Tag Info

New answers tagged

0

I guess it depends on what your threat model is, if the NSA are after you as a person of interest, full disk encryption is the least of your problems, you should probably just stick to a one time pad and abacus. If you want to keep a potential thief from accessing your documents if your laptop should get stolen, any of the closed source full disk encryption ...


0

There's no real answer to that question other than that you cannot completely trust anything you didn't put together yourself. It's unfortunate, but true. I am certain there are many other available open-source options for WDE, which may allay some of your fears in trusting a company. Then again, some open-source software has been compromised by ...


1

As far as other Certificate Authority (CA) root certificates security is concern, then there are no issues expect if you are going to implement an authorized Certificate Authority SSL certificate. If the security certificate is from Trusted Certificate authority then it never cause any security trouble.


10

So we had to install their root certificate "PrivateCompany Root CA" to our web server. Why? Globally trusted CAs are useful in general purpose clients like browsers. But if you consume specific web services from a custom client you can add that CA locally. Every decent SSL client allows you to influence certificate validation. For example by ...


1

To some extent what the company can do is limited to The Name Constraints (if any are applied) The Enhanced Key Usages (if any are applied) Of course this depends on client software properly validating each and every intermediate certificate in the chain, up to and including the root. It's possible that some client software doesn't validate the chain ...


6

There is only full trust with the CAs you've installed. This means, that there is no restriction which certificates a trusted CA can sign. So it can also sign fake certificates for sites they don't own (e.g. banking.com) and you will accept them. I don't know how you communicate with the partner, but with languages like Perl, Python etc you can specify a CA ...


0

To your initial question, the only thing I've seen that purports to really do this is the Gartner Magic Quadrant. However, given the overall complexity of firewall solutions, it's hard to make an apples to apples comparison in my opinion. Often they are used for their basic layer 3 and layer 4 filtering capabilities, and as such are treated like commodity ...


0

Create a cryptographic random number and add it as a parameter in the url (token). Then only show the real page if it's a valid token. Tokens should be store in the database obviously. How safe is it to provide a link within that app generated email to allow his friends to directly respond to the question without having to register on the site? It ...


1

Including a unique identifier or token in each individual e-mail is the best bet. This token should be random, hard to guess, and from a large enough set that randomly entering values will never produce a valid token. These tokens should then be limited in the scope and scale of what they can do. If you limit the users to a single response and ...


2

I would create tokens (Some random hashes or GUID) for each intended user and pass that in the query string to identify the individual users. Kind of like a temporary password. Putting this on the query string is convenient for the user. Making the GUID, or token temporary or for one time use will reduce sharing. Your only concern after this is if your users ...



Top 50 recent answers are included