|Travis Burtrum 53b1fbb689||2 years ago|
|.ci||2 years ago|
|.gitignore||2 years ago|
|.travis.yml||2 years ago|
|LICENSE.md||2 years ago|
|Makefile||2 years ago|
|README.md||2 years ago|
|pegh.c||2 years ago|
|test.sh||2 years ago|
pegh is a file encryption tool using passwords and authenticated encryption. It is simple, secure, and returns proper exit codes so you can tell whether encryption or decryption failed or not.
pegh is Klingon for secret
# encrypt file.txt to file.txt.pegh with password SUPER_SECRET_1942 pegh -e SUPER_SECRET_1942 <file.txt >file.txt.pegh # decrypt file.txt.pegh to file.txt with password SUPER_SECRET_1942 pegh -d SUPER_SECRET_1942 <file.txt.pegh >file.txt # make enrypted backup tar czv -C /path/to/dir/ . | pegh SUPER_SECRET_1942 -o foo.tar.gz.pegh # extract encrypted backup pegh SUPER_SECRET_1942 -d -i foo.tar.gz.pegh | tar xzv
The easiest way to scale cost/time it takes for bruteforcing is simply to continue doubling -s, on both encryption and decryption commands.
$ pegh -h usage: pegh [options...] password -e encrypt input to output, default mode -d decrypt input to output -i <filename> file to use for input, default stdin -o <filename> file to use for output, default stdout -a append to -o instead of truncate -c <max_mb> maximum megabytes of ram to use per encrypted chunk, so while decrypting/encrypting twice this will be used, the same amount will be needed for decryption as encryption and is saved in the file format, so decryption will fail if this isn't set high enough, these are only allocated after scrypt is finished so max usage will be the highest of these only, not both combined, max: 2047, default: 16 -m <max_mb> maximum megabytes of ram to use when deriving key from password with scrypt, applies for encryption AND decryption, must almost linearly scale with -N, if too low operation will fail, default: 64 -N <num> scrypt parameter N, only applies for encryption, default 32768 this is rounded up to the next highest power of 2 -r <num> scrypt parameter r, only applies for encryption, default 8 -p <num> scrypt parameter p, only applies for encryption, default 1 -s <num> multiplication factor to apply to both -N and -m for easy work scaling, rounded up to the next highest power of 2, BEWARE: -s 32 requires 2G ram, -s 64 requires 4G and so on, default: 1 -h print this usage text -q do not print error output to stderr -V show version number and format version support then quit For additional info on scrypt params refer to: https://blog.filippo.io/the-scrypt-parameters/ https://tools.ietf.org/html/rfc7914#section-2
pegh implements a simple versioned file format so encryption parameters can change in the future. Numbers here are inclusive 0-based byte array indices, 0th byte is always version number, everything else depends on version number, currently only version 0 exists.
Version 0, scrypt key derivation, aes-256-gcm encryption, 43 byte header, 16 byte auth tag per chunk. The 12-byte IV for the first chunk is 0, and is incremented by 1 for each successive chunk, if it ever rolls back over to 0 encryption should be aborted (chunk size should be increased).
|0||8 bit unsigned byte||pegh file format version|
|1-4||32 bit unsigned integer in big endian order||scrypt N parameter|
|5||8 bit unsigned byte||scrypt r parameter|
|6||8 bit unsigned byte||scrypt p parameter|
|7-10||32 bit unsigned integer in big endian order||aes encrypted chunk size|
|11-42||32 randomly generated bytes||scrypt key derivation seed|
|43+end||any number of chunks, chunk_size + 16 long||chunks followed by AES-256-GCM auth tag|
pegh.c: AGPLv3 for now, message me if you have a problem with this
documentation/file format: consider this your choice of MIT, Apache 2, or public domain