I need to setup a 2 way SSL communication channel between a .NET/WCF application and a third party web server. Right now I am trying get a successful handshake with the host in order to validate that all the elements are setup correctly (client certificate, server certificate authority, network communication...). I'm using the openSSL command line tool to try and validate this, using the s_client command.
Here is what is stopping me, and what I don't understand:
whatever I do, openSSL is expecting to find a private key for the
client certificate
the client certificate was given to me by the third party, but it
does not contain any private key
if I just generate my own private key file using openSSL, I'm getting
a key values mismatch error
Keep in mind that I have just started getting my hands into SSL so I have a very basic understanding of the whole protocol. From what I've been reading, it seems that both server and client need their private key in a 2 way SSL setting. However, I can't figure out how to get a working private key on my client (working with the client certificate that was given to me).
I would very much appreciate if somebody could shed some light on client certificate private keys, as this is giving me a major headache.
Certificates on their own are only public pieces of information. What links a public key certificate to the name it contains is the fact that whoever has legitimate control over that name (e.g. your name or your server's name) also has the private key for it.
Certificates are used to prove the identity of the remote party by challenging the remote party to perform an operation that can only be done with the corresponding private key: signing something (which can be verified with the public key) or deciphering something that was encrypted with the public key. (Both can happen in the SSL/TLS handshake, depending on the cipher suite.)
During the SSL/TLS handshake, the server sends its certificate (in clear) and proves to the client that it has the corresponding private key using an authenticated key exchange.
In your case, you also want to use client-certificate authentication. It's not enough to send the client certificate during the handshake: the client must also prove it has the private key. Otherwise, anyone who receives that certificate could clone it. The point of using certificates is to prevent any cloning, in such a way that you never have to show your own secret (the private key).
More specifically, the client has to sign the handshake messages in the Certificate Verify message of the TLS handshake so that the server can verify it against the public key sent in the client certificate. Without this step, no client-certificate authentication would be taking place.
the client certificate was given to me by the third party, but it does
not contain any private key
Giving you your certificate without its private key seems a bit pointless, unless you're expected to have generated a certificate request on your side beforehand (in which case you would have the private key).
Indeed, rather than being given a certificate and its private key, it's better practice for you to generate your key-pair, create a certificate request (CSR), and have your CA issue a certificate from that CSR (but without them ever knowing your private key). In this case, you would be expected to have kept your private key, and you would be able to use it with the cert you would have received.
This is an interesting question. I was puzzled by TLS related errors while configuring LDAPS, my learnings here.
The names, "public key" and "private key" sound like it's an one-way operations, but it works both ways, anything encrypted by one key and be decrypted by another key - just that one key is made public
Info = openly available data, with unknown integrity - it could be tampered by 3rd party
Signature = Info encrypted by a private key, accompanies Info
Verifying signature = use public key to decrypt the signature and get Info back thus proving "whoever created the signature knows the private key", or, the Info is untouched after it's signed
That is, unless the private key is stolen
Certificate = Info + Signature + public key + private key
Info = the meat
Signature = to verify integrity of Info
public key = to verify other certificates, see Certificate Chain
private key = not included in certificate itself but should be kept by certificate owner, could be used to sign other certificates
Certificate Authority (CA) = Anyone who holds a public / private key pair and creates certificates
CA identity is indicated in Info on a certificate
Certificate Chain = Use public key on mother-certificate to verify the signature on a child-certificate, to ensure child-certificate was created and signed by mother CA
And mother created and signed by grandma, so forth
Root CA = Root CA is the CA you trust unconditionally
Because their certificates (Root Certificates) are pre-loaded in your browser or OS as a source of trust
Also because their certificates are self-signed, its Signature is created using its own private key, verified by its own public key
This could be confusing, but think it as the origin of certificate-based encryption - only when people found out that relying on a few Root CA's to create and sign countless certificates are insecure (private keys might get stolen) and impossible (rapidly increasing demands of certificates), they came up with the idea of certificate chains
Root CA's only create and sign certificates for some Intermediate CA's so private keys are not used too often and is safer
Intermediate CA's can create and sign millions of certificates and get their certificate revoked when a private key is known to be leaked
This is the reason you don't often see the Root Key Ceremony, which signifies the importance of the private key for a Root Certificate: the less the private key is used, the less it could be exposed
Self-Signed Root Certificate = virtually the same as a Root Certificate - just no one but the creator would trust
TLS Handshake = server sends certificate, client validate is against the certificate chain, when verified, exchange a "password" securely
The "password" is then used to encrypt actual data transfer using symmetric encryption
Asymmetric encryption is not used for actual data due to its need for intense computation powers
Client Certificate Authentication = in addition to previous process which authenticates the server, the server demands a valid certificate to be provided by the client, also a signature of previous handshake messages
The signature is created using the private key of the certificate and can be verified by the public key on the certificate
This proves the ownership of the certificate
This is used usually for restricting access to known individual clients
Using certificates means no man-in-the-middle or replay attack could happen
This is often seen in finance or IoT
This is also utilized by LDAPS, which caused mystic error messages when I had a valid certificate in my local keystore but it was not accompanied by its private key
...So, in terms of openssl, when you use -cert you indicate Client Certificate Authentication, with which you will need -key to specify the corresponding private key.
