<?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 Jake Williamson on Medium]]></title>
        <description><![CDATA[Stories by Jake Williamson on Medium]]></description>
        <link>https://medium.com/@jwilliamson7176?source=rss-af78b4da01cb------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*91Eh0t91uITulNvh</url>
            <title>Stories by Jake Williamson on Medium</title>
            <link>https://medium.com/@jwilliamson7176?source=rss-af78b4da01cb------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 22:12:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jwilliamson7176/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[Sauna]]></title>
            <link>https://medium.com/hack-the-box-machines/sauna-fcb78781db17?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/fcb78781db17</guid>
            <category><![CDATA[pass-the-hash]]></category>
            <category><![CDATA[kerberos]]></category>
            <category><![CDATA[hackthebox]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Tue, 07 Jul 2020 23:39:59 GMT</pubDate>
            <atom:updated>2020-07-07T23:39:59.827Z</atom:updated>
            <content:encoded><![CDATA[<p>My first machine on Hack The Box, and we’re dealing with Kerberos</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*wMFxiRyYszPKjPmPpub1wQ.png" /><figcaption>Figure 1 — Sauna Difficulty</figcaption></figure><p>This machine was classed as Easy, although the majority of people who completed the box rated it Medium. Nevertheless, it said Easy on the front so I went for it.</p><h3>First Thing’s First</h3><p>Once connecting to the VPN, I did the first thing any sane person would do … ping the box. As this was my first time connecting to the Hack The Box environment, I wanted to make sure that everything was working. I also checked to make sure I’d been assigned an IP address on HTB’s network.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/542/1*vx4D0LoMvgKswZVEwMn2-w.png" /><figcaption>Figure 2 — ping and IP address results</figcaption></figure><p>With everything working as it should, it was time to start seeing what I could find out about the box I was dealing with.</p><h3>Enumeration</h3><p>As most of us do at the beginning, I ran <a href="https://nmap.org/">Nmap</a> to see what services were running on the box, and well, there were quite a few services.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wmurxn2Z5NR4ey4Q1P00eQ.png" /><figcaption>Figure 3 — Nmap output</figcaption></figure><p>The first ports that immediately jumped out at me were 80, 88, and 389. I decided to turn my reconnaissance towards these ports. Port 80 told me that it was running a <a href="https://www.iis.net/">Microsoft IIS</a> web server and therefore could have a vulnerable webpage. Port 88 showed that it was using <a href="https://web.mit.edu/kerberos/#what_is">Kerberos</a> for authentication to the server and if we get the right credentials I could use that for a connection. Port 389 identified the system as a Domain Controller, I could use some enumeration techniques more suited for Active Directory if the webserver didn’t play out. The remaining ports I wrote down (even though they were saved in the output file), I would come back to those later if needed.</p><p>In an attempt to gain more information about the system, I also ran Nessus. Nothing noteworthy turned up, so I left that dead-end and decided to move on to the website.</p><h3>The Website</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WhRyUKZuU9_AzFP_vTDZ6Q.png" /><figcaption>Figure 4 — The Website</figcaption></figure><p>Whilst their website looked fully operational, a quick dig around found that clicking on links or submitting forms returned a 405 Error stating that the HTML verb couldn’t be used. My first thought was to change the verb used from POST to GET to see if we could access any other webpages or submit any information to the machine. This was to no avail, everything tried kept returning a 405 error. With this information, I turned to <a href="https://tools.kali.org/web-applications/dirbuster#">Dirbuster</a> to see if that could find any hidden pages or config files on the website. After letting it run for a while, and eventually stopping it because it wasn’t returning anything extra, I ended up with the following pages:</p><ul><li>index.html</li><li>blog.html</li><li>contact.html</li><li>single.html</li><li>about.html</li></ul><p>Reading through each of the pages, there wasn’t anything exploitable that I hadn’t already tried. My initial thought was that the website was a dead-end and wouldn’t contain any information until I got to the bottom of <em>about.html</em>. There was a team list detailing all the staff that worked for Egotistical Bank (granted it was only 6 people). This would be useful as we were leaving the website behind and moving on to Kerberos and Active Directory, at least we would have a list of users to go off.