If your terminal is also configured for light-on-dark, or if you're on a tty console, your screen should look something like this:
$ select-screen-profile
Select a screen profile:
1. plain
2. ubuntu-light ---- recommended
3. ubuntu-dark
Choose: 1-3 [2]: 3
My laptop can crank out about 7000 of these per second. So this should only take about 35 minutes to generate the complete table. You could also add an if-statement in there to spit out the matching passphrase, testing against the md5sum I gave you in the shadow file. And, again note that you could easily split this search across multiple CPUs for parallel scaling.
#!/bin/sh
# generate_passphrases.sh
CHARS="a b c d e f g h i \
j k l m n o p q r s t u v \
w x y z A B C D E F G H I \
J K L M N O P Q R S T U V \
W X Y Z 0 1 2 3 4 5 6 7 8 9"
for i in $CHARS; do
for j in $CHARS; do
for k in $CHARS; do
for l in $CHARS; do
md5=`echo -n "$i$j$k$l" | md5sum`
echo "$i$j$k$l $md5"
done
done
done
done
Start with the sha512sum of the following seed:Alright! So now you just need to write a simple shell script, right? Perhaps something like:
$ echo "Daemon" | sha512sum
Write the computed sum to the first line of a new file.
sha512sum that new file.
Write the resulting sum to the second line of the new file.
sha512sum that new file.
Write the resulting sum to the third line of the new file.
...
Continue doing this until you have exactly 1,000,000 lines in the file.
Hint: the first two lines of this file should be:
b48400ad0b48c66d6f7d538cc6aff0dab594dc78721059cae20d68322363924d6c5c6894d3eaee5c90b975188cbe0e2c2ea09d209a3dd263c425e84d0fbfd2be
c5f5f35f7661849a6a9ddaa6f8951c1bad7ea05777b15967a7f64248635a1e48153715182a95c125caf5d762417fcb3df05e6e6f279a65a860877b7def856207
...
Take the last 6 digits of the 1,000,000th line, and the first six digits of
line 961,325.
Take the difference of these two numbers.
Convert that difference from hex to decimal.
Who's number is this?
Nice, now watch your sums grow. It will crank through $i pretty quickly...at first. But note that with every run /tmp/sums is getting bigger. Every run is harder than the previous. Eeeek!
#!/bin/sh
sha=`echo "Daemon" | sha512sum - | awk '{print $1}'`
echo $sha > /tmp/sums
i=0
while [ $i -lt 1000000 ]; do
sha=`sha512sum /tmp/sums | awk '{print $1}'`
echo $sha >> /tmp/sums
i=`expr $i + 1`
done
In less than 30 seconds, this script will generate precisely the same output that the shell script above will produce 4 days later! Amazing, huh!
#!/usr/bin/python
import hashlib, os, sys
m = hashlib.sha512()
m.update("Daemon\n")
digest = m.hexdigest() + "\n"
m = hashlib.sha512()
out = file('growing.sums','w')
top = 1000000
for i in range(0,top):
out.write(digest)
m.update(digest)
digest = m.hexdigest() + "\n"
if (i % 100) == 0:
print "%d %0.1f\r" % (i, i * 100 / top),
sys.stdout.flush()
out.close()
561093-da7080 = 845FEDConvert that to decimal, and you get: 8675309.
Eight - Six - Seven - Five - Three - Oh - Niiiiieeeeeiiiiiine
...
I sincerely appreciate the kind words. Daemon (and the upcoming sequel) are labors of love to me, and knowing that they strike a chord with folks in the open-source community makes my day.
...
