Reputation: 1586
I observed that the size of the tmpfs
volumes created by docker
is roughly half the size of the machine's physical memory.
For example, on a machine with 22GB of RAM, I got this:
Filesystem Size Used Avail Use% Mounted on
overlay 970G 130G 840G 14% /
tmpfs 64M 0 64M 0% /dev
tmpfs 12G 0 12G 0% /sys/fs/cgroup
tmpfs 20G 0 20G 0% /ramdisk
/dev/sda1 970G 130G 840G 14% /etc/hosts
tmpfs 12G 180K 12G 1% /dev/shm
tmpfs 12G 0 12G 0% /proc/acpi
tmpfs 12G 0 12G 0% /proc/scsi
tmpfs 12G 0 12G 0% /sys/firmware
I would like to increase this size. Could anybody please tell me how to do that?
Thank you very much in advance for your help!
Update: Let me add some context to this question.
In my docker I have a /ramdisk
volume whose size is large (here 20GB) because one of my programs needs that:
nvidia-docker run --ipc=host -h $HOSTNAME --mount type=tmpfs,destination=/ramdisk,tmpfs-mode=1770,tmpfs-size=21474836480
When running the program, at the moment its memory usage surpasses 12GB of ramdisk
, it crashes (while ramdisk
still has 8GB left). Note that 12GB is the size of the other tmpfs
system volumes.
So, one solution I could think of is to increase the size of those other volumes, which is my question.
Upvotes: 2
Views: 8882
Reputation: 826
As per docker docs (20GiB as per example):
docker run -d \
-it \
--name tmptest \
--mount type=tmpfs,destination=/app,tmpfs-mode=1770,tmpfs-size=21474836480 \
nginx:latest
PS: Docs specify that by default tmpfs volumes have unlimited size, so the calculations here might be rounded down to the amount of free memory in the host OS.
SRC: https://docs.docker.com/storage/tmpfs/#specify-tmpfs-options
Upvotes: 2