You can use this guide to install Active Directory Certificate Services (AD CS) as an Enterprise root certification authority (CA) and to enroll server certificates to servers that are running Network Policy Server (NPS), Routing and Remote Access service (RRAS), or both NPS and RRAS.
In the illustration above, multiple servers are depicted: DC1, CA1, WEB1, and many SDN servers. This guide provides instructions for deploying and configuring CA1 and WEB1, and for configuring DC1, which this guide assumes you have already installed on your network. If you have not already installed your Active Directory domain, you can do so by using the Core Network Guide for Windows Server 2016.
In this scenario, the Enterprise Root certification authority (CA) is also an issuing CA. The CA issues certificates to server computers that have the correct security permissions to enroll a certificate. Active Directory Certificate Services (AD CS) is installed on CA1.
When you deploy server certificates, you make one copy of the RAS and IAS servers certificate template and then configure the template according to your requirements and the instructions in this guide.
You utilize a copy of the template rather than the original template so that the configuration of the original template is preserved for possible future use. You configure the copy of the RAS and IAS servers template so that the CA can create server certificates that it issues to the groups in Active Directory Users and Computers that you specify.
The CA publishes a certificate revocation list (CRL) that computers must check to ensure that certificates that are presented to them as proof of identity are valid certificates and have not been revoked. You must configure your CA with the correct location of the CRL so that computers know where to look for the CRL during the authentication process.
After you configure the certificate template on the CA, you can configure the default domain policy in Group Policy so that certificates are autoenrolled to NPS and RAS servers. Group Policy is configured in AD DS on the server DC1.
You must create an alias (CNAME) resource record for the Web server to ensure that other computers can find the server, as well as the AIA and the CRL that are stored on the server. In addition, using an alias CNAME resource record provides flexibility so that you can use the Web server for other purposes, such as hosting Web and FTP sites.
On the CA, configure a copy of the RAS and IAS Servers certificate template. The CA issues certificates based on a certificate template, so you must configure the template for the server certificate before the CA can issue a certificate.
Configure server certificate autoenrollment in Group Policy. When you configure autoenrollment, all servers that you have specified with Active Directory group memberships automatically receive a server certificate when Group Policy on each server is refreshed. If you add more servers later, they will automatically receive a server certificate, too.
Refresh Group Policy on servers. When Group Policy is refreshed, the servers receive the server certificate, which is based on the template that you configured in the previous step. This certificate is used by the server to prove its identity to client computers and other servers during the authentication process.
All domain member computers automatically receive the Enterprise Root CA's certificate without the configuration of autoenrollment. This certificate is different than the server certificate that you configure and distribute by using autoenrollment. The CA's certificate is automatically installed in the Trusted Root Certification Authorities certificate store for all domain member computers so that they will trust certificates that are issued by this CA.
Server certificates are an extremely recurrent piece of software that the average network user has no knowledge of but encounters every time the user accesses the internet. To any person within the technology industry, they are known as a method of confirming the identity of a server to a client.
The two most common types of server certificates are a web server certificate and a RADIUS server certificate. Both server certificate types will be explored in more detail, but the common thread between them is how they identify the involved parties with a certificate chain of trust.
To establish trust, server certificates should always be issued from a Certificate Authority (CA) that can be recognized by every client device. The reason is that operating systems and web browsers have certificate stores. These come with a set of trusted certificate authorities by default, which is known as public certificate authorities. If your server certificate is signed by a public CA, then the client devices would not see any SSL errors if the server certificate is correctly used.
You can use server certificates signed by private certificate authorities, but you need to go through additional steps as an administrator to make every single device trust the issuing private CA to prevent users from seeing SSL errors or warnings. Some operating systems are not friendly as well. For example, iOS would show a warning if you try installing a private certificate (a certificate that is not recognized by default by iOS).
A web server certificate binds to a particular web server and activates the HTTPS protocol to protect communications between the client and the webserver. For those unaware, a website URL is preceded by a transfer protocol, either HTTP or HTTPS. HTTPS means that the communications with this web server are secure, and it is a fast-growing trend to include a server certificate among popular websites.
To use a server certificate on a web server, the Subject Alternative Name (SAN)/Domain Name System (DNS) should match the Fully Qualified Domain Name (FQDN) of the webserver. Or at least it should be a wildcard entry, which is a public key certificate that can be used with several subdomains of a website domain.
Next, the server presents the web server certificate to prove its identity. The web server certificate is identified based on its SAN and the client deciphers whether the certificate has been issued from a trusted CA.
If the certificate is trusted, the client moves on to the final key exchange. The encrypted message is exchanged based on the symmetric algorithm established in the hello message. Both the client and server agree to use a key to decrypt the encrypted messages and communicate securely while the webserver is in use. This process occurs faster than a human can detect, but ensures that all communications will only be read by the server and client.
An interesting use case involves the process of equipping Windows servers with web server certificates. In order to provision a Windows web server with a certificate, the server account Web Hosting must have a server certificate with Common Name (CN) that matches the URL hosted by the server. This is often used with Internet Information Service (IIS) and presents a unique method of establishing trust between client and server.
A RADIUS server certificate is used to prove that the RADIUS server a client is authenticating to is in fact the correct server. Based on the CN on the certificate, the end-user can feel secure knowing that they will not fall victim to a Man-In-The-Middle attack.
The process of confirming the identity of a RADIUS is called server certificate validation. When a user connects an EAP secured network, a TLS handshake will be initiated (this is similar in concept to the webserver certificate handshake).
When using an EAP authentication method, an encrypted EAP tunnel is created to prevent anyone but the client and server from viewing the messages being exchanged. Within the EAP tunnel, a certificate chain of trust is between the server certificate, an intermediate certificate, and the client. Trust is established by validating the server with the help of Root CA and the CN of the server certificate. Then both parties can communicate securely. While it is a best practice to include the intermediate certificate in the chain, the process can be completed with just the root certificate. Because clients use Root CA and name to validate the server. Whereas servers show the missing piece, Intermediate CA, to complete the certificate chain and prove their identity in the process.
Server certificates are just one of many possible use cases for the certificates. Beyond server identification, they can be used for VPN authentication, email security, Smart Card security, managed device authentication, and much more.
Those problems are the inability to login remotely via Remote Desktop due to a "RPC server not available" error, and the lost ability to create new certificates. We tried stopping and restarting some of the services, but several of them remain stuck in the "stopping" phase. The server uptime is something close to a year and a half, and the assumption is that a restart ought to bring everything back up fresh.
This is simply not true. All certificates issued by the server are signed with it's private key. That signature is what's checked by clients using the public key certificate that has also been generated by the server and installed on the clients. Nothing needs to be verified on the server in order for connections to succeed. The only thing that would ever be checked is revocations. So go ahead and reboot it. In fact, go ahead and start installing security updates on it as well. Since you haven't rebooted it in a long time, I assume it hasn't been updated in a very long time. That's kind of worrisome...
Client Certificates are digital certificates for users and individuals to prove their identity to a server. Client certificates tend to be used within private organizations to authenticate requests to remote servers. Whereas server certificates are more commonly known as TLS/SSL certificates and are used to protect servers and web domains. Server Certificates perform a very similar role to Client Certificates, except the latter is used to identify the client/individual and the former authenticates the owner of the site.