Daniel Suarez (aka Leinad Zeraus)
Note the sample artist, Jeff Luna...one of the funniest guys I know! Head over to http://cdbaby.com/cd/jeffluna where you can listen to this album for free, and buy it if you like it. My Mom's Van is my favorite ;-)
$ time seq 1 100 | xargs -i ecryptfs-unwrap-passphrase wrapped-passphrase {} 2>&1 >/dev/nullOkay, so I was able to attempt 100 passphrases in 23 seconds. That's about 4.3 attempts per second.
real 0m23.487s
user 0m21.481s
sys 0m0.180s
#!/bin/sh
# generate_passphrases.sh
CHARS="a b c d e f g h i \
j k l m n o p q r s t u v \
w x y z A B C D E F G H I \
J K L M N O P Q R S T U V \
W X Y Z 0 1 2 3 4 5 6 7 8 9"
for i in $CHARS; do
for j in $CHARS; do
for k in $CHARS; do
echo "$i$j$k"
done
done
done
$ ./generate_passphrases.sh > /tmp/passphrasesOkay, so we have 238,328 passphrases we need to test. So you could start crunch through these at this point, in a single thread, on a single system. But most laptops these days have multi-core cpu's. And you might even have access to multiple machines.
$ wc -l /tmp/passphrases
238328 /tmp/passphrases
#!/bin/sh
# try_passphrases.sh
INPUT="$1"
passphrases=`cat $INPUT`
total=`wc -l $INPUT`
i=0
for p in $passphrases; do
ecryptfs-unwrap-passphrase wrapped-passphrase "$p" 2>&1 >/dev/null
if [ "$?" = "0" ]; then
echo "FOUND: $p"
echo "FOUND: $p" | mail me@example.com
exit 0
fi
i=`expr $i + 1`
echo "($p) $i / $total"
done
exit 1
$ time ./try_passphrase.sh xap2 | tee /tmp/outAnd, in less than hour, one of the processes hit the winner:
FOUND: 9zDSweet, now, let's decrypt the mount passphrase:
$ ecryptfs-unwrap-passphrase wrapped-passphrase 9zDSo there's the mount passphrase! Also, let's look at that warning message briefly... I also gave the hint in the challenge that we're using the default eCryptfs salt. For added security, you should always consider adding a salt to such passphrases. There's a simplicity/complexity trade off. Just something you should consider in your own environments and use cases.
Unable to read salt value from user's .ecryptfsrc file; using default
904e740f417955457a5d56accd40ee01
$ mkdir /tmp/1 /tmp/2And we can view the contents of challenge_2.txt:
$ cp challenge_2.txt /tmp/1
$ sudo mount -t ecryptfs /tmp/1 /tmp/2
Select key type to use for newly created files:
1) pkcs11-helper
2) openssl
3) passphrase
4) tspi
Selection: 3
Passphrase: 904e740f417955457a5d56accd40ee01
Select cipher:
1) aes: blocksize = 16; min keysize = 16; max keysize = 32 (not loaded)
2) blowfish: blocksize = 16; min keysize = 16; max keysize = 32 (not loaded)
3) des3_ede: blocksize = 8; min keysize = 24; max keysize = 24 (not loaded)
4) twofish: blocksize = 16; min keysize = 16; max keysize = 32 (not loaded)
5) cast6: blocksize = 16; min keysize = 16; max keysize = 32 (not loaded)
6) cast5: blocksize = 8; min keysize = 5; max keysize = 16 (not loaded)
7) arc4: blocksize = 1; min keysize = 1; max keysize = 256 (loaded)
Selection [aes]: aes
Select key bytes:
1) 16
2) 32
3) 24
Selection [16]: 16
Enable plaintext passthrough (y/n) [n]: n
Attempting to mount with the following options:
ecryptfs_key_bytes=16
ecryptfs_cipher=aes
ecryptfs_sig=dfa024f15f2d12e7
Mounted eCryptfs
$ head -n 5 /tmp/2/challenge_2.txtPage down a bit, if you want the solution to this riddle. Or don't go below here, if you want to try and solve it for fun :-)
Challenge 2:
--------------------------------------------------------------------------
What is the full name of the research partner of a cryptographer who's first name matches the name of a dog on a tv show which has featured a vessel named after the video game played in a movie filmed near the largest city served by the United States telephone area code matching the greatest prime factor of the number of the asteroid named after the person possibly killed by a squeamish ossifrage?