However, for many Web scenarios, even if the certificate for a server is signed by a homemade CA, one only need to add the homemade CA to truststore or keychain then server certificate can be validated. This is done by using -CAfile.
Bottom-line.. If you're sending client certificates, then you would need the client certificates' private key to encrypt the pre-master secret
Related
I'm looking at this example, and I can see that as part of client authentication the user should pass the --key holding the private key used for certificate signing request to the curl command, I'm not sure why this is needed if it's for encryption shouldn't the public key of the server to be used?
Having the certificate itself doesn't prove anything. Certificate is never secret, can be shared freely and as such multiple parties could be in a possession of a valid client certificate.
As part of the TLS handshake with mutual authentication, client sends CertificateVerify message to prove it also has the private key matching the certificate it sent in Certificate message before. With TLS server already having client's certificate (and thus its public key), it can verify the signature.
I have done some research on this but still have some trouble connecting the dots as to what exactly happens when the server send its certificate to the client in terms of verifying signature and root certificates.
When you create a CSR, it generates a private key along with it, and you send it to the CA to generate the cert and sign it, but doesn't the CA use it's own private key to sign? So what is the point of the private key you generated with your CSR?
When server sends its certificate for client to validate, how exactly does client validate that it is a valid CA cert. It has a collection of trusted CA certs, ok - but how exactly are they used to verify that it was a valid CA that signed the server's certificate using the signature and public key of the server certificate? What things are compared to make sure it was not forged?
Is there any point in encrypting your internal self signed certs? How about an internal root cert? Or is the private key the only one worth encrypting?
If we don't keep a database of encrypted data for our web service (over SSL) for example, would we ever care about storing our own private key once we generated the self signed cert, and if we do, they why?
When you create a CSR, it generates a private key along with it
Or you have already generated your own private key.
and you send it to the CA to generate the cert and sign it
You send the CSR. You don't send your private key. It's private. You don't send it to anyone.
but doesn't the CA use it's own private key to sign?
Yes.
So what is the point of the private key you generated with your CSR?
It pairs with the public key contained in the certificate and it is part of the process used to prove that you and only you own that certificate, as only you can generate digital signatures with that private key that can be verified by the public key in the certificate.
When server sends its certificate for client to validate, how exactly does client validate that it is a valid CA cert. It has a collection of trusted CA certs, ok - but how exactly are they used to verify that it was a valid CA that signed the server's certificate using the signature and public key of the server certificate? What things are compared to make sure it was not forged?
The certificate itself is verified, by verifying its digital signature; it is checked for being within its validity period; and then an attempt is made to form a certificate chain using the alleged signer of the certificate (the 'issuer') and the trusted certificates in the collection.
Is there any point in encrypting your internal self signed certs?
No. They are public documents. Only the private key is private, and that isn't in the certificate.
How about an internal root cert?
No.
Or is the private key the only one worth encrypting?
Yes.
If we don't keep a database of encrypted data for our web service (over SSL) for example, would we ever care about storing our own private key once we generated the self signed cert, and if we do, they why?
Because it's private. It is a critical part of your identity. If you leak it, others can impersonate you.
As I understand the working of digital certificates:
1. The server generates public and private key
2. It then generates a CSR and submits it to the CA
3. The CA signs it with its private key and returns the certificate to server
While sending data to server:
1. Encrypt the data using server's public key and transmit. Only server will be able to decrypt it as it has the private key
While downloading data from server:
1. The server encrypts the data using its private key. It trasmits the certificate and the data.
2. User decrypts the certificate using CA's public key.
3. Then decrypts the data using the public key obtained from decrypted certificate.
I think what server needs is only private key and the certificate issued by CA
Client needs is CA public key
What I don't understand is the server config too has an option of specifying CA cert file path. Will this ever be used if we are talking about server only security that is the client is not using its own keys ? Is that a mandatory parameter for server config with SSL/TLS ?
