Date: 2026apr7 Update: 2026apr10 OS: Linux Language: bash Q. Linux: Make a fixed size file of random bytes A. You might want to do this to reserve some disk space case of an emergency. There are several ways, here they are ordered by my preference: 1. With fallocate and shred

# Make it fallocate -l 1G /home/myrandomfile # Randomize it shred --iterations=1 /home/myrandomfile

The `fallocate` command uses the fallocate() system call which seems like the best way of conveying your intentions to the filesystem. https://man7.org/linux/man-pages/man2/fallocate.2.html By default `shred` does 3 iterations so we add `--iterations=1` to reduce the number. This is random enough for our purposes and faster. 2. With truncate and shred

# Make it truncate --size 1G /home/myrandomfile # Randomize it (same as above) shred --iterations=1 /home/myrandomfile

Its safe to bet that the `truncate` command uses the ftruncate() system call. https://man7.org/linux/man-pages/man2/truncate.2.html Its been around since 2007 so any distribution will have it by now. 3. With head

head -c 1G /dev/urandom > /home/myrandomfile

Modern `head` copes with binary data. 4. With dd Use `dd` to read random bytes into a file:

dd if=/dev/urandom of=/home/myrandomfile bs=1M count=(size in megabytes)

For example, to make a 1GiB file:

dd if=/dev/urandom of=/home/myrandomfile bs=1M count=1024

This way is fine, but the syntax isn't so pretty.