SaMOSA: Sandbox for Malware Orchestration and Side-Channel Analysis CCS: Security and privacy Malware and its mitigation CCS: Security and privacy Virtualization and security CCS: Security and privacy Distributed systems security
Meet Udeshi, Venkata Sai Charan Putrevu, Prashanth Krishnamurthy, Ramesh Karri, Farshad Khorrami · 2025
Read SaMOSA: Sandbox for Malware Orchestration and Side-Channel Analysis CCS: Security and privacy Malware and its mitigation CCS: Security and privacy Virtualization and security CCS: Security and privacy Distributed systems security by Meet Udeshi, Venkata Sai Charan Putrevu, Prashanth Krishnamurthy, Ramesh Karri, Farshad Khorrami (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-14261.txt.
Introduction
Meet Udeshi and co-authors build a sandbox for dissecting malware that targets embedded devices, in this 2025 paper. Gadgets running factories and grids face rising attack, but safe test beds lag behind. Malware often hides until its command server calls or a trigger fires.
Their system, SaMOSA, replays captured command traffic on demand, so samples act out without live criminal servers. It tailors the fake environment across the sample's life cycle and records four side channels in sync. Several processor types are supported. A cryptominer case study on ARM hardware shows the sandbox catching real behavior.
Catalog record
| CREATOR | Meet Udeshi, Venkata Sai Charan Putrevu, Prashanth Krishnamurthy, Ramesh Karri, Farshad Khorrami |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-14261
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
SaMOSA: Sandbox for Malware Orchestration and Side-Channel Analysis CCS: Security and privacy Malware and its mitigation CCS: Security and privacy Virtualization and security CCS: Security and privacy Distributed systems security Meet Udeshi, Venkata Sai Charan Putrevu, Prashanth Krishnamurthy, Ramesh Karri, Farshad Khorrami Abstract . Cyber-attacks on operational technology (OT) and cyber-physical systems (CPS) have increased tremendously in recent years with the proliferation of malware targeting Linux-based embedded devices of OT and CPS systems. Comprehensive malware detection requires dynamic analysis of execution behavior in addition to static analysis of binaries. Safe execution of malware in a manner that captures relevant behaviors via side-channels requires a sandbox environment. Existing Linux sandboxes are built for specific tasks, only capture one or two side-channels, and do not offer customization for different analysis tasks. We present the SaMOSA Linux sandbox that allows emulation of Linux malwares while capturing time-synchronized side-channels from four sources. SaMOSA additionally provides emulation of network services via FakeNet, and allows orchestration and customization of the sandbox environment via pipeline hooks. In comparison to existing Linux sandboxes, SaMOSA captures more side-channels namely system calls, network activity, disk activity, and hardware performance counters. It supports three architectures predominantly used in OT and CPS namely x86-64, ARM64, and PowerPC 64. SaMOSA fills a gap in Linux malware analysis by providing a modular and customizable sandbox framework that can be adapted for many malware analysis tasks. We present three case studies of three different malware families to demonstrate the advantages of SaMOSA. 1. Introduction The proliferation of embedded computers connected to the internet has grown immensely with the modernization of operational technology (OT) and cyber-physical systems (CPS). Majority of these run a Linux operating system (OS) on ARM, PowerPC or x86 architectures ( Alhanahnah et al. 2018 ; Carrillo-Mondéjar et al. 2020 ) . In effect, malwares targetted towards Linux-based embedded devices have seen a proportional spike. While malware analysis has long focused on Windows due to widespread adoption and higher prevalence of attacks, analysis of Linux binaries is extremely important for OT and CPS security, so that malwares are detected before they cause lasting damage. While static analysis can reveal known malwares based on code signatures, an unseen malware’s intentions become more apparent during execution, thus requiring dynamic analysis of execution behavior. Dynamic analysis of malware involves analysis of its execution trace captured via different side-channel sources such as system calls (syscall), hardware performance counters (HPC), network activity, and disk activity ( Cozzi et al. 2018 ; Udeshi et al. 2025 ) . Many approaches use statistical and machine learning based methods to detect anomalous execution behavior ( Abed et al. 2015 ) . While methods focused on embedded devices of OT and CPS primarily use