Code Red: check the measuring instrument, too
PERIOD
July 2001
EDITION
Short field file
REVIEWED
2026-11-19
The short version
CAIDA’s July 19, 2001 study examined the random-seed variant of Code Red, called CRv2. It observed spreading attempts against vulnerable IIS web servers using traffic collected at UCSD and Lawrence Berkeley Laboratory.
The flaw and the conditions
Microsoft’s bulletin describes an unchecked buffer in an Index Server ISAPI extension. The relevant mapping and reachable web server mattered; the indexing service itself did not need to be running. “We do not use indexing” was therefore not a complete exposure assessment.
The peak that needed an explanation
CAIDA identified a conspicuous spike in its graph as a collection-gap artifact. Observations accumulated during missing collection time were represented when collection resumed. The peak did not establish that all those infections occurred in one sudden burst.
From the workbench
Before explaining a dramatic chart, ask how the measurements were made. A graph can reflect both the event and the instrument watching it.
Limits of this file
This concerns the study’s CRv2 scope, not every Code Red variant. It omits definitive global counts and claims about the author’s identity or intent.
