Zen

Translation

Since 2014, latency has become an important concern for web services. As access networks bandwidth continue to grow, latency is becoming a key factor that affects the performance of interactive web services. With TLS 1.2, the download of a web page requires a minimum of four round-trip-times, one to create the underlying TCP connection, one to exchange the ClientHello/ServerHello, one to exchange the keys and then one to send the HTTP GET and retrieve the response. This can be very long when the server is not near the client. TLS 1.3 aimed at reducing this handshake to one round-trip-time and even zero by placing some of the cryptographic handshake in the TCP handshake. This part will be discussed in the TCP chapter. We focus here on the reducing the TLS handshake to a single round-trip-time.
0/8030 · 803

Sign in to save translations.

English French Actions
a `TLSPlainText` that contains the encrypted data
 
TLS supports several methods to encrypted records. The selected method depends on the cryptographic algorithms that have been negotiated for the TLS session. A detailed presentation of the different methods that can be used to produce the `TLSPlainText` from the user data is outside the scope of this book. As an example, we study one method: Stream Encryption. This method is used with cryptographic algorithms which can operate on a stream of bytes. The method starts with a sequence of bytes provided by the user application: the plain text. The first step is to compute the authentication code to verify the integrity of the data. For this, TLS computes :math:`MAC(SeqNum, Header, PlainText)` using HMAC where `SeqNum` is a sequence number which is incremented by one for each new TLS record transmitted. The `Header` is the header of the TLS record described above and `PlainText` is the information that needs to be encrypted. Note that the sequence number is maintained at the two endpoints of the TLS session, but it is not transmitted inside the TLS record. This sequence number is used to prevent replay attacks.
 
MAC-then-encrypt or Encrypt-then-MAC
 
When secure protocols use Message Authentication and Encryption, they need to specify how these two algorithms are combined. A first solution, which is used by the current version of TLS, is to compute the authentication code and then encrypt both the data and the authentication code. A drawback of this approach is that the receiver of an encrypted TLS record must first attempt to decrypt data that has potentially been modified by an attacker before being able to verify the authenticity of the record. A better approach is for the sender to first encrypt the data and then compute the authentication code over the encrypted data. This is the encrypt-then-MAC approach proposed in :rfc:`7366`. With encrypt-then-MAC, the receiver first checks the authentication code before attempting to decrypt the record.
 
Improving TLS
 
During the last two decades, the deployment of TLS has continued to grow. The early TLS servers were only used for critical services such as e-commerce websites or online banks. As CPU performance improved, it became much more cost-effective to use TLS to secure non-critical parts of web servers, including the delivery of HTML pages and even video services. There is now a growing number of applications that rely on TLS [AM2019]_.
 
In 2013, the statistics collected by the Firefox Telemetry project [#ftelemetry]_ revealed that 30% of the web pages loaded by Firefox users were done over HTTPS. In October 2019, 80% of the web pages are loaded over HTTPS. In six years, HTTPS became the dominant protocol to access web services. Another look at the deployment of HTTPS on web sites may be found in [Helme2019]_.
 
Measurement studies that analyzed the evolution of TLS over the years have identified several important changes in the TLS ecosystem [KRA2018]_. First, the preferred cryptographic algorithms have changed. While RC4 was used by 60% of the connections in 2012, its usage has dropped since 2015. AEA started to be deployed in 2013 and is now used for more than 90% of the connections. The deployed versions of TLS have also changed. TLS 1.0 and TLS 1.1 are now rarely used. The deployment of TLS 1.2 started in 2013 and reached 70% of the connections in 2015. Version 1.3 of TLS, that is described below, is also widely deployed.
 
