Research Archives | JUMPSEC LABS

Online Machine Learning: how to integrate user feedback

by Newt Tan | Dec 12, 2022 | Research

When designing and implementing a machine learning model, ensuring it is continually updated is a challenge that all engineers encounter. In this article, I explore the online machine learning technique that I used during a project and present how it was implemented for effective results. Choosing a machine learning method Machine learning solutions can be mainly split into offline and online methods. Online machine learning is a method in which data becomes available in a sequential...

Implementation and Dynamic Generation for Tasks in Apache Airflow

by Newt Tan | Nov 23, 2022 | Detection, Research

I recently worked on a project focused on log anomaly detection using manageable machine learning pipelines. The pipelines mainly include data collection --- feature extraction --- feature engineering --- detection/prediction --- updating (maintenance). It’s important to have a solid UI to manage the pipelines so I can easily review the chain of pipelines. After much research, I found many engineers recommended Airflow. In airflow, the core concept is the Directed Acyclic Graph...

Advisory CVE-2021-41551 Leostream Connection Broker – Authenticated Zip Slip

by lrckt | Jan 26, 2022 | Research, Security Bug, Vulnerability

Software: Leostream Connection BrokerAffected Versions: 9.0.40.17Vendor page: https://leostream.com/CVE Reference: CVE-2021-41551Published: 25/01/2022Attack Vector: path traversal, authenticatedCredits: Andrei Constantin Scutariu, Lenk Ratchakrit Seriamnuai, Andrea Malusardi Summary Leostream Connection Broker 9.0.40.17 allows administrators to conduct directory traversal attacks by uploading a ZIP file that contains a symbolic link. Mitigation The Leostream has released a patch for this...

Advisory CVE-2021-41550 Leostream Connection Broker – Authenticated Remote Code Execution

by lrckt | Jan 26, 2022 | Research, Security Bug, Vulnerability

Software: Leostream Connection BrokerAffected Versions: 9.0.40.17Vendor page: https://leostream.com/CVE Reference: CVE-2021-41550Published: 25/01/2022Attack Vector: Remote, authenticatedCredits: Andrei Constantin Scutariu, Lenk Ratchakrit Seriamnuai, Andrea Malusardi Summary As the Leostream Connection Broker version: 9.0.40.17 allowed an attacker to upload any content through Third Party Content functionality, it was found that the application allowed the listed filenames below the ability to...

Burp Suite and Beyond: Exploring non-HTTP protocols using MITM_RELAY

by editor | Aug 24, 2021 | Application Security, burpsuite, network, Network Tools, Research

In this article, Muhammet takes us on a deep technical journey to persevere beyond the limitations of the proxy tool Burpsuite, and explore non-HTTP, application-layer protocols using ‘MITM RELAY’.

Detecting known DLL hijacking and named pipe token impersonation attacks with Sysmon

by xnand | Nov 13, 2020 | Detection, Jumpsec, Research

Recently we posted a bunch of advisories relating to Ivanti Unified Endpoint Manager, a couple of which are for vulnerabilities which can be used to achieve local privilege escalation. We will give a brief explanation of the vulnerabilities and an example of Sysmon configuration rules to log exploitation attempts, along with the rationale behind them so you can adapt them to your existing configuration if needed.

Advisory CVE-2020-13773 – Ivanti Unified Endpoint Manager Reflected XSS

by xnand | Nov 13, 2020 | Jumpsec, Research

Software: Ivanti Endpoint ManagerAffected Versions: <= 2020.1.1Vendor page: www.ivanti.comCVE Reference: CVE-2020-13773Published: 13/11/2020CVSS 3.1 Score: 5.5 - AV:N/AC:L/PR:L/UI:R/S:U/C:L/I:L/A:LAttack Vector: Remote, authenticatedCredits: Andrei Constantin Scutariu, Lenk Ratchakrit, Calvin Yau Summary Various web pages on Ivanti Unified Endpoint Manager web management console lack proper input validation on parameters passed in HTTP request, leaving the application vulnerable to...

Advisory CVE-2020-13769 – Ivanti Unified Endpoint Manager SQL injection

by xnand | Nov 13, 2020 | Jumpsec, Research

A number of web components in Endpoint Manager do not properly sanitize user input when executing SQL queries, leaving the application vulnerable to injection attacks towards the underlying database. On a standard installation with default options, the account used to query the database is database administrator.

JUMPSEC LABS

Online Machine Learning: how to integrate user feedback

Implementation and Dynamic Generation for Tasks in Apache Airflow

Advisory CVE-2021-41551 Leostream Connection Broker – Authenticated Zip Slip

Advisory CVE-2021-41550 Leostream Connection Broker – Authenticated Remote Code Execution

Burp Suite and Beyond: Exploring non-HTTP protocols using MITM_RELAY

Detecting known DLL hijacking and named pipe token impersonation attacks with Sysmon

Advisory CVE-2020-13773 – Ivanti Unified Endpoint Manager Reflected XSS

Advisory CVE-2020-13769 – Ivanti Unified Endpoint Manager SQL injection

[Latest Posts]

[Categories]

GitHub

Follow JumpsecLabs

Disclaimer