</p><h3>Kerberos</h3><p>The first thing I had to do, before getting dirty with Kerberos, generate a list of usernames. I’m sure there’s a tool out there that does this for me, but I ended up writing this out by hand until I had a list I was comfortable to use for a first pass.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/820/1*IudfHFkrBX5GqgF6JAz9gA.png" /><figcaption>Figure 5 — The Userlist</figcaption></figure><p>After a bit of Googling, I found a tool that could be used to brute force Kerberos, <a href="https://github.com/TarlogicSecurity/kerbrute">Kerbrute.py</a>. I ran this script with my user list and a password list from Seclists. Eventually, Kerbrute found what I was looking for, a user that didn’t require pre-authentication (if you want to know why this is important, <a href="https://m0chan.github.io/2019/07/31/How-To-Attack-Kerberos-101.html#as-rep-roasting">this article</a> provides a good explanation).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/1*RJfoBVcOYpkMzKDLjelIUg.png" /><figcaption>Figure 6 — Kerbrute output</figcaption></figure><p>Whilst Kerbrute wasn’t able to crack the password, it did show me which user didn’t require pre-authentication. With this information, I was able to use GetNPUsers.py (part of <a href="https://github.com/SecureAuthCorp/impacket">Impacket</a>) to extract a hash of the FSmith’s password. GetNPUsers returned a file containing the hash of FSmith’s password in a format readable by John The Ripper, so I ran this file through John and was presented with a password that I’d be able to use to login as FSmith.</p><blockquote>Note: Due to HTB’s rules, I won’t show the output of John or the hash of the password. Don’t want to be told off after my first box.</blockquote><h3>Win-RM</h3><p>So, I had a username and a valid password. My next goal was to use these to log in to the machine. It took me a while to figure out how to use the credentials, but after banging my head against the wall I decided to go back to the beginning. I checked back over the output of Nmap and the ports I had written down and Googled each one of them. Eventually, I got down to port 5985 and found an article telling me what Win-RM was. It essentially told me that I could use Win-RM to communicate with Windows computers on my network and send commands; promising. Additional searching brought me to <a href="https://github.com/Hackplayers/evil-winrm">evil-winrm</a>, a tool that used this service to provide a shell (assuming we had the correct credentials).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*IMPqgzCiJp-5KmuwwMhvYA.png" /><figcaption>Figure 7 — User Level Access</figcaption></figure><h3>Evil-WinRM</h3><p>I’m in. As shown in Figure 7, I’m connected to the Windows Box and logged in as FSmith. For those of you who have done HTB before, you’ll know where to head for the flag. For those that haven’t, it’s only a quick Google away.</p><p>My next job was to obtain root access. Having little experience in escalating privileges on AD, I resorted to a fair amount of research. This research leads to another of Impacket’s modules: SecretsDump.py. I would need to have valid credentials from a Domain Admin, which turned out not to be FSmith. So I took back to Google and found a page from Microsoft that describes how to configure automatic logon, rather than configuring it, I queried that registry key and found a service account with a username and password.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/674/1*iE61qh7AeKku_jasgJikGg.png" /><figcaption>Figure 8 — A service account with automatic login</figcaption></figure><p>Figure 8 show’s the registry file containing the username (svc_loanmanager) and password (slightly offscreen) for this login. So, I went back to SecretsDump.py and attempted it again with these credentials. It didn’t work. But this time, it gave me a different error to the one I got from FSmith. This time, it told me that it was due to a bad username. This threw me as I presumed the login was valid from the registry dump. After a while of Googling, I decided to retry SecretsDump.py but instead of using “svc_loanmanager” as the login, I would try various spellings. The second spelling hit; “svc_loanmgr” dumped a wonderful load of usernames and hashes onto my screen, including one for Administrator. Heading back to Impacket, I had a look through their tools to see if they had something available for Pass The Hash attacks.</p><h3>wmiexec.py</h3><p>wmiexec.py is a tool by Impacket that allows a user to perform a Pass The Hash attack. All I had to do was run the script with the domain, username, and hash of the password, and it would log me in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/317/1*2PIKKiBP0rm3PFSWSgg-xg.png" /><figcaption>Figure 9 — ROOTED!</figcaption></figure><p>Which thankfully, it did. I had successfully got to Administrator level on the machine. All that was left to do was get the flag and enter it into HTB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*Idqsjl9Rh-_JJX4q.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fcb78781db17" width="1" height="1" alt=""><hr><p><a href="https://medium.com/hack-the-box-machines/sauna-fcb78781db17">Sauna</a> was originally published in <a href="https://medium.com/hack-the-box-machines">Hack The Box Machines</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Malicious Word Documents | Android Phishing Attacks | CISCO Forensic Guides | AI Voice Fraud]]></title>