Another interesting fact is are the key exchange schemes. In 2012, RSA was the dominant solution, used by more than 80% of the observed connections [KRA2018]_. In 2013, Edward Snowden revealed the surveillance activities of several governments. These revelations had a huge impact on the Internet community. The IETF, which standardizes Internet protocols, considered in :rfc:`7258` that such pervasive monitoring was an attack. Since then, several IETF working groups have developed solutions to counter pervasive monitoring. One of these solutions is to encourage `Perfect Forward Security`. Within TLS, this implies replacing RSA by an authenticated Diffie Hellman key exchange such as ECDHE. Measurements indicate that since summer 2014, ECDHE is more popular than RSA. In 2018, more than 90% of the observed TLS connections used ECDHE.
 
The last point is the difficulty of deploying TLS servers [KMS2017]_. When TLS servers are installed, the system administrator needs to obtain certificates and configure a range of servers. Initially, getting certificates was complex and costly, but initiatives such as https://letsencrypt.org have simplified this workflow.
 
In 2014, the IETF TLS working started to work on the development of version 1.3 of the TLS protocol. Their main objectives [Rescorla2015]_ for this new version were:
 
simplify the design by removing unused or unsafe protocol features
 
improve the security of TLS by leveraging the lessons learned from TLS 1.2 and some documented attacks
 
improve the privacy of the protocol
 
reduce the latency of TLS
 
Since 2014, latency has become an important concern for web services. As access networks bandwidth continue to grow, latency is becoming a key factor that affects the performance of interactive web services. With TLS 1.2, the download of a web page requires a minimum of four round-trip-times, one to create the underlying TCP connection, one to exchange the ClientHello/ServerHello, one to exchange the keys and then one to send the HTTP GET and retrieve the response. This can be very long when the server is not near the client. TLS 1.3 aimed at reducing this handshake to one round-trip-time and even zero by placing some of the cryptographic handshake in the TCP handshake. This part will be discussed in the TCP chapter. We focus here on the reducing the TLS handshake to a single round-trip-time.
 
To simplify both the design and the implementations, TLS 1.3 uses only a small number of cipher suites. Five of them are specified in :rfc:`8446` and ``TLS_AES_128_GCM_SHA256`` must be supported by all implementations. To ensure privacy, all cipher suites that did not provide Perfect Forward Secrecy have been removed. Compression has also been removed from TLS since several attacks on TLS 1.2 exploited its compression capability :rfc:`7457`.
 
Enterprises, privacy and TLS
 
By supporting only cipher suites that provide Perfect Forward Secrecy in TLS 1.3, the IETF aims at protecting the privacy of users against a wide range of attacks. However, this choice has resulted in intense debates in some enterprises. Some enterprises, notably in financial organizations, have deployed TLS, but wish to be able to decrypt TLS traffic for various security-related activities. These enterprises tried to lobby within the IETF to maintain RSA-based cipher suites that do not provide Perfect Forward Secrecy. Their arguments did not convince the IETF. Eventually, these enterprises moved to ETSI, another standardization body, and convinced them to adopt `entreprise TLS`, a variant of TLS 1.3 that does not provide Perfect Forward Secrecy [eTLS2018]_.
 
The TLS 1.3 handshake differs from the TLS 1.2 handshake in several ways. First, the TLS 1.3 handshake requires a single round-trip-time when the client connects for the first time to a server. To achieve this, the TLS designers look at the TLS 1.2 handshake in details and found that the first round-trip-time is mainly used to select the set of cryptographic algorithms and the cryptographic exchange scheme that will be used over the TLS session. TLS 1.3 drastically simplifies this negotiation by requiring to use the Diffie Hellman exchange with a small set of possible parameters. This means that the client can guess the parameters used by the server (i.e. the modulus, p and the base g) and immediately start the Diffie Hellman exchange. A simplified version of the TLS 1.3 handshake is shown in the figure below.
 
