I am expecting that only on the server will the certificate chain be checked. Is that true? Are there exceptions? If the client side does not check the certificate chain then there would be no need for a trust store on the client side, right?
Related
I know that certificates that are sent by the server cant be faked (still there is MD5 collisions but costy) but what about faking the client ..
in man in the middle attack:
cant we tell the server that we are the legitimate client and take data from that server manipulate it then encrypt it again with legitimate client public key ? how does the client be sure that the data came really from the server ?
in theory .. can we inject any data into the response sent by the server to the client ?..
How are you authenticating the client? SSL client certificates? Or some application level system (cookies etc)?
Here's what SSL does in a nutshell:
Negotiates a Diffie-Hellman shared session key between the two parties
Has the server sign the session key and send the result to the client. Once the client verifies this, the client knows there is no MITM, and the server is who they say they are.
If client certificates are enabled, has the client sign the session key and send the signature to the server. The server now knows there is no MITM and the client is who they say they are.
Encrypts all data in both directions using the shared session key
Typically when you use SSL you won't use client certificates. Strictly speaking, the server does not know if the connection is MITM'd. However, most clients will disconnect if the server certificate is bad. The server assumes that if the client pushes forward with the connection, there is no MITM. Even if Mallory, doing the MITM, chooses not to propagate the disconnect from the client, he has no new information now; all he's done is connected to the server himself. Without intercepting the client's session cookie or other authentication information (which is only sent by the client after verifying the connection is secure) the MITM is useless.
So in short, as long as one end or the other verifies the certificate of the other end before initiating any high-level communication of sensitive information, SSL is secure in both directions.
You're right -- without secure certificate authentication on the client and server there is an opening for a man in the middle attack.
SSL can be "secure both ways" if you use mutual authentication also called two-way SSL.
I am trying to setup a Mosca server with SSL/TLS encryption.
Looking at the Mosca wiki at the page https://github.com/mcollina/mosca/wiki/TLS-SSL-Configuration suggests that we will require a private key and a certificate for the broker.
While that page is silent about the the configuration on the client side for the mqtt over SSL/TLS, I found an article by Mattino Collina himself on SSL/TLS configuration on the client side. Here http://www.hivemq.com/blog/mqtt-client-library-mqtt-js
This article states that for mqtts we need to provide a key and a certificate on the client side too. Should they be the same key and certificate that we provided while setting up the broker or different for every client that we connect to the broker?
Are these keys and certificates a way of authenticating the client for the broker?
How many certificates and keys depends on exactly what you are trying to achieve.
If you just need a secure connection then you only need a certificate and private key on the broker. (you may need to supply the CA certificate to the client depending on if you sign your own certificates or if you use certificates signed by recognised public CA).
If you want to uniquely identify the client via SSL then the client will also require it's own unique certificate and matching private key. These will be different from the brokers certificate/key pair, but probably signed by the same CA.
I have an asp.net mvc web app that has been running in production for about 4 years. Suddenly since about a week ago, I am getting this error being returned for all calls to 3rd-party secure API's:
System.Net.WebException: The underlying connection was closed: Could
not establish trust relationship for the SSL/TLS secure channel. --->
System.Security.Authentication.AuthenticationException: The remote
certificate is invalid according to the validation procedure.
This is for calls to SendGrid for sending emails, calls to Azure Blob Storage for uploading of documents, calls to Connect.io for logging.
I have managed to resolve the Azure Blob Storage problem temporarily by changing the connection string to use http instead of https.
Clearly something has broken on my app server, and I have no idea where to start looking.
Please help.
Edit:
Turns out I was using a sample library provided by one of my (lesser-used) 3rd party API's, and this library had an override of
System.Net.ServicePointManager.ServerCertificateValidationCallback += (sender, certificate, chain, sslPolicyErrors)
which had it's own logic about what constitutes a valid certificate!!! AARGH!
This part become key information for your problem:
I am getting this error being returned for all calls to 3rd-party
secure API's
According to MSDN blog:
This error message is caused because the process is not being able to
validate the Server Certificate supplied by the Server during an HTTPS
(SSL) request. The very first troubleshooting step should be to see
if the server supplied certificate and every certificate in the chain
is trouble free.
Because it seems that one or more third party certificates are rejected, you may configure Trusted Roots part of your certificate trust lists to include all required third party CA as part of chain to work with secure APIs from trusted sources, including reissued certificates if any.
