Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Logjam downgrades a TLS connection to 512-bit export-grade Diffie-Hellman, which an attacker can break in real time thanks to precomputation against a handful of shared prime numbers.

What Is the Logjam Attack?
Logjam, published in May 2015 by a team of researchers from INRIA, Microsoft Research, Johns Hopkins, and the Universities of Michigan and Pennsylvania, is a downgrade attack against the Diffie-Hellman key exchange in TLS. An active attacker on the network convinces a server to negotiate deliberately weakened 512-bit export-grade parameters, then breaks the resulting key and reads or modifies the session. It is tracked as CVE-2015-4000, and its real contribution was showing that the cost of attacking Diffie-Hellman had been badly misjudged.
Until the late 1990s, United States export regulations limited cryptography in software sold abroad to 512 bits. Vendors complied by building two modes into their products: full strength for domestic use and “export grade” for everything else. The rules were relaxed in 2000, but the weak cipher suites stayed in the code, dormant, for the next fifteen years.
Logjam woke them up, as did FREAK for RSA a few months earlier. Both are reminders that a protocol option retained for compatibility is an option an attacker can select on your behalf.
In a Diffie-Hellman key exchange the server chooses the group parameters—a large prime and a generator—and sends them to the client during the handshake. Critically, in TLS 1.2 and earlier the parameters themselves are not bound to the negotiated cipher suite in a way the client can verify before the handshake completes.
The attacker sits in the path and edits the client’s list of offered cipher suites so that only DHE_EXPORT remains. The server, if it still supports export mode, replies with a 512-bit prime. The client has no way to tell that it did not ask for this, because the server’s signature covers the parameters but not the suite the client originally offered. Both sides complete the handshake, and the attacker only needs to solve one 512-bit discrete logarithm to recover the session key and finish the handshake in both directions.
Solving a discrete logarithm with the number field sieve has two stages. The first and by far the most expensive depends only on the prime, not on the particular key exchange. Once that work is done, individual logarithms for that prime can be computed in about a minute.
The researchers found that servers overwhelmingly reused a tiny number of hard-coded primes. A single 512-bit prime shipped with Apache covered a large share of vulnerable hosts; another came from mod_ssl. Precomputing against two primes therefore put real-time attacks within reach of a well-resourced individual, and the authors estimated that an academic team could reach 768-bit, while a nation state could plausibly have completed the precomputation for a 1024-bit prime—which, they pointed out, would explain passively decrypted VPN traffic described in leaked intelligence documents.
At publication, 8.4% of the top one million HTTPS sites were vulnerable, along with substantial proportions of SMTP StartTLS, IMAPS, POP3S, and IKE endpoints used for IPsec VPNs.
The remediation is configuration rather than patching, and it has not changed:
EXP and LOW.Browsers responded by refusing Diffie-Hellman groups below 1,024 bits, which broke a visible number of badly configured sites and forced the issue. Modern scanners, including the Qualys SSL Labs test and testssl.sh, flag weak groups automatically, so this is easy to verify as part of routine vulnerability scanning.
Logjam belongs to a family that includes POODLE, DROWN, and the ROBOT attack, all of which exploit obsolete options that were never removed. The practical lesson for anyone managing TLS is that a cipher suite list is a security control with a maintenance schedule, not a one-off setting, and that shared cryptographic parameters concentrate risk in a way individual key sizes do not suggest.
Our Network Security Services include protocol and cipher configuration review across web, mail, and VPN endpoints, which is where weak groups usually survive longest.
Against well-configured modern servers, no. Export cipher suites have been removed from every current TLS library, browsers reject groups under 1,024 bits, and TLS 1.3 eliminates custom Diffie-Hellman parameters altogether. The residual risk is in legacy VPN concentrators, mail servers, and embedded management interfaces that were configured years ago and never reviewed, which is exactly where scanners still find weak groups.
Both downgrade a connection to export-grade cryptography, but they target different key exchanges. FREAK forces the use of 512-bit export RSA and exploits a client-side state machine bug, whereas Logjam targets Diffie-Hellman and exploits the fact that the server chooses the group and the client cannot verify what was originally offered. Logjam also benefits from precomputation against widely shared primes.
Elliptic curve, in almost all cases. ECDHE gives equivalent security to finite-field Diffie-Hellman at a fraction of the computational cost, uses standard named curves rather than server-chosen parameters, and avoids the precomputation weakness that makes shared primes dangerous. If finite-field groups are required for compatibility, use a unique group of at least 2,048 bits generated for your own environment.
Run an external scan with the Qualys SSL Labs test for public web endpoints, and testssl.sh or sslyze for internal and non-HTTP services. Both report the offered cipher suites and the size of any Diffie-Hellman group. Remember to check mail servers, VPN gateways, and management interfaces as well as websites, since those are routinely excluded from web-focused assessments.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.