microarchitectural side-channels such as HPC ( Krishnamurthy et al. 2024 ; Wang et al. 2016a ; Wang et al. 2016b ) , some incorporate multiple sources simultaneously for more elaborate detection ( Zheng et al. 2023 ) . For dynamic analysis, safe execution of binaries requires a sandbox environment where a malware is sufficiently isolated so that it cannot affect important systems, yet it is provided all capabilities to perform malicious actions. Additionally, the sandbox should capture all relevant information about the binary’s execution such that any malicious behavior is meaningfully tracked and can be detected. We present the SaMOSA sandbox that is tailored for Linux malware analysis, supports multiple processor architectures popular in OT and CPS, captures multiple side-channels for in-depth analysis, provides pipeline hooks for custom orchestration of the sandbox environment, and automates the end-to-end execution pipeline. Alrawi et al. ( Alrawi et al. 2024 ) studied malware sandboxes of the last 20 years and proposed sandbox design guidelines in which they outlined that the sandbox should rely on emulation or virtualization, be modular and customizable, perform monitoring outside the sandbox environment as much as possible, provide a realistic OS along with an emulated network environment to trigger malware behavior, and collect monitoring data from multiple components for comparative analysis. SaMOSA is designed in line with these guidelines. The contributions of this paper are threefold: (1) The SaMOSA sandbox for automated malware emulation and analysis on Linux for x86, ARM, and PowerPC (2) Orchestration framework to run user-defined commands and scripts via pipeline hooks, providing versatility and customization for different malware analysis (3) Capturing four time-synchronized side-channels (system calls, hardware performance counters, network activity, disk activity) for deep insights into malware behavior 2. Related Work Figure 1. Overview of the SaMOSA sandbox. SaMOSA involves three stages: orchestration where the user configures hooks to customize the sandbox environment, execution which executes the binary while collecting side-channels, and analysis where the time-synchronized data is post-processed and plotted to observe correlations across multiple side-channels. Table 1. Comparison of existing Linux sandboxes Sandbox Multi Arch Network Mon. HPC Mon. Disk Mon. Syscall Mon. Network Emu. Orchestration Limon ( Monnappa 2015 ) ✗ ✓ ✗ ✗ ✓ ✗ ✗ F-Sanbox ( Phu et al. 2019 ) ✗ ✓ ✗ ✗ ✓ ✗ ✗ Detux ( Detux [n. d.] ) ✓ ✓ ✗ ✗ ✓ ✗ ✗ Padawan ( Cozzi et al. 2018 ) ✓ ✓ ✗ ✗ ✓ ✓ ✗ ELFEN ( ELFEN [n. d.] ) ✓ ✓ ✗ ✗ ✓ ✗ ✗ LiSa ( Uhrıcek 2020 ) ✓ ✓ ✗ ✗ ✓ ✗ ✗ SaMOSA (ours) ✓ ✓ ✓ ✓ ✓ ✓ ✓ Several sandboxes have been developed for binary analysis. Cuckoo Sandbox ( Cuckoo [n. d.] ) and its open-source successor CAPE Sandbox ( CAPE [n. d.] ) are widely used for Windows, however they do not support Linux binaries and require complex, time-consuming configuration. To address this gap, several Linux sandboxes were developed. Limon ( Monnappa 2015 ) and Detux ( Detux [n. d.] ) are Python-based sandboxes for emulating Linux malwares. However, their scripts and VM images are out-dated, posing challenges in updating to latest OS and software needed to analyze recent malware. F-sandbox ( Phu et al. 2019 ) , derived from Firmadyne ( Chen et al. 2016 ) and Detux frameworks, is specifically designed for ELF malware targeting MIPS architecture. Padawan sandbox ( Cozzi et al. 2018 ) performs static analysis via binary disassembly and employs QEMU for sandboxing and dynamic analysis with support for multiple architectures. It captures low-level kernel and userspace events via SystemTap. However, Padawan is closed-source and not publicly available, posing similar challenges in updating to latest software. LiSa Sandbox ( Uhrıcek 2020 ) and ELFEN ( ELFEN [n. d.] ) are recent fully automated Linux malware analysis platforms, supporting static analysis via strings and YARA signatures and dynamic analysis via QEMU emulation. LiSa also captures system calls via SystemTap and additionally captures network activity from inside the sandbox, while ELFEN relies on eBPF filters. Both LiSa and ELFEN use buildroot-based custom-compiled OS images that provide a minimal Linux environment which may not represent a full-featured OS environment like Ubuntu. Also, using SystemTap or eBPF for capturing system calls and events requires implementing custom filters for specific events which is more suitable for debugging specific issues rather than capturing comprehensive system information for malware analysis. Although these sandboxes target Linux, they are limited in the environments they provide and the execution