Further details: https://technet.microsoft.com/en-us/library/dn265983.aspx
NB (Optional):
As temporary measure, you can implement this certificate validation handler in WebRole.cs until all related third-party certificates has reissued (remember this setting will trust all issued certificates, hence it's not recommended for long term usage):
System.Net.ServicePointManager.ServerCertificateValidationCallback += (sender, certificate, chain, sslPolicyErrors) => true;
Additional reference: http://robertgreiner.com/2013/03/could-not-establish-trust-relationship-for-the-ssl-tls-secure-channel/
Similar thing happened in our system. Our problem was TLS version. The SSL offload appliance was configured to accept only TLS 1.2. One week ago this configuration accepted all TLS versions 1.0 to 1.2.
We had to reconfigure .NET's SecurityProtocol settings like:
System.Net.ServicePointManager.SecurityProtocol = SecurityProtocolType.Ssl3 | SecurityProtocolType.Tls12;
You can use this site to test which TLS version you are using: https://www.ssllabs.com/ssltest/index.html
Try to get some information about the certificate of the servers and see if you need to install any specific certs.
The server(s) may had a cert signed by a 3rd party CA which you hadn't trusted yet. The solution is to add that CA to the Trusted Root CA list.
I understand the use and necessity of SSL Certificates. I am looking for a detailed explanation of the relationship between http protocol and SSL Certificate and also between https and SSL Certificate.
In AFNetworking defining _AFNETWORKING_ALLOW_INVALID_SSL_CERTIFICATES_ allows invalid certificates.
Assuming this flag is not set, the library should deny invalid certificates by default. But what if the server has no certificate installed at its end? I see that I can access a server with no certificate using http protocol.
And does the default certificate validation occur for https protocol only?
I want to access only those servers with a certificate and which are valid over http and https? Is this feature required and if so is it possible?
I think you don't understand what HTTP and HTTPS are about:
http is the session protocol to communicate with servers without encryption. Everything is sent in the clear and there is also no way to verify the authenticity of the server (i.e. that the server responding is indeed having the correct hostname). A server doesn't need an SSL certificate in this case.
https is used when you want to ensure that communication with the server is encrypted. For this, the server needs an SSL certificate and before anything is sent over the HTTP session, a handshake will take place allowing the client to fetch the SSL certificate of the server in order to encrypt the data that is sent. So an SSL certificate is only used for HTTPS connections.
It is also possible to verify the authenticity of the server, depending on the SSL certificate used. All browsers (and by default on iOS, unless you explicitely change this in your app) will only accept SSL certificates that have been issued by known certification authorities. This way you can be sure that the server is indeed located at the hostname you requested. Depending on the SSL certificate the owner of the server might also be verified (i.e. that the server indeed belongs to the company it claims to belong to).
The Application i am working on needs to connect to a webservice over https, The certificate is trusted and valid.
I have used NSURLConnection is previous projects to use soap over http
Can anybody please point the difference between the two above mentioned scenarios,
I also need to understand what exactly happens when connecting over https, is the certificate stored automatically on the device, how does ssl handshake happen.
Any Pointers in this direction will be really helpful.
Regards,
Ishan
I need some clarification. Is the certificate signed by Apple for use with notifications or is it signed by an SSL root certificate authority (like VeriSign)?
Apple signed certificates are only to be used with WebServer to Apple Server communications like the Apple Push Notification Service. They are not intended for iOS device to WebServer.
A SSL certificate signed by a SSL root certificate authority should just work.
I think you are looking for an HTTP over SSL/TLS primer. So, here it goes.
HTTP is an unencrypted channel. The request and response are in a plain text data stream. HTTPS is an encrypted channel. The request and response are in a data stream encrypted using a shared master key. The magic of SSL/TLS is how this encrypted channel is created.
First, the client and server say hello to each other (in a clear channel).
Next, the client downloads the server's public certificate (in a clear channel).
At this point, the client has some work to do. It needs to verify the certificate. It needs to know that it understands the certificate, that the date range is valid, that the certificate is signed by a trusted certificate authority, and that the certificate has not been revoked.
Now, the client knows that it can trust the server.
Next, It sends a few short messages encrypted with the public key of the server (which is in the server's public certificate). These messages can only be decrypted by the server's private key (which only the server knows about). These messages allow the client and the server to negotiate a master key.
Finally, the client and the server begin the normal HTTP request and response using the newly created encrypted channel.
I hope this is what you are looking for. For a more detailed description see: http://www.moserware.com/2009/06/first-few-milliseconds-of-https.html
If the certificate was issued by a chain of certificate authorities whose root is trusted by Apple, then there is nothing to do. The iOS device will accept the certificate, as long as it is otherwise valid (ie not expired, not revoked, etc).
If the CA chain's root is not trusted by Apple, you will need to download the root's certificate to the phone. This can be done (I think) via the iPhone Configuration Utility. Enterprise provisioning scenarios undoubtedly support this also.