Change SWAP Size to Improve Low Memory Cloud VM Reliability

I am having a couple of low memory vm in the cloud to run some basic testing lab. It always gets me a notification for the performance , especially on memory since they are only have 587M memory.
This process is good for both Ubuntu and CentOS.

Performance Issue on Low Memory Cloud VM

You might experience this because of either CPU or Memory, or both. One way to improve system reliability to avoid system crush is to increase our swap file size. This post is to summarize some steps to configure swap file and swappiness value. 

One Command to Add/Delete SWAP File Size

Command (not support OpenVZ)  from root user：

wget https://raw.githubusercontent.com/51sec/swap/main/swap.sh && bash swap.sh

When choose add swap, please enter number for size. Default metric is MB。

In screenshot, the priority is showing 0. Script in github has set to 100. 

Github: https://github.com/51sec/swap

Original script is from : https://yun.94ish.me/sh/swap.sh

Adjusting the Swappiness Value

Swappiness is a Linux kernel property that defines how often the system will use the swap space. It can have a value between 0 and 100. A low value will make the kernel to try to avoid swapping whenever possible, while a higher value will make the kernel to use the swap space more aggressively.

On Ubuntu, the default swappiness value is set to 60. You can check the current value by typing the following command:

cat /proc/sys/vm/swappiness

60

While the swappiness value of 60 is OK for most Linux systems, for production servers, you may need to set a lower value.

For example, to set the swappiness value to 10, run:

sudo sysctl vm.swappiness=10

To make this parameter persistent across reboots, append the following line to the /etc/sysctl.conf file:

/etc/sysctl.conf

vm.swappiness=10

The optimal swappiness value depends on your system workload and how the memory is being used. You should adjust this parameter in small increments to find an optimal value.

Manually Change SWAP File Size

Linux provides for two types of swap space. By default, most Linux installations create a swap partition, but it is also possible to use a specially configured file as a swap file. A swap partition is just what its name implies—a standard disk partition that is designated as swap space by the mkswap command.  If you don’t have enough memory, swap will be used quite often and noticeably more during any memory requirement spikes. If you don’t have enough memory and no swap space, this will often cause failure to allocate memory for requests needing more memory pages. As a last resort, the Kernel will deploy OOM killer to nuke high-memory processes (usually MySQL, java, etc), which shows in my previous post - Linux Virtual Machine PHP Invoked OOM-killer (Out of Memory) Issue.

Here comes recommended SWAP spaces:

Amount of system RAM	Recommended swap space	Recommended swap with hibernation
less than 2 GB	2 times the amount of RAM	3 times the amount of RAM
2 GB - 8 GB	Equal to the amount of RAM	2 times the amount of RAM
8 GB - 64 GB	0.5 times the amount of RAM	1.5 times the amount of RAM
more than 64 GB	workload dependent	hibernation not recommended

In my case, my VM is only having 587M memory with 0K swap file size.

Based on recommendation, I am adding a 1GB swap file for it.

[[email protected] ~]# sudo fallocate -l 1G /swapfile
[[email protected] ~]# sudo dd if=/dev/zero of=/swapfile bs=1024 count=1048576
1048576+0 records in
1048576+0 records out
1073741824 bytes (1.1 GB) copied, 27.5052 s, 39.0 MB/s
[[email protected] ~]# sudo chmod 600 /swapfile
[[email protected] ~]# sudo mkswap /swapfile
Setting up swapspace version 1, size = 1048572 KiB
no label, UUID=d126e13f-162d-4a1d-998c-f8e4152e4f8b
[[email protected] ~]# sudo swapon /swapfile
[[email protected] ~]# vi /etc/fstab

Command Explanation:

• create a file which will be used as swap space: sudo dd if=/dev/zero of=/swapfile bs=1024 count=1048576
• Ensure that only the root user can read and write the swap file by setting the correct permissions:sudo chmod 600 /swapfile
• set up a Linux swap area on the file:sudo mkswap /swapfile
• Run the following command to activate the swap: sudo swapon /swapfile
• Make the change permanent by opening the /etc/fstab file: sudo vi /etc/fstab

Add a new line into /etc/fstab file:
/swapfile swap swap defaults 0 0

Notes: How do I allocate memory to work as swap space in an Amazon EC2 instance by using a swap file?

