Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Comments
Safely Storing User Passwords: Hashing vs. Encrypting
Newest First  |  Oldest First  |  Threaded View
Jose ManuelG297
50%
50%
Jose ManuelG297,
User Rank: Apprentice
7/14/2017 | 11:56:35 AM
Re: Good overview
Nice article, there's another issue that needs to be taken into account: the moment when you hash.

In a typical multi-tier architecture, the user application (client) requests the password which is then sent to a server during the login time. Subsequent requests usually use a session token.

The hashing of the password can be done either on the client side, on the server side, or on both.

1- If done by the client, even if the communication channel is attacked, the original password is safe, avoiding the login through the application client user interface... but towards the server, the hashed password is effectively the password, so if we get to the hash password database, we can authenticate against the server.

2- If the hashing is done at server side, the problem is that the password needs to be secured while in transit.

3-A mixed of both is to encrypt or hash between the client and the server, and again on the server side. This looks the safest way to me.

I would like to know your thoughts about it...

Thanks and best regards
dgp101
50%
50%
dgp101,
User Rank: Apprentice
9/7/2016 | 8:54:39 PM
Re: Good overview
Good question. The point of the salt is that it is different for each account. It's true that you could use it, together with the knowledge of the hashing algorithm and number of iterations, to produce one rainbow table for each user, one user at a time. However, without the salt you just make the rainbow table once -- even before compromising the system -- and then compare every user's hash to your rainbow table and reveal many passwords quickly.

With a hash, you have to (1) wait till you know my salt to make the rainbow table for my account, and (2) when/if you find one match, that's the only one you'll get with it because every password in the table also has a unique salt.

Finally, using "slow" hashing algorithms with many cycles can increase the amount of time per attempt to a point where even making one complete rainbow table should take years. That works in concert with the salt: The salt makes you do X amount of work per user you want to crack instead of X amount of work to get "All/most users" - and the iterations (slowness) make you do 1000X the amount of work PER attempt. These two work together to make cracking too costly to be useful. Ideally that cost should be on the order of requiring a large cluster of dedicated hardware to churn for many years to crack one password. That's good since future hardware will be faster than today's.
rhusain0
50%
50%
rhusain0,
User Rank: Apprentice
11/17/2015 | 3:39:35 PM
Re: Good overview q re password transmission
How do u handle securing password transmission on over the air. Or capture of hash over air Im doing password for iot and looking for best practices Rafe
rhusain0
50%
50%
rhusain0,
User Rank: Apprentice
11/17/2015 | 3:39:33 PM
Re: Good overview q re password transmission
How do u handle securing password transmission on over the air. Or capture of hash over air Im doing password for iot and looking for best practices Rafe
TejGandhi1986
50%
50%
TejGandhi1986,
User Rank: Apprentice
6/13/2014 | 9:01:50 PM
Preventing the password file from getting stolen
Thanks for the article,itis very informative and provide details on the foundations related to hashing and encryption.

Considering different chain of thoughts ,along with encryption and hashing that is used to secure passwords it is essential that the password file is well protected SAM file in windows and etc/shadow or etc/passwd file in Linux access must be restricted with multiple layers of defense to prevent it from getting stolen.

Thanks

Tej Gandhi

 
MichaelCoates
100%
0%
MichaelCoates,
User Rank: Author
6/12/2014 | 4:16:00 PM
Re: Good overview
Good question. The stolen database would indeed include the salts. However, exposure of random per-user salts does not undermine their purpose and security benefits.

There are two benefits to using per-user salts

1. When using per-user salts an attacker cannot simply review the stolen password hash databse for duplicate hashes (which would indicate the same original password for both accounts). The introduction of a per-user salt ensures that even the same password will result in unique hashes.

2. An attacker cannot download a rainbow table and use it against the password hashes. A rainbow table is a large database of precomputed hashes for a variety of common passwords (or even all possible passwords of certain character sets and lengths). Without per-user salts an attacker could do a simple lookup of the stolen hash within the rainbow table to determine the original password. The introduction of per-user salts means the rainbow table is useless.


Sure, an attacker could incorporate the salt into a brute force attack. But the purpose of a salt isn't to stop brute force. It's to accomplish the two items mentioned above (duplicate hashes and rainbow tables). As I mentioned in the article, the iterative hashing approach that exists in bcrypt/scrypt/PBKDF2 is the defense against brute force attacks on the password hash.