I hope I am making sense
While sending data to server: 1. Encrypt the data using server's
public key and transmit. Only server will be able to decrypt it as it
has the private key
While downloading data from server: 1. The server encrypts the data
using its private key. It trasmits the certificate and the data. 2.
User decrypts the certificate using CA's public key. 3. Then decrypts
the data using the public key obtained from decrypted certificate.
That's completely wrong (remember, you don't actually encrypt with a private key). That's not at all how SSL works. A new symmetric key is generated for the encryption, per session. See this question for more details.
The CA certificate is used as a trust anchor to verify the identity of the remote party.
Strictly speaking the server itself doesn't need to have a copy of the CA used to sign its own certificate, since it's up to the remote party (the client) to have it as a trust anchor (i.e. something it already trusts).
Nevertheless, it can send the CA certificate as part of its server chain. It is in fact recommended to do so if the CA certificate is itself signed using another CA certificate up the chain. Doing so increases the chances of being authenticated by the remote party.
Servers generally can have options to specify the CA cert for two reasons (depending on what the option is):
Presenting their own certificate chain (i.e. with intermediate CA certificates).
Verifying the client certificate when used (in this case the server has its own trust anchors that lets it verify the remote certificate, which is the client certificate).
Can anybody clarify to me what kind of "Private Key" is located in a PFX file ?
I am asking this because after having read quite a lot on digital certification I am still a bit confuse on the Private Key part, especially when Certificate Authority is involved.
The part I am missing is where it is said that the Private Key is kept secret on the CA side but if I go in my Personal Certificate Store and I try to export a certificate, I am able to export it with the private key in it.
Is this private key is the same as THE private key that we are not supposed to know ?
Public key cryptography is used for two different purposes when it comes to X.509 certificates used for SSL/TLS.
Each X.509 certificate contains a public key, which has a matching private key held only by the identity this certificate belongs to (the subject of the certificate).
One of the purposes is to build trust in another certificate. A CA uses its private key to sign other certificates: possibly other (intermediate) CA certificates or End-Entity Certificates (the ones that are actually going to be used for the SSL/TLS connection itself).
For example, the Root CA's private key can be used to sign (i.e. issue) an Intermediate CA's certificate. The Intermediate CA certificate can be verified using the Root CA certificate's public key. The Intermediate CA's own key-pair isn't involved in this verification (although the verification process ensures that the public key in that cert belongs to the Intermediate CA).
The Intermediate CA's private key can be used to sign your server certificate (an EEC). The Intermediate CA's public key can be used to verify the authenticity of your server certificate. Someone who would only have the Root CA certificate can therefore check your server certificate by building a chain. Again, your server public and private keys are not involved in this verification at all, but the result of these operations proves that the public key in your server certificate belongs to your server name.
There is no relationship between any of the key pairs involved in a certification chain. Even the CA issuing a certificate will not (or should not) have access to the private key of the certificate it is issuing.
The second usage of public key cryptography is during the SSL/TLS handshake. The exact mechanism depends on the cipher suite, but this allows the client to agree with the server on a secret that only the server with the private key matching the public key in the server certificate will be able to see. It's this private key that is the private key of the certificate itself.
In a PKCS#12 file, you will find an End-Entity Certificate and the private key matching the public key in that certificate, to be used by the entity to which this certificate was issued, and generally a chain of CA certificates (from the one directly issuing the EEC to other CA certificate further up the chain, possibly up to a Root CA).
Here, I've taken this example of an EEC for an SSL/TLS server, but the same could apply for EEC to be used in other contexts (e.g SSL/TLS client or S/MIME).
If you are talking about SSL-Certificates and SSL I think you mean X509-Certificates in common. Let me make it clear first: the passwords of the private keys are different.
The magic here is Chaining of the Certificates, i.e.
Root Certificate from CA
Intermediate Certificate from CA (for instance Class 2)
Your Certificate
If you look at your SSL-Certificate, you will find this structure. Every certificate within this hierarchy can be seen as a certificate for its own. You can find more information within this msdn article. A briefly description from my point of view:
The Certifcates within this chain can be seen are separate Certificates. With their own ability to sign Information with the private key and validating signature with their private key. They are basically linked within this chain.
You might ask Why are the Certificates linked?
There are two main reasons: Security and Trust.