There are several important differences with the TLS 1.2 handshake. First, the Diffie Hellman key exchange is required in TLS 1.3 and this exchange is initiated by the client (before having validated the server identity). To initiate the Diffie Hellman key exchange, the client needs to guess the modulus and the base that can be accepted by the server. Either the client uses standard parameters that most server supports or the client remembers the last modulus/base that it used with this particular server. If the client guessed incorrectly, the server replies with the parameters that it expects and one round-trip-time is lost. When the server sends its `ServerHello`, it already knows the session key. This implies that the server can encrypt all subsequent messages. After one round-trip-time, all data exchanged over the TLS 1.3 session is encrypted and authenticated. In TLS 1.3, the server certificate is encrypted with the session key, as well as the `Finished` message. The server signs the handshake to confirm that it owns the public key of its certificate. If the server wants to send application data, it can already encrypt it and send it to the client. Upon reception of the server Certificate, the client verifies it and checks the signature of the handshake and the `Finished` message. The client confirms the end of the handshake by sending its own `Finished` message. At that time, the client can send encrypted data. This means that the client only had to wait one round-trip-time before sending encrypted data. This is much faster than with TLS 1.2.
 
For some applications, waiting one round-trip-time before being able to send data is too long. TLS 1.3 allows the client to send encrypted data immediately after the `ClientHello`, without having to wait for the `ServerHello` message. At this point in the handshake, the client cannot know the key that will be derived by the Diffie Hellman key exchange. The trick is that the server and the client need to have previously agreed on a `pre-shared-key`. This key could be negotiated out of band, but usually it was exchanged over a previous TLS session between the client and the server. Both the client and the server can store this key in their cache. When the client creates a new TLS session to a server, it checks whether it already knows a pre-shared key for this server. If so, the client announces the identifier of this key in its `ClientHello` message. Thanks to this identifier, the server can recover the key and use it to decrypt the 0-rtt Encrypted record. A simplified version of the 0-rtt TLS 1.3 handshake [#fhandshake]_ is shown in the figure below.
 
On the web, TLS clients use certificates to authenticate servers but the clients are not authenticated. However, there are environments such as enterprise networks where servers may need to authenticate clients as well. A popular deployment is to authenticate remote clients who wish to access the enterprise network through a Virtual Private Network service. Some of these services run above TLS (or more precisely a variant of TLS named DTLS that runs above UDP [MoR2004]_ but is outside the scope of this chapter). In such services, each client is authenticated thanks to a public key and a certificate that is trusted by the servers. To establish a TLS session, such a client needs to prove that it owns the public key associated with the certificate. This is done by the server thanks to the CertificateRequest message. The TLS handshake becomes the following one:
 
The server sends a CertificatRequest message. The client returns its certificate and signs the Handshake with is private key. This confirms to the server that the client owns the public key indicated in its certificate.
 
There are many more differences between TLS 1.2 and TLS 1.3. Additional details may be found in their respective specifications, :rfc:`5246` and :rfc:`8446`.
 
Footnotes
 
See http://www.iana.org/assignments/tls-parameters/tls-parameters.xhtml#tls-parameters-4
 
A detailed explanation of the TLS 1.3 handshake may be found at https://tls13.ulfheim.net/
 
See https://letsencrypt.org/stats/ for a graph and https://docs.telemetry.mozilla.org/datasets/other/ssl/reference.html for additional information on the dataset
 

Loading…

User avatar None

Source string added
Since 2014, latency has become an important concern for web services. As access networks bandwidth continue to grow, latency is becoming a key factor that affects the performance of interactive web services. With TLS 1.2, the download of a web page requires a minimum of four round-trip-times, one to create the underlying TCP connection, one to exchange the ClientHello/ServerHello, one to exchange the keys and then one to send the HTTP GET and retrieve the response. This can be very long when the server is not near the client. TLS 1.3 aimed at reducing this handshake to one round-trip-time and even zero by placing some of the cryptographic handshake in the TCP handshake. This part will be discussed in the TCP chapter. We focus here on the reducing the TLS handshake to a single round-trip-time.
4 years ago
Browse all string changes

Glossary

English French
No related strings found in the glossary.

32230 String information

Source string location
../../protocols/tls.rst:408
String age
4 years ago
Last updated
a month ago
Source string age
4 years ago
Translation file
locale/fr/LC_MESSAGES/protocols/tls.po, string 80