Score:1

Performance test went wrong and the dd command created 13TB of data on /dev/mapper/device. Why system didn't crash? HDD-250GB

gs flag

So I wanted to do some performance test with encrypted and normal data storage on my embedded device.

That is not what I was expected to see at all!

Can you please explain it to me what just happend. Why dd comand output was 1843200+0 records in and out but df -h show file system disk space usage as 13TB?

This is my workflow:

dd if=/dev/urandom of=enc_per_test.img bs=512 count=2097152

dd if=/dev/urandom of=normal_per_test.img bs=512 count=2097152

And receive 2 images 1GB each - as I predicted.

losetup /dev/loop1 enc_per_test.img 

losetup /dev/loop2 normal_per_test.img

After that I perform:

dmsetup -v create enc_per_test --table "0 $(blockdev --getsz /dev/loop1) crypt <crypt_setup> 0 /dev/loop1 0 1 sector_size:512"

mkfs.ext4 /dev/mapper/enc_per_test

mkdir /mnt/enc_per_test

mount -t ext4 /dev/mapper/enc_per_test /mnt/enc_per_test/

As I expected df-h showed mounted enc_per_test:

 Filesystem ############## Size ### Used ## Avail ## Use% ### Mounted on #####           

 /dev/mapper/enc_per_test ## 976M ## 2.6M ## 907M ## 1% #### /mnt/enc_per_test

I clear cache:

echo 3 > /proc/sys/vm/drop_caches

And finally perform dd comand to fill up the enc_per_test:

time dd if=/tmp/random of=/dev/mapper/enc_per_test conv=fsync

1843200+0 records in
1843200+0 records out
943718400 bytes (944 MB, 900 MiB) copied, 152.098 s, 6.2 MB/s

So I was like, ok that's fine. This is what I wanted. Let's see how it's look like in df -h:

 Filesystem ############## Size ### Used ## Avail ## Use% ### Mounted on #####           

 /dev/mapper/enc_per_test ## 13T ## 13T ## 0 ## 100% #### /mnt/enc_per_test

What happend here?

Thank you!

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.