If you loose your private key or if your certificates is broken, it's need to be revocated. This basically says, that your Certificates should not be trusted anymore. It's listed on the Revocation-List on the parent Certificate to ensure that this security break can be fixed very fast. This behavior explains, why root certficates creating IntermediateCertificates and not your requested Certificate - they want to be sure that they can revocate their Intermediate Certificate as well.
Additionally to this functional explanation there is a mathematical, but I am not able to explain this is an easy way. You can find Information about this within it's article on Wikipedia. Basically it says, that you can't calculate the password from the parent and / or public certificate to use the private certificate.
I'm consuming a web service from an Android device using HTTP(s) with SSL. A self-signed (untrusted) certificate is used for client authentication.
I have a general understanding of how public/private keys are used for SSL. From my understanding I can clearly see how a certificate can be used to set up a secure connection and transmit data securely. However I do not understand how they are used for client authentication since the certificate contains the public key and is not kept a secret.
I have a few questions:
Where can I read about how SSL and certificates can be used for client authentication?
Even if the certificate was not made public...by visiting the HTTPS URL in a browser I can view and save the certificate. I can then package up the certificate in a key store and use it from an application.
In this post Jeremy Huiskamp writes
client auth will automatically be performed when the server requests
it
...so client authentication as well as encryption of data can be performed with certificates?
Edited to answer the first part of my question: The client keystore should contain not only the server's public key but also the client's private key. The server must then be able to decrypt using the client's public key? Does this mean the keystore should have two certificates?
First, a quick point about the terminology in public key cryptography:
you sign and decrypt/decipher using a private key,
you verify (a signature) and encrypt/encipher using a public key.
(You don't really "decrypt" using a public key.)
Using SSL/TLS with or without client-authentication, the server presents a certificate (*) for which it has the private key. The server sends its certificate during the SSL/TLS handshake (at the beginning of the connection) and is able to decipher what the client sends using its private key (which it keeps private). The private key and certificates are stored in the server's keystore (or equivalent if it's not implemented in Java).
As part of this, the client uses its truststore, which is a form a keystore that contains trusted certificates, to verify the server certificate. The server certificate could be trusted by being explicitly in the truststore or, in most cases, trusted by linking in to a trusted CA certificate in the truststore (PKI).
The terminology between keystore and truststore in Java can be a bit confusing, you can find more details in this answer.
Regarding your question, the client's truststore doesn't contain the server's public key, but either its certificate or a CA certificate with which it should be verifiable. (It's not just about having the public key, but knowing whose it is, using the other pieces of information in the certificate.)
When you use client-certificate authentication in addition to this, there is a truststore (or equivalent) on the server side and a keystore on the client side too, since the roles are reversed for this purpose.
In the SSL/TLS handshake that uses client-authentication, the server requests a certificate from the client, who sends it (if available).
At the end of this handshake, the client sends a CertificateVerify message, which signs all the messages exchanged so far between the client and the server (so it's something known to both) using the client certificate private key. The server is then able to verify this signature against the public key within the client certificate it has obtained as part of this exchange. This proves to the server that whoever is on the client side has the private key corresponding to the public key in the certificate it has sent.
The next step for the server is to verify whether to trust this certificate, i.e. whether to trust the binding between identity and public key as presented and "sealed" within the certificate.
This is usually done using a PKI, whereby you check the certificate against a known CA, or if your deployment environment is sufficiently small, against a fixed set of trusted certificates. (There can be alternative methods of verification, but their usability will really depend on the circumstances where you want to deploy this system.)
Therefore, for your second question:
The client keystore should contain at least the client's certificate and its private key.
The client truststore should contain the server certificate or a CA certificate with which the server certificate can be verified.
Since both keystore and truststore are a keystore (in the storage format sense, usually a file) used for a different purpose, it's often possible to use the same keystore to serve both the purpose of keystore and truststore.
(*) There are cipher suites that don't rely on certificates, but that's unusual and off topic for this question.
A certificate just binds an identity to a public key. This binding is not secret, so there is no need to keep the certificate secret. If I have John Smith's certificate, I can prove that John Smith owns the secret key corresponding to a particular public key. But since I don't know that secret key, the certificate is of no use to me.
When authentication occurs by certificate, one step is always to have whoever presents the certificate prove they know the secret key corresponding to the public key in the certificate. If you cannot pass that step, the authentication fails.
The server's keystore should have the server's certificate. The client's keystore should have the client's certificate. The client will present its certificate to the server, so the server will learn the client's public key that way. (And vice-versa.)