df -h /. If it didn't grow, or you don't want to restart, do it by hand (works while running, no restart needed).

Manual extension without a restart#

1. Find out which partition holds the system and what file system it has:

   findmnt -n -o SOURCE,FSTYPE /

   • findmnt – lists mounted file systems.
   • -n – no header.
   • -o SOURCE,FSTYPE – only the device and the file system type.
   • / – for the root folder (the system).

   /dev/sda1 ext4

   The system is on the partition /dev/sda1 (disk sda, partition number 1) with the ext4 file system. Yours may be disk vda or a different partition number; in that case use your values in the next steps.

2. Compare the disk and partition size:

   lsblk -l -o NAME,SIZE,TYPE,MOUNTPOINTS

   • lsblk – lists disks and partitions.
   • -l – as a list, without the tree graphics.
   • -o … – columns: name, size, type (disk / part) and mount point.

   The line of type disk (for example sda 60G) shows the new disk size; the line with the mount point / (for example sda1 49.9G) shows the size of the system partition.

   If the disk line still shows the original size, the system hasn't noticed the change yet. Ask it to check the disk again and repeat the lsblk command:

   echo 1 > /sys/class/block/sda/device/rescan

   • echo 1 > …/rescan – tells the kernel to detect the size of disk sda again. It doesn't change anything on the disk. For a disk with a different name, replace sda.

   If the size still doesn't change, restart the VPS with the Restart button and continue after the restart.

3. Grow the partition into all the free space:

   growpart /dev/sda 1

   • growpart – grows a partition to take up the free space after it.
   • /dev/sda – the disk (without a number).
   • 1 – the partition number. There must be a space between the disk and the number.

   You'll see a line starting with CHANGED: partition=1. If the command is missing, install it: apt install cloud-guest-utils.

Finally, extend the file system according to the type step 1 printed:

New IPv4 addresses#

After payment we usually assign the addresses automatically. If that can't happen right away, the Change plan tab shows “IPv4 addresses waiting for assignment”. Once we assign them, we send an email and you'll see the addresses in the Network tab. You have to configure them inside the VPS yourself following the IPv6 and additional IP addresses guide.

Verification#

nproc
free -h
df -h /

• nproc – the number of CPU cores the VPS sees.
• free -h – the RAM size (the total column on the Mem line). The value is usually a bit lower than ordered, because the system kernel reserves part of it.
• df -h / – the size and free space of the system partition (the Size and Avail columns).

You'll also see the used space in the customer area in the Disk tab.

Troubleshooting#

RAM or the number of cores is the same after a restart#

You restarted the VPS with the reboot command. Restart it with the Restart button in the Console tab, or stop it and start it again. If the Change plan tab still shows “Upgrade waiting for a restart” even after a restart with the button, the system most likely didn't shut down within 20 seconds and the VPS was hard-restarted without loading the changes. Run poweroff inside the VPS, wait for the Stopped status and click Start. If even that doesn't help, contact us.

NOCHANGE: partition 1 is size … it cannot be grown#

The partition is already as big as it can be. Either the system already extended it by itself (check df -h / and, if needed, run just resize2fs), or another partition sits after the system partition. In the second case, contact us.

resize2fs prints Nothing to do!#

The file system is already the same size as the partition. Check that you ran growpart first.

df -h / still shows the old size#

Check in lsblk that both the disk and the partition are bigger. If only the disk is bigger, the growpart step is missing. If the partition is bigger too, resize2fs (or xfs_growfs) is missing.

The upgrade can't be ordered#

The tab itself shows the possible reasons: an unpaid upgrade order (finish it or cancel it), a previous upgrade still being completed, or a pending service transfer. The service must be active.

Further reading#

• resize2fs manual page (opens in a new window)
• growpart manual page (opens in a new window)
• lsblk manual page (opens in a new window)