--------------------------------------------------------------------------
Answer: ?
$ x=`grep "$USER.*Private.*ecryptfs" /proc/mounts | awk '{print $4}' | sed 's/sig=[0-9a-f]*,/sig=xxx,/'`Then, you can obtain the md5 message digest of $x with:
$ echo $x
rw,ecryptfs_sig=xxx,ecryptfs_cipher=aes,ecryptfs_key_bytes=16
$ echo $x | md5sumThe value, da9ae8f980ec845914220cfe727a8b16, is then used as the symmetric key to decrypt the challenge text file:
da9ae8f980ec845914220cfe727a8b16 -
Challenge 1:Daniel Suarez originally published Daemon under the pseudonym, Leinad Zeraus, which, of course, is his name spelled backwards.
--------------------------------------
Leinad_Zeraus : Daniel^Suarez
::
Nitsud_Dnalkrik : ???????????????
--------------------------------------
Dustin Kirkland (Twitter, LinkedIn) is an engineer at heart, with a penchant for reducing complexity and solving problems at the cross-sections of technology, business, and people.
With a degree in computer engineering from Texas A&M University (2001), his full-time career began as a software engineer at IBM in the Linux Technology Center working on the Linux kernel and security certifications, including a one-year stint as an dedicated engineer-in-residence at Red Hat in Boston (2005). Dustin was awarded the title Master Inventor at IBM, in recognition of his prolific patent work as an inventor and reviewer with IBM's intellectual property attorneys.
Dustin then first joined Canonical (2008) as an engineer (eventually, engineering manager), helping create the Ubuntu Server distribution and establishing Ubuntu as the overwhelming favorite Linux distribution in Amazon, Google, and Microsoft's cloud platforms, as well as authoring and maintaining dozens of new open source packages.
Dustin joined Gazzang (2011), a venture-backed start-up built around an open source project that he co-authored (eCryptFS), as Chief Technology Officer, and helped dozens of enterprise customers encrypt their data at rest and securely manage their keys. Gazzang was acquired by Cloudera (2014).
Having effectively monetized eCryptFS as an open source project at Gazzang, Dustin returned to Canonical (2013) as the VP of Product for Ubuntu and spent the next several years launching a portfolio of products and services (Ubuntu Advantage, Extended Security Maintenance, Canonical Livepatch, MAAS, OpenStack, Kubernetes) that continues to deliver considerable annual recurring revenue. With Canonical based in London, an 800+ work-from-home employee roster and customers spread across 40+ countries, Dustin traveled the world over, connecting with clients and colleagues steeped in rich cultural experiences.
Google Cloud (2018) recruited Dustin from Canonical to product manage Google's entrance into on-premises data centers with its GKE On-Prem (now, Anthos) offering, with a specific focus on the underlying operating system, hypervisor, and container security. This work afforded Dustin a view deep into the back end data center of many financial services companies, where he still sees tremendous opportunities for improvements in security, efficiencies, cost-reduction, and disruptive new technology adoption.
Seeking a growth-mode opportunity in the fintech sector, Dustin joined Apex Clearing (now, Apex Fintech Solutions) as the Chief Product Officer (2019), where he led several organizations including product management, field engineering, data science, and business partnerships. He drastically revamped Apex's product portfolio and product management processes, retooling away from a legacy "clearing house and custodian", and into a "software-as-a-service fintech" offering instant brokerage account opening, real-time fractional stock trading, a secure closed-network crypto solution, and led the acquisition and integration of Silver's tax and cost basis solution.
Drawn back into a large cap, Dustin joined Goldman Sachs (2021) as a Managing Director and Head of Platform Product Management, within the Consumer banking division, which included Marcus, and the Apple and GM credit cards. He built a cross-functional product management community and established numerous documented product management best practices, processes, and anti-patterns.
Dustin lives in Austin, Texas, with his wife Kim and their wonderful two daughters.
