ํ์ดํ ์น dataloader๋ก ํ์ตํ ๋ฐ์ดํฐ๋ฅผ ๋ถ๋ฌ์จ ํ ํ์ต์ ์งํํ ๋ ๋ค์๊ณผ ๊ฐ์ ์๋ฌ๊ฐ ๋ฐ์ํ์๋ค..
RuntimeError: DataLoader worker (pid 21427) is killed by signal: Bus error. It is possible that dataloader's workers are out of shared memory. Please try to raise your shared memory limit.
์ด๋ docker ์ปจํ ์ด๋๋ฅผ ๋์ธ ๋ ๋์คํฌ ์ฉ๋์ ๊ณต์ ํ๋๋ฐ ๋ฉ๋ชจ๋ฆฌ๊ฐ ๋ถ์กฑํด์ ๋ฐ์ํ๋ ์๋ฌ์ด๋ค.
shm ์ฌ์ด์ฆ๊ฐ ์ปจํ ์ด๋๋ฅผ ์์ฑํ ๋ ๋ํดํธ๊ฐ์ด 64m ์ด๋ค
๋ฐ๋ผ์
docker run --name 1_ne_ -ti --shm-size=400G -v
์ด์ ๊ฐ์ด ์งํ ํด์ฃผ๋ฉด
๋ฉ๋ชจ๋ฆฌ๊ฐ ์ฆ๊ฐํจ์ ์ ์ ์๋ค.
1ne : ์ปจํ ์ด๋ ์ด๋ฆ
--shm-size=400G : ์ฉ๋ 400G ์ค์
overlay 879G 739G 95G 89% /
tmpfs 64M 0 64M 0% /dev
tmpfs 126G 0 126G 0% /sys/fs/cgroup
shm 400G 28K 400G 1% /dev/shm
/dev/sdb1 3.5T 2.7T 576G 83% /Data
/dev/sda2 879G 739G 95G 89% /Files
tmpfs 126G 12K 126G 1% /proc/driver/nvidia
tmpfs 26G 1.5G 24G 6% /run/nvidia-persistenced/socket
udev 126G 0 126G 0% /dev/nvidia0
tmpfs 126G 0 126G 0% /proc/asound
tmpfs 126G 0 126G 0% /proc/acpi
tmpfs 126G 0 126G 0% /proc/scsi
tmpfs 126G 0 126G 0% /sys/firmware