cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 
Newcomer I

Re: NIST new ruling on passwords

Auditors dont know everything and if they push back and if you state that you align with NIST CSF hit them with NIST Special Publication 800-63 Rev 3, Digital Identity Guidelines

Newcomer III

Re: NIST new ruling on passwords

While all good suggestions, why has nobody mentioend password managers?

 

I'd much rather see corporate support for a "blessed" Password Manager, and require its use.  Each site gets a random, unique, strong password.  No need for constant password rotation, easy to change a specific password in case of a site-specific breach.

 

The most secure password is one the user *can't* remember.

 

This circumvents most of these issues.

Newcomer I

Re: NIST new ruling on passwords

It's a kind of horses-for-courses thing. I have about four long passwords that I memorise at any one time: two are for password managers, one for my laptop, one for my desktop PC. The rest go in the password vault.

 

Worth also thinking about a proper product over the freemium cloud based apps. I use LastPass at home for passwords that are mine: for shared system passwords (eg root passwords for systems) a system that logs when passwords are checked in and out, and by which users, is a better bet. You'd probably be looking at Thycotic or CyberArk for that.

 

James

Newcomer III

Re: NIST new ruling on passwords

James, your use case is almost exactly like mine.  Long, memorised passwords for my machine login and password manager. Everything else in password manager.  Which is great for personal use.

 

I agree that for corproate use, you'd want something with more access control and logging and central admin features.

Newcomer I

Re: NIST new ruling on passwords

I can see it was mentioned about the use of password managers. These are great tools to help protect passwords without the user ever having to see the password or having to think about creating them. The problem here is that NIST guidelines are recommended for all users, so when you have a large user community the cost becomes prohibative. This is why tools such as BeyondTrust, CyberArk, Thycotic typically are used for privileged accounts. Of course if you have an organisation with unlimited budget then hey presto, however its never that simple.

 

If you are advising a small number of individuals then they're some great tools out there such as 1password, lastpass, keepass etc. These all help individuals manage a large set of passwords and help with generating and auditing the vast number of accounts we all need these days.

 

Implementing something wider and larger for all users would require going to a fully fledged Identity and Access Management Solution! Great, got 2 years and maybe 2 mil to spare then sorted.

 

Look at the scale of the problem, what solutions you already have and if they are being used to their best ability and work from there. This way you can decide on the appropraite solution!

Newcomer I

Re: NIST new ruling on passwords

Agree with the others that password managers should be encouraged for use. This is especially useful to those in IT operations.

Viewer II

Re: NIST new ruling on passwords

 At a Cyber Security briefing delivered by an ex GCHQ professional we were shown how readily available software can guess a password of 12 characters in 2 seconds. Therefore, recommending that for every additional character security was increased. Surely recommending 8 then is outdated?  Pass phrases or several unconnected words and no forcing password changes were also recommended as more effective and secure controls. You will never stop people writing down passwords somewhere if they have complex, hard to remember passwords. Because of this it was also suggested that it was better to advise people that if passwords are written down and not in a password manager, then store them away from any PCs, media etc that uses them, rather than say 'don't ever write them down', because inevitably some people will.  Although we recommend the above now as part of 27001 consultancy we find we are often in conflict with outsourced IT service providers who set the password policy at 8 complex characters with forced changes.

 

How would others faced with this challenge address this?

Viewer II

Re: NIST new ruling on passwords

I've already had users asking about this.  As soon as I can identify an efficient way to check the passwords against a common dictionary attack, I will consider implementing.  Anyone using a tool they would recommend that is cost-effective (or free), secure and can be automated (or at least requires minimal manual intervention)?  Are you checking these quarterly?

Newcomer I

Re: NIST new ruling on passwords

A very good and free solution is Hydra. You can pass it a list of users you wish to check and a password you want to test/guess or you can provide a dictionary of passwords you wish to validate. The only problem with the last option is if you run this against active directory and you have a lock out policy configured after so many attempts then you could potentially lock out your users. A good way around this (providing you have a reset on the counter after so many minutes) is to implement a delay between checks, e.g. check users against password1, wait for a given period of time greater than that of your reset time, check users against password2, repeat steps.

Newcomer III

Re: NIST new ruling on passwords

As a one-day hack-a-thon project, I downloaded the 300 million + sha1 hashes from the breach corpus Troy Hunt made available, and on a t2.micro instance in AWS, I could search for and compare (with a binary search, the hashes are sorted) in sub one second.  Refusing all passwords which appear in that big a corpus of known breached passwords is pretty good coverage.  It's just a chunk (about 12 gig uncompressed) of disk and a teeny bit of CPU to check.