 

Hope that helps.

Michael

 
chiefwilson
50%
50%
chiefwilson,
User Rank: Apprentice
6/11/2014 | 9:57:04 PM
Re: Good overview
Michael,

Thank you for a well-written article. I agree that hashing passwords with added salt provides far greater security than simply encrypting passwords. My question is simple: If a malicious actor steals a database of password hashes, won't this database include the salts as well, thereby nullifying the purpose of the salt, which is to defend against brute force dictionary and rainbow table attacks?
Marilyn Cohodas
100%
0%
Marilyn Cohodas,
User Rank: Strategist
6/4/2014 | 2:51:46 PM
Re: Good overview
Thanks, Michael. One of the things I've been hearing about more and more is that personal information has become much more valuable a target for cybercrime than, for example credit cards. If that's the case, then your message about hashing versus encryption is one that InfoSec pros should definitely take to heart. 
MichaelCoates
100%
0%
MichaelCoates,
User Rank: Author
6/4/2014 | 1:02:50 PM
Re: Good overview
The largest misconception is that since encryption is good for protecting information in some situations it is therefore appropriate for all situations involving sensitive data. As discussed above, encryption is really the wrong choice for protecting passwords.

Second, that any hashing algorithm is sufficient for password hashing. Selecting a weak algorithm like md5 or failing to user per user salts places passwords at extreme risk if the hash file is stolen.

-Michael
Marilyn Cohodas
100%
0%
Marilyn Cohodas,
User Rank: Strategist
6/4/2014 | 12:34:09 PM
Good overview
Thanks for your detailed overview, Michael. You say that to propertly secure user invormation today, application developers must starts with "a proper understanding of fundamental security controls and the protection of user passwords using modern hashing algorithms." What do you think is the biggest misunderstanding of security that app developers have?


COVID-19: Latest Security News & Commentary
Dark Reading Staff 9/25/2020
Hacking Yourself: Marie Moe and Pacemaker Security
Gary McGraw Ph.D., Co-founder Berryville Institute of Machine Learning,  9/21/2020
Startup Aims to Map and Track All the IT and Security Things
Kelly Jackson Higgins, Executive Editor at Dark Reading,  9/22/2020
Register for Dark Reading Newsletters
White Papers
Video
Cartoon
Current Issue
Special Report: Computing's New Normal
This special report examines how IT security organizations have adapted to the "new normal" of computing and what the long-term effects will be. Read it and get a unique set of perspectives on issues ranging from new threats & vulnerabilities as a result of remote working to how enterprise security strategy will be affected long term.
Flash Poll
How IT Security Organizations are Attacking the Cybersecurity Problem
How IT Security Organizations are Attacking the Cybersecurity Problem
The COVID-19 pandemic turned the world -- and enterprise computing -- on end. Here's a look at how cybersecurity teams are retrenching their defense strategies, rebuilding their teams, and selecting new technologies to stop the oncoming rise of online attacks.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2020-15208
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, when determining the common dimension size of two tensors, TFLite uses a `DCHECK` which is no-op outside of debug compilation modes. Since the function always returns the dimension of the first tensor, malicious attackers can ...
CVE-2020-15209
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, a crafted TFLite model can force a node to have as input a tensor backed by a `nullptr` buffer. This can be achieved by changing a buffer index in the flatbuffer serialization to convert a read-only tensor to a read-write one....
CVE-2020-15210
PUBLISHED: 2020-09-25
In tensorflow-lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, if a TFLite saved model uses the same tensor as both input and output of an operator, then, depending on the operator, we can observe a segmentation fault or just memory corruption. We have patched the issue in d58c96946b and ...
CVE-2020-15211
PUBLISHED: 2020-09-25
In TensorFlow Lite before versions 1.15.4, 2.0.3, 2.1.2, 2.2.1 and 2.3.1, saved models in the flatbuffer format use a double indexing scheme: a model has a set of subgraphs, each subgraph has a set of operators and each operator has a set of input/output tensors. The flatbuffer format uses indices f...
CVE-2020-15212
PUBLISHED: 2020-09-25
In TensorFlow Lite before versions 2.2.1 and 2.3.1, models using segment sum can trigger writes outside of bounds of heap allocated buffers by inserting negative elements in the segment ids tensor. Users having access to `segment_ids_data` can alter `output_index` and then write to outside of `outpu...