Note: No need to reboot system

Verification by using commands:

• top
• htop
• free -h

To Sum up all commands again:

1. sudo fallocate -l 1G /swapfile ##it won't work if fallocate is not installed. using dd if=/dev/zero of=/swapfile bs=1024 count=1048576 command instead
2. sudo chmod 600 /swapfile
3. sudo mkswap /swapfile
4. sudo swapon /swapfile
5. echo '/swapfile none swap sw 0 0' | sudo tee -a /etc/fstab

Performance Testing

With the methods posted in my post : VPS Bench Comparison for AWS,Azure,Google Cloud and Oracle Cloud (Free Tier & Regular Tier)

I could not tell much difference from before and after swap increase change, but it definitely improves reliability to reduce system crush risk. It might be will tell difference when there are some applications heavyly using memory.

Before SWAP Increase

After SWAP Increase

nench.sh v2019.07.20 -- https://git.io/nench.sh  benchmark timestamp:    2020-03-18 21:17:43 UTC ------------------------------------------------- Processor:    Intel(R) Xeon(R) CPU @ 2.30GHz CPU cores:    1 Frequency:    2300.000 MHz RAM:          587M Swap:         - Kernel:       Linux 3.10.0-1062.12.1.el7.x86_64 x86_64 Disks: sda     10G  HDD CPU: SHA256-hashing 500 MB     1.719 seconds CPU: bzip2-compressing 500 MB     CPU: AES-encrypting 500 MB     1.052 seconds ioping: seek rate     min/avg/max/mdev = 452.8 us / 5.33 ms / 51.9 ms / 8.48 ms ioping: sequential read speed     generated 2.46 k requests in 5.00 s, 614 MiB, 490 iops, 122.7 MiB/s dd: sequential write speed     1st run:    35.86 MiB/s     2nd run:    36.05 MiB/s     3rd run:    36.05 MiB/s     average:    35.99 MiB/s IPv4 speedtests     your IPv4:    35.226.94.xxxx     Cachefly CDN:         110.45 MiB/s     Leaseweb (NL):        17.40 MiB/s     Softlayer DAL (US):   60.76 MiB/s     Online.net (FR):      17.77 MiB/s     OVH BHS (CA):         44.79 MiB/s No IPv6 connectivity detected -------------------------------------------------

nench.sh v2019.07.20 -- https://git.io/nench.sh  benchmark timestamp:    2020-03-18 21:36:09 UTC ------------------------------------------------- Processor:    Intel(R) Xeon(R) CPU @ 2.30GHz CPU cores:    1 Frequency:    2300.000 MHz RAM:          587M Swap:         1.0G Kernel:       Linux 3.10.0-1062.12.1.el7.x86_64 x86_64 Disks: sda     10G  HDD CPU: SHA256-hashing 500 MB     1.731 seconds CPU: bzip2-compressing 500 MB     CPU: AES-encrypting 500 MB     1.059 seconds ioping: seek rate     min/avg/max/mdev = 338.6 us / 4.13 ms / 68.9 ms / 6.66 ms ioping: sequential read speed     generated 2.46 k requests in 5.00 s, 615.5 MiB, 492 iops, 123.1 MiB/s dd: sequential write speed     1st run:    35.86 MiB/s     2nd run:    36.05 MiB/s     3rd run:    36.05 MiB/s     average:    35.99 MiB/s IPv4 speedtests     your IPv4:    35.226.94.xxxx     Cachefly CDN:         101.90 MiB/s     Leaseweb (NL):        17.43 MiB/s     Softlayer DAL (US):   60.28 MiB/s     Online.net (FR):      17.80 MiB/s     OVH BHS (CA):         32.00 MiB/s No IPv6 connectivity detected -------------------------------------------------

In the following example, we’ll extend the swap space available in the /swapfile from 0 GB to 1 GB.

Turn off all swap processes

• sudo swapoff -a

fallocate -l 1G /swapfile

Resize the swap

• sudo dd if=/dev/zero of=/swapfile bs=128M count=8

sudo dd if=/dev/zero of=/swapfile bs=1024 count=1048576

if = input file
of = output file
bs = block size
count = multiplier of blocks

Make the file usable as swap

• sudo mkswap /swapfile

Activate the swap file

• sudo swapon /swapfile

Check the amount of swap available

• grep SwapTotal /proc/meminfo

But this settings will not survive a reboot.
To make the change permanent open the /etc/fstab file:

sudo nano /etc/fstab

and paste the following line:

/etc/fstab

/swapfile swap swap defaults 0 0

If you want to set the swap priority to 100, the command will changed to :

/swapfile swap swap defaults 0 100

References

• Create a Linux Swap File