            <link>https://medium.com/jakes-security-blog/malicious-word-documents-android-phishing-attacks-cisco-forensic-guides-ai-voice-fraud-154894cce0bf?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/154894cce0bf</guid>
            <category><![CDATA[cisco]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[malware-analysis]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[android]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Thu, 05 Sep 2019 21:59:55 GMT</pubDate>
            <atom:updated>2019-09-05T22:00:21.219Z</atom:updated>
            <content:encoded><![CDATA[<h4>05/09</h4><p><strong>Encrypted Resume Malicious Word Documents</strong></p><p>Brad Duncan (<a href="https://blog.paloaltonetworks.com/author/bduncan/">https://blog.paloaltonetworks.com/author/bduncan/</a>) has an incredibly educational malware analysis post on the latest version of the encrypted resume Malspam. These Word documents arrive encrypted and the body of the email will list the password. The sample that Brad looked at ended up delivering the Remcos Remote Access Trojan (<a href="https://www.virustotal.com/gui/file/fa9a94b32f7fa1e1e3eef63d3fb9003fda8d295e1f1a3e521691725e4c7da9f3/detection">VT</a>). Brad provides access to the samples and the PCAP he analysed.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/754/0*7gKIdKE1kLZVoFZQ.gif" /></figure><p><a href="https://isc.sans.edu/forums/diary/Malspam+using+passwordprotected+Word+docs+to+push+Remcos+RAT/25292/">https://isc.sans.edu/forums/diary/Malspam+using+passwordprotected+Word+docs+to+push+Remcos+RAT/25292/</a></p><p><strong>Android SMS Phishing Attack</strong></p><p>Checkpoint, in a blog post, disclosed an interesting weakness in many Android phones that can be used to send SMS messages with OTA provisioning data. OTA provisioning is used by many network providers to supply phones with basic configuration options, such as the APN settings and proxy IP addresses. Spoofed messages being sent to the phone can then be used to change the proxy settings, granting the attack a MITM position. Particularly vulnerable phones are the Samsung mobile devices which accept unauthenticated OTA provisioning messages, so all the attacker needs to do is send a specifically formatted message to the phone as an SMS. To create the SMS message, the attacker needs a simple USB modem that allows them to send these binary SMS messages to the victim. Other phones use weak authentication and in order to break this, the only additional thing the attacker needs is the phones IMSI number (the identifier for the phone); if that isn’t available then the attacker can send a PIN-protected message to the phone. By doing it with a PIN, it is a little more complex as there is no guarantee that the user will enter the PIN, Checkpoint is suggesting that attackers are sending an ordinary SMS first informing users of an update and then listing the PIN that they will need to perform this update, then sending the update and prompting users for the PIN. The processing of these messages is not part of the Android Operating System which is why some phones are more vulnerable than others, as it depends on the manufacturer. The post also doesn’t mention iOS and whether it is vulnerable to these types of attacks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Mt3VZlAa6-_IEo4v.jpg" /></figure><p><a href="https://research.checkpoint.com/advanced-sms-phishing-attacks-against-modern-android-based-smartphones/">https://research.checkpoint.com/advanced-sms-phishing-attacks-against-modern-android-based-smartphones/</a></p><p><strong>CISCO Releases Forensic Guides</strong></p><p>CISCO has released a series of guides detailing how to provide forensic analysis on its devices. Four guides have been published for ASA, Firepower, iOS, and iOS XE. These guides go over how to do many things such as verifying the integrity of the firmware using hashes; how to read out log files; and how to create memory dumps for future analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aoj6rIvQyym114He.png" /></figure><p><a href="https://tools.cisco.com/security/center/tacticalresources.x">https://tools.cisco.com/security/center/tacticalresources.x</a><br><a href="https://tools.cisco.com/security/center/resources/asa_forensic_investigation">https://tools.cisco.com/security/center/resources/asa_forensic_investigation</a><br><a href="https://tools.cisco.com/security/center/resources/ftd_forensic_investigation">https://tools.cisco.com/security/center/resources/ftd_forensic_investigation</a><br><a