Skip to main content
In the software world, it is expected for security vulnerabilities to be immediately announced, thus giving operators an opportunity to take protective measure against attackers. Vulnerabilities typically take two forms:
  1. Vulnerabilities that, if exploited, would harm the software operator. In the case of Parallax examples would be:
    • A bug that would allow remote reading or writing of OS files, or
    • Remote command execution, or
    • Bugs that would leak cryptographic keys
  2. Vulnerabilities that, if exploited, would harm the Parallax mainnet. In the case of the Parallax client, examples would be:
    • Consensus vulnerabilities, which would cause a chain split,
    • Denial-of-service during block processing, whereby a malicious transaction could cause the network to crash.
    • Denial-of-service via p2p networking, whereby portions of the network could be made inaccessible due to crashes or resource consumption.
In most cases so far, vulnerabilities in the Parallax client have been of the second type, where the health of the network is a concern, rather than individual node operators. For such issues, the Parallax reserves the right to silently patch and ship fixes in new releases.

​
Why silent patches

In the case of Parallax, it takes a lot of time (weeks, months) to get node operators to update even to a scheduled hard fork. If we were to highlight that a release contains important consensus or DoS fixes, there is always a risk of someone trying to beat node operators to the punch, and exploit the vulnerability. Delaying a potential attack sufficiently to make the majority of node operators immune may be worth the temporary loss of transparency. The primary goal for the Parallax team is the health of the Parallax network as a whole, and the decision whether or not to publish details about a serious vulnerability boils down to minimising the risk and/or impact of discovery and exploitation. At certain times, it’s better to remain silent. This practice is also followed by other projects such as Bitcoin.

​
Public transparency

Our policy on public transparency is:
  • If we silently fix a vulnerability and include the fix in release X, then,
  • After 4-8 weeks, we will disclose that X contained a security-fix.
  • After an additional 4-8 weeks, we will publish the details about the vulnerability.
We hope that this provides sufficient balance between transparency versus the need for secrecy, and aids node operators and downstream projects in keeping up to date with what versions to run on their infrastructure. In keeping with this policy, we have taken inspiration from Solidity bug disclosure - see below.

​
Disclosed vulnerabilities

The Parallax client has a built-in command to check whether it is affected by any publically disclosed vulnerability, using the command prlx version-check. This command will fetch the latest json file (and the accompanying signature-file, and cross-check the data against its own version number. The JSON file of known vulnerabilities below is a list of objects, one for each vulnerability, with the following keys:
  • name
    • Unique name given to the vulnerability.
  • uid
    • Unique identifier of the vulnerability. Format Prlx-<year>-<sequential id>
  • summary
    • Short description of the vulnerability.
  • description
    • Detailed description of the vulnerability.
  • links
    • List of relevant URLs with more detailed information (optional).
  • introduced
    • The first published Parallax client version that contained the vulnerability (optional).
  • fixed
    • The first published Parallax client version that did not contain the vulnerability anymore.
  • published
    • The date at which the vulnerability became known publicly (optional).
  • severity
    • Severity of the vulnerability: low, medium, high, critical.
    • Takes into account the severity of impact and likelihood of exploitation.
  • check
    • This field contains a regular expression, which can be used against the reported web3_clientVersion of a node. If the check matches, the node is with a high likelihood affected by the vulnerability.
  • CVE
    • The assigned CVE identifier, if available (optional)

​
What about GitHub security advisories

We prefer to not rely on GitHub as the only/primary publishing protocol for security advisories, but we plan to use the GitHub-advisory process as a second channel for disseminating vulnerability-information. Advisories published via GitHub can be accessed here.
⌘I