Redis: Warning Memory overcommit must be enabled

[ManOki]

WARNING Memory overcommit must be enabled! Without it, a background save or replication may fail under low memory condition.

If you see a warning like this, you have to enable overcommit on your host machine (outside docker).

Solution

• echo "vm.overcommit_memory = 1" | sudo tee /etc/sysctl.d/nextcloud-aio-memory-overcommit.conf should be set, to enable Memory overcommit on reboot, change will apply after reboot or see the line below how to activate it temporary without reboot
• sysctl "vm.overcommit_memory=1" enables Memory overcommit on-the-fly temporary, then restart Nextcloud AIO from the AIO Interface

Note for Docker Desktop instances: See https://stackoverflow.com/a/69294687 for how to apply sysctl config values.

Background

Due to the fact that (for good reasons) the Redis container does not run in privileged mode, there is no way that Nextcloud AIO or Redis can enable this for you, i.e., /proc is read-only for containers.

See Redis documentation, corresponding docker issue and kernel documentation

[szaimen]

Wow, thanks a lot for this nice guide! :)

[alkhachatryan]

it helped me with laravel, psql and redis stack

[loren-klemesrud-veritone]

This was so helpful!

[NiciTheNici]

Thank you! This has been an issue for a while for me <3

[Irdiism]

Just noticed the issue on one of my Nextcloud instances and solved it in a couple of minutes. Thanks a lot for your contribution!

[woiwoiwoiwoi]

Unfortunately, I can't get this to work, and am thankful for any pointers.

I hope I'm correct in understanding that I am to run these commands on my host system? My host system is a QNAP NAS. The first command results in No such file or directory, so I created /etc/sysctl.d and re-ran the command, which succeeded. Then I ran the second command, to which the response is sysctl: error: 'vm.overcommit_memory=1' is an unknown key. I then tried the instructions from the redis log, which are slightly different, and added 'vm.overcommit_memory = 1' to /etc/sysctl.conf, which produced the same error message.

What can I do to fix this?

Also, is this a rare misconfiguration on some systems (like mine) or is this a common error message that most people don't notice since it's buried in the redis logs?

[EVOTk]

Hello,

I think you need to create the file manually on a QNAP NAS. This command should allow you to do that:
echo 1 > /proc/sys/vm/overcommit_memory

QTS will delete this file each time the NAS is booted or restarted, but you can solve this problem by including it in startup.sh : https://www.qnap.com/en/how-to/faq/article/running-your-own-application-at-startup

[woiwoiwoiwoi]

Fantastic, this works, thank you!

Regarding autorun.sh, I'll report back once I've tested it properly.