href="https://tools.cisco.com/security/center/resources/ios_forensic_investigation">https://tools.cisco.com/security/center/resources/ios_forensic_investigation</a><br><a href="https://tools.cisco.com/security/center/resources/iosxe_forensic_guide">https://tools.cisco.com/security/center/resources/iosxe_forensic_guide</a></p><p><strong>AI Mimics CEO’s Voice To Transfer Money</strong></p><p>CEO fraud has just gone up a level, upgrading from emails to voice phone calls claiming to be a high-level executive within the company. A UK company fell for this scam after the voice at the end of the phone claimed to come from the German HQ of this company. The recipient of this scam thought they recognised the voice of the CEO and followed the instructions to transfer money. Without hearing the call, it is difficult to say how sophisticated this attack is, but the victim claimed that they have spoken to the CEO in the past, and therefore should be able to recognise the voice.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/794/0*PAH9Er2LHEEwmct4.jpg" /></figure><p><a href="https://www.wsj.com/articles/fraudsters-use-ai-to-mimic-ceos-voice-in-unusual-cybercrime-case-11567157402">https://www.wsj.com/articles/fraudsters-use-ai-to-mimic-ceos-voice-in-unusual-cybercrime-case-11567157402</a><br><a href="https://www.scmagazineuk.com/ai-mimics-ceo-voice-scam-uk-energy-firm-200k/article/1595277">https://www.scmagazineuk.com/ai-mimics-ceo-voice-scam-uk-energy-firm-200k/article/1595277</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=154894cce0bf" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jakes-security-blog/malicious-word-documents-android-phishing-attacks-cisco-forensic-guides-ai-voice-fraud-154894cce0bf">Malicious Word Documents | Android Phishing Attacks | CISCO Forensic Guides | AI Voice Fraud</a> was originally published in <a href="https://medium.com/jakes-security-blog">Jake’s Security Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Trick Bot | Supermicro Vulnerability | Facebook Loses Control of Secret Key]]></title>
            <link>https://medium.com/jakes-security-blog/trick-bot-supermicro-vulnerability-facebook-loses-control-of-secret-key-929d37ebbb86?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/929d37ebbb86</guid>
            <category><![CDATA[trickbot]]></category>
            <category><![CDATA[trojan]]></category>
            <category><![CDATA[free-basics]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[supermicro]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Wed, 04 Sep 2019 22:32:31 GMT</pubDate>
            <atom:updated>2019-09-10T08:51:04.966Z</atom:updated>
            <content:encoded><![CDATA[<h4>04/09</h4><p><strong>Trick Bot</strong></p><p>There is an interesting variant of <a href="https://blog.malwarebytes.com/detections/trojan-trickbot/">Trick Bot</a> out in the wild recently. The interesting aspect of this is the link file that is used to spread this variant. Link files usually link to other files, but usually can contain content themselves; in this case, in addition to the link file itself, there is data being appended to it. Windows usually ignores data that is appended to a link file unless it is properly formed, however, the script inside the link file will extract the next stage of the payload from this data. The code, once extracted from the link file, will download additional stages that turns out to be another variant of Trick Bot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/472/0*IcZ8YM33e_7kwVwa.png" /></figure><p><a href="https://isc.sans.edu/forums/diary/Guest+Diary+Tricky+LNK+points+to+TrickBot/25290/">https://isc.sans.edu/forums/diary/Guest+Diary+Tricky+LNK+points+to+TrickBot/25290/</a></p><p><strong>Supermicro Virtual USB Vulnerability</strong></p><p>The security company <a href="https://eclypsium.com/">Eclypsium</a> released details about a vulnerability in some Supermicro BMC motherboard management controllers. One of the features implemented by these controllers is Virtual USB devices. These virtual devices are essentially USB drives that can be connected to servers across the internet. According to Eclypsium, authentication to connect these USB devices isn’t implemented correctly in some of Supermicro’s BMCs and as a result, an attacker can connect their USB device without authentication. The problem is if an administrator did connect a USB device and authenticated correctly, then disconnected the USB device, the internal state about the connection remains and it is now possible for an attacker to connect a different device without having to authenticate first. This state is only cleared correctly after the BMC itself gets rebooted. So, rebooting the server may not clear this out; this state may remain for an extended time after the administrator has disconnected their device. Typically only storage devices are connected using this method, but the interface being