Certificate authority eMudhra has revoked eight additional intermediate CA certificates after an investigation into its recent certificate security incident revealed mishandling of private keys during the certificate generation process.
The new revocations bring the total number of affected intermediate CA certificates to 16, including eight previously compromised certificates whose private keys were so weak that anyone could recover them almost instantly.
In a new incident report Published on Tuesday, eMudhra said both sets of certificates were affected by a custom Python script used to transfer private keys between hardware security modules (HSM) in April 2025.
HSMs are devices designed to securely generate and store cryptographic keys, preventing sensitive private keys from being exposed outside of hardware.
eMudhra claims to have developed the script to move certificate authority keys between different HSM providers in its facilities after delays in obtaining new hardware.
However, the script contained a bug that reduced eight ECDSA private keys to a single byte, making them extremely easy to recover.
The same script also processed eight RSA private keys, which managed sensitive key information in the memory of the computer running the script rather than protecting it in the HSM.
Although the RSA keys were not weakened like the original eight certificates, eMudhra says it cannot demonstrate that it handled them in accordance with its certificate security policies.
“A reimport performed from host process memory does not comply with CPS §6.2.6 and RSA objects are therefore treated as not meeting this requirement,” the incident report states.
For this reason, the company revoked the eight additional RSA Intermediate CA certificates on November 7, including certificates used to issue TLS and code signing certificates.
Some of the affected RSA intermediaries have been actively issuing certificates since May 2026, with the last certificate being issued on November 3.
eMudhra says the affected client certificates have also been replaced or migrated.
The investigation also found that an earlier issue in June 2025 may have exposed the flaw more than a year before it was discovered by a security researcher known as “Wayne.”
At that time, eMudhra detected certificate revocation lists (CRLs) signed with private keys that did not match the corresponding CA certificates.
The company resolved the problem by signing the revocation lists with the corresponding keys, but did not investigate the reason for the problem.
The new report confirms that this was a warning sign of a faulty key generation process that went undetected until October 2026.
The investigation also revealed that eMudhra did not verify the strength of the generated private keys or that the keys used for certification matched those initially created in the HSM.
Major browsers have blocked compromised certificates
As BleepingComputer previously reported, the incident first came to light on October 17, when a security researcher named Wayne discovered that eight eMudhra intermediate CA private keys could be recovered by testing just 256 possible values.
Major browsers and operating systems trusted compromised certificates, potentially allowing attackers to forge certificates for legitimate websites and intercept encrypted communications.
Mozilla, Google, Apple and other software companies subsequently released protections to block compromised certificates.
Microsoft also released out-of-band Windows updates on October 26 to remove trust from affected eMudhra certificate trust chains on systems that were unable to receive automatic certificate trust updates.
Following the incident, eMudhra announced that it would exit the public trust TLS certificate business and transfer its remaining customers to another provider.
The company announced plans to stop issuing certificates under its remaining G1 hierarchy by November 16.
eMudhra has also commissioned an independent review from KPMG, with results expected by December 4, to identify improvements to its certificate management and key generation procedures.
The company says it has not detected any evidence that the compromised keys were used maliciously.
Adaptive Security’s AI agents analyze email intent, context, and behavior to uncover phishing, business email compromise, and invoice fraud.
See what’s happening with a 14-day read-only scan of Microsoft 365 or Google Workspace, without MX modifications.
