Skip to content

Response, New York. Noah Park: Check Point Patches Actively Exploited Management Server Path Traversal. Detection, Dallas. Ava Okello: Report Finds Nearly Half of Deployed Detections Need Attention Before SOCs Can Trust Them.

SOCtember

Always First. Fast SOC News.

MISP User Guide page on Warning Lists, showing the warninglist index in MISP 3.0.
Screenshot: MISP User Guide, Warning Lists chapter (MISP 3.0.0), published by CIRCL, dual-licensed AGPLv3 and CC BY-SA 4.0 (fair use).

Threat Intel

Did You Know: A MISP Warning List Hit Can Mean an Indicator Is Probably Noise or That It Sits on Shared Infrastructure, and Neither One Is a Verdict

MISP warning lists flag values such as private addresses, public DNS resolvers, empty-file hashes and cloud ranges. Each list carries a category, false_positive by default or known, and the two are different assessments that call for different handling. A hit means review the indicator, and a miss does not mean it is safe.

Priya Shah, Threat IntelSingapore3 min read

SINGAPORE - An indicator that sets off a warning in MISP is not automatically harmless, and one that sets off nothing is not automatically safe. The misp-warninglists project describes its lists as well-known indicators that can be associated with potential false positives, errors or mistakes. MISP shows a warning box at the event and attribute levels when a value matches one of them, and the lists are also used at the API level to help filter likely false positives. The project's README lists 164 of them, from private network ranges and public DNS resolvers to top-site rankings, cloud provider ranges and the hashes of empty files, all released under the CC0 public domain dedication.

misp-warninglists README describing the list types and the false_positive and known categories.
Screenshot: GitHub, MISP/misp-warninglists README, type and category of warning list (CC0 1.0).

The detail that matters most is the category field, which states what a hit actually means. The README defines two values. A false_positive list, the default, holds values very unlikely to be the indicator itself, such as RFC 1918 addresses, public DNS resolvers and popular domains, and the project says reports of them are usually extraction or logging artifacts. A known list holds shared or multi-tenant infrastructure: cloud and CDN ranges, URL shorteners, dynamic DNS, VPN providers and research scanners. For those, the README says a report naming the value as malicious can still be true, but blocking it or seeking a takedown will also hit unrelated users of the same infrastructure.

MISP User Guide table of warninglist matching rules and its note that false_positive and known are different assessments.
Screenshot: MISP User Guide, Warning Lists, Interpret the matching rules, published by CIRCL, dual-licensed AGPLv3 and CC BY-SA 4.0 (fair use).

The list files show the difference. The empty-hashes list sets no category, so it takes the false_positive default, and it includes d41d8cd98f00b204e9800998ecf8427e, the MD5 of an empty file. The dynamic-dns list is marked known and, in the version fetched on November 18, carried 45,603 hostname entries. The amazon-aws list is also marked known, with 4,083 address ranges. A sample hash that matches the first is almost certainly an artifact. A command-and-control domain that matches the second may be real, and the hit is a warning about what a block would break.

The dynamic-dns warning list on GitHub, marked with the category known.
Screenshot: GitHub, MISP/misp-warninglists, lists/dynamic-dns/list.json (CC0 1.0).

The MISP User Guide, written against MISP 3.0.0, is direct about the limits. It says a hit is a reason to review the observation and its intended use, not proof that it is benign or malicious, and that neither category establishes ground truth about an incident. It also says a lookup that returns no hit is not validation that a value is safe, because the relevant list may be disabled, absent, stale or unsuitable for that input. Only enabled lists take part in lookups. Warnings are normally checked only on attributes with the IDS flag set unless an administrator turns on MISP.warning_for_all, and changing the warning display does not change the IDS flag, block publication or delete data. The enforceWarninglist option on supported restSearch exports can drop matching attributes, and the guide notes that a private address may still matter inside an investigation even when it does not belong in a detection feed.

PyMISPWarningLists README table comparing its default exact-match search with slow_search matching by list type.
Screenshot: GitHub, MISP/PyMISPWarningLists README, Search modes (fair use).

Tools outside MISP can behave differently. The PyMISPWarningLists README says its default search checks only for exact membership in each list. Matching by list type, such as checking an IP address against CIDR ranges or a hostname against its parent domains, requires slow_search=True, and regex lists are not implemented in that mode. The package also loads a bundled snapshot unless it is told to fetch current lists. The MISP guide adds that an external consumer's local list set is not automatically the same as the server's enabled configuration.

For a threat intelligence team, the practical checks are short. Read the category before acting on a hit: treat false_positive matches as likely parsing noise, and treat known matches as real infrastructure that needs a narrower response than a blanket block. Confirm which lists are enabled on the server and in any script that consumes them, and how recently each was updated. Keep warned values in investigation context even when a detection export leaves them out, and never read an empty lookup as a clean bill of health.

About the Author

Priya Shah

Priya Shah covers threat intelligence, intrusion analysis, and adversary tradecraft for SOCtember from Singapore.

Related stories

Threat Intel desk