implemented is essentially for a generic USB device, so the attacker could connect a USB device of their choice and use that device to compromise the server. Supermicro has developed a patch, so applying these patches should be a priority. However, for just a quick fix, rebooting the BMC without rebooting the entire server should resolve any issues for the time being. This service listens on port 623, so on the firewall restrict any remote access to the network using this port. Also, in addition to this exploit, Eclypsium points out three other vulnerabilities. This entire system uses plaintext for authentication, the network traffic is either not encrypted or it uses RC4 with a fixed key that is compiled into the BMC firmware — which is the same key across all BMCs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*tNlitWynIV6zmtO0.png" /></figure><p><a href="https://eclypsium.com/2019/09/03/usbanywhere-bmc-vulnerability-opens-servers-to-remote-attack/">https://eclypsium.com/2019/09/03/usbanywhere-bmc-vulnerability-opens-servers-to-remote-attack/</a></p><p><strong>Facebook Key Used to Sign Unrelated Android Apps</strong></p><p>Facebook has lost control over one of its software signing used to sign its Free Basics application. Free Basics is a lightweight version of the Facebook Android app for users in areas with limited internet access and, like all Android applications, it is signed and of course, a secret key is needed to create these signatures. The same secret key has now been spotted signing unrelated applications that are not part of Facebook, suggesting that the actual secret key that was used to sign the Free Basics application has now been leaked. This means that anyone who has the secret key would be able to release a malicious update for the Free Basics application. Facebook has responded to this by publishing a new version of the application that is signed with a different key, so definitely update if you are using this application.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/512/0*r7bi9CzSzr5wMVv3.png" /></figure><p><a href="https://www.androidpolice.com/2019/08/29/cryptographic-key-used-to-sign-one-of-facebooks-android-apps-compromised/">https://www.androidpolice.com/2019/08/29/cryptographic-key-used-to-sign-one-of-facebooks-android-apps-compromised/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=929d37ebbb86" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jakes-security-blog/trick-bot-supermicro-vulnerability-facebook-loses-control-of-secret-key-929d37ebbb86">Trick Bot | Supermicro Vulnerability | Facebook Loses Control of Secret Key</a> was originally published in <a href="https://medium.com/jakes-security-blog">Jake’s Security Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Malware Installs Node.js | Dovecot Vulnerability | Cloudflare Spreads Malware]]></title>
            <link>https://medium.com/jakes-security-blog/malware-installs-node-js-dovecot-vulnerability-cloudflare-spreads-malware-5581de576fc7?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/5581de576fc7</guid>
            <category><![CDATA[malware]]></category>
            <category><![CDATA[nodejs]]></category>
            <category><![CDATA[cloudflare]]></category>
            <category><![CDATA[security]]></category>
            <category><![CDATA[dovecot]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Tue, 03 Sep 2019 19:02:19 GMT</pubDate>
            <atom:updated>2019-09-10T08:51:25.249Z</atom:updated>
            <content:encoded><![CDATA[<h4>03/09</h4><p><strong>Malware Installs Node.js</strong></p><p>Xavier Mertens (<a href="https://xavier.mertens.consulting/">https://xavier.mertens.consulting</a>) has come across another malware sample that actually downloads, and installs, a complete copy of Node.js on Windows systems. Node.js has become a really popular development platform due to the libraries and capabilities that it offers; in this case, it is not really clear why the attacker is going to the trouble of downloading it all. Xavier took a look at the malware and it appears to implement a Command and Control channel, this may not be a finalised edition of the malware and it may indicate that a more detrimental piece of malware will appear soon. Node.js, obviously, is not malware by itself and isn’t flagged as such by any antivirus, this could mean that it is an attempt to slip past the antivirus and load miscellaneous JavaScript snippets that aren’t detected — or even analysed. In this particular case, the attack JavaScript that is implementing this Command Control channel is actually being delivered as a Base64 Encoded comment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MuVcoMPyJcq8G4qf5s3HGQ.png" /></figure><p><a href="https://isc.sans.edu/forums/diary/Malware+Dropping+a+Local+Nodejs+Instance/25284/">https://isc.sans.edu/forums/diary/Malware+Dropping+a+Local+Nodejs+Instance/25284/</a></p><p><strong>PigeonHole and Dovecot Vulnerability</strong></p><p>One of the most popular servers for IMAP when creating a mail server in a UNIX environment is Dovecot. A remote code execution vulnerability was patched 5 days ago (28/08/2019), the notes do mention a Proof of Concept on how to execute this exploit and also states that actual remote code execution is difficult to accomplish in this case. As well as remote code execution, this vulnerability could also be used to read memory from Dovecot. So whilst there hasn’t been an exploit yet, and there is unlikely to be an exploit soon, patches should be available for the UNIX distribution you are using.