CTU-Capture-Malicious-Malware-GPC-1-1
This data was generated as part of a research project by the Stratosphere Laboratory, AI Center, FEE, Czech Technical University in Prague, Czechia. The goal is to store long-lived real botnet traffic and generate labeled netflow files for academic research.
These captures were created by Sebastian Garcia and Vojtěch Uhlíř. The captures were curated and verified in 2024 by Veronica Valeros. Contact us at stratosphere@aic.fel.cvut.cz, sebastian.garcia@agents.fel.cvut.cz.
Citation
Cite as: Garcia, Sebastian, and Uhlíř, Vojtěch. (2013). CTU-Capture-Malicious-Malware-GPC-1-1: a labeled dataset of real malicious network traffic [Data set] Zenodo. https://doi.org/10.5281/zenodo.10546994
Dataset specifications
- Dataset name: CTU-Capture-Malicious-Malware-GPC-1-1
- Dataset description: This dataset contains network traffic Windows VM (Win8)
- Dataset duration: 26 days
- Endpoint model: Windows VM
- Endpoint private IP: 10.0.2.22
- Endpoint public IP: N/A
- Endpoint default GW: 10.0.2.2
- Malware threat label: zbot
- Malware SHA256: 8330196e9f62ab96fde8d184d7629d73cd30127dc65050c7c55d586ce367c9c8
- Malware First Submission to Virus Total: 2013-08-17
- Number of days from first submission to VT to execution: 19 days
Dataset file description
The following files are included in the dataset:
- raw/
- .pcap: original packet capture file in pcap format
- zeek/
- conn.log.labeled
- Additional Zeek logs depending on the network traffic of each capture
- artifacts/
- labels.config: netflow labeler rule configuration file
- bin/
- .zip: malicious executable compressed with password “infected”
- README.md: Markdown README
- README.html: HTML README
Dataset timeline
- Capture started on: Thu Sep 5 15:40:07 CEST 2013
- Capture stopped on: Tue Oct 1 13:37:58 CEST 2013
Labels
This dataset was labeled by hand by security experts by analyzing the traffic and creating labeling rules. The program used was https://github.com/stratosphereips/netflowlabeler. The labels rules in the file labels.config, condense all the information needed to understand the labels in this capture.