Description

When IMAP compression is enabled, the same compression state is reused across responses in a session, so response sizes depend on both attacker-supplied mail and other mail in the same mailbox. An attacker that can send mail to a user and can also observe the sizes of that user's IMAP traffic can confirm whether the body of a small message matches a guessed text. Recovery of arbitrary unknown content was not demonstrated, but the attack can disclose whether a secret-like message body matches a candidate. Disable IMAP compression. Update to non-vulnerable version. No publicly available exploits are known.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "3.7",
  "pubDate": "2026-08-28T12:16:28.897Z",
  "pubdate": "2026-08-28T12:16:28.897Z",
  "executiveSummary": "A side-channel vulnerability exists within the IMAP protocol implementation when compression is enabled. The vulnerability stems from the reuse of the compression state across multiple responses within a single IMAP session, which correlates response sizes with the content of the transmitted data.\nThis flaw allows an attacker capable of sending mail to a target user and monitoring the resulting network traffic size to infer sensitive information. By observing variations in encrypted packet sizes, an attacker can confirm guesses regarding the content of private messages through a side-channel analysis of the compression ratio.\nThe attack is essentially a chosen-plaintext side-channel attack targeting the protocol-level compression. While the current disclosure indicates that full recovery of arbitrary unknown content has not been demonstrated, it poses a significant risk to the confidentiality of mail contents, specifically for confirming guessed text or secret strings.\nExploitation requires the attacker to have the ability to send emails to the victim's account and perform traffic analysis on the victim’s IMAP session. Impact is limited to the disclosure of sensitive information via pattern matching rather than direct system compromise or remote code execution. Immediate mitigation involves disabling IMAP compression.",
  "technicalDetails": "The vulnerability is a cryptographic side-channel attack targeting the interaction between the IMAP protocol and its compression mechanism. When compression is negotiated for an IMAP session, the system maintains a persistent compression context across the session lifecycle. Because this context is reused, the compression algorithm (e.g., DEFLATE) generates output sizes that are highly dependent on the redundancy between the attacker-supplied input and the existing content in the mailbox.\nThe root cause lies in the stateful nature of the compression engine. As the IMAP server processes requests and responses, the compression window updates based on previously transmitted data. An attacker can leverage this by sending a crafted email message to a victim's mailbox. When the victim (or their IMAP client) accesses this mailbox, the server compresses the response. If the attacker can measure the size of the resulting encrypted network packets, they can deduce whether the victim's private email content contains strings that align with the attacker’s input, thereby achieving higher compression ratios (smaller packet sizes).\nThe attack flow follows these steps: 1) The attacker sends a series of emails to the victim containing various candidate guesses for the secret content. 2) The attacker initiates an IMAP session or observes an existing session where compression is enabled. 3) The attacker triggers the synchronization of the mailbox, forcing the server to compress the responses that include the targeted private messages alongside the attacker's previously injected content. 4) The attacker monitors the size of the encrypted traffic flows. 5) By comparing the size of responses containing different guesses, the attacker determines a successful match when a specific guess results in a smaller-than-average response size due to increased dictionary matches within the compression algorithm.\nThis vulnerability is distinct from traditional buffer overflows or injection flaws, as it focuses on information leakage via traffic metadata. It requires no specific privileges other than the ability to deliver mail to the user and perform passive or active network traffic monitoring to observe session frame lengths. The lack of state isolation between mail items within the same compression stream is the primary design flaw facilitating this leakage. The scope of the risk includes the potential to confirm sensitive information such as password resets, tokens, or private communications contained within user mailboxes."
}