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/1*4KMq4XBzvuMBBE95tjU-HA.png" /></figure><p><a href="https://www.openwall.com/lists/oss-security/2019/08/28/3">https://www.openwall.com/lists/oss-security/2019/08/28/3</a></p><p><strong>Cloudflare Workers Used to Spread Malware</strong></p><p>Malware authors often use Cloudflare in order to obfuscate the actual origin of their malware and also to bypass some blocklists. Typically this is done by using Cloudflare as proxy service, according to Marcel Afrahim (<a href="https://www.linkedin.com/in/marcel~/">https://www.linkedin.com/in/marcel~/</a>) however, the <a href="https://www.symantec.com/security-center/writeup/2019-071014-5108-99">Astaroth trojan</a> took this a step further by using Cloudflare’s Workers Feature. Cloudflare Workers is essentially code that you run within Cloudflare; it is intended for things like more intelligent load balancing. It works by allowing the user to upload JavaScript snippets to Cloudflare, then it is run in Cloudflare to serve requests being sent to specific URLs. This type of serverless computing is becoming more popular recently, with Amazon having its own Lambda service to do similar things. However, Cloudflare offers a free tier that allows users to run 100,000 workers per day for free. Apparently, this is what Astaroth is doing in its latest version to make it more difficult to block these requests. Marcel has created a blog post on Medium that goes into considerable detail on how this is accomplished and how the random URLs are being generated using this method to make it more difficult to determine any meaningful indicators of compromise.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/532/0*gtvubZE9S3x5I2tl.jpg" /></figure><p><a href="https://medium.com/@marcelx/threat-actor-behind-astaroth-is-now-using-cloudflare-workers-to-bypass-your-security-solutions-2c658d08f4c">https://medium.com/@marcelx/threat-actor-behind-astaroth-is-now-using-cloudflare-workers-to-bypass-your-security-solutions-2c658d08f4c</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5581de576fc7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jakes-security-blog/malware-installs-node-js-dovecot-vulnerability-cloudflare-spreads-malware-5581de576fc7">Malware Installs Node.js | Dovecot Vulnerability | Cloudflare Spreads Malware</a> was originally published in <a href="https://medium.com/jakes-security-blog">Jake’s Security Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[iOS Exploits | Twitter CEO Hacked]]></title>
            <link>https://medium.com/jakes-security-blog/ios-exploits-twitter-ceo-hacked-831721881823?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/831721881823</guid>
            <category><![CDATA[twitter]]></category>
            <category><![CDATA[ios]]></category>
            <category><![CDATA[security]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Mon, 02 Sep 2019 21:45:57 GMT</pubDate>
            <atom:updated>2019-09-02T21:45:57.759Z</atom:updated>
            <content:encoded><![CDATA[<h4>02/09</h4><p><strong>iOS Exploits</strong></p><p>Over the weekend, Google came out with a set of blog posts detailing a targeted attack against iPhone users. These attacks were targeted in the sense that the exploit code was deposited on very specific websites, however, Google does not mention these sites. But they categorise it as a water hole attack which typically means that the website infected with the malicious code is usually targeting a particular community, therefore any user visiting this website was infected if their phone was vulnerable. Typically, when we discuss mobile malware, we talk about malware that a user specifically installs; not knowing that what they are installing is actually malicious. These exploits are different. These exploits are a more dangerous “drive-by” exploits where a user visits a website that is infected with the malicious code, then as a result, the malicious code is installed on the user’s phone. If done correctly, the user is completely unaware of what is happening and has no idea that the phone is infected at all. These exploits are, by default, more complicated and typically require an exploit chain to be successful. In this case, the initial exploit was targeting the Safari web browser; the user would visit the malicious website, there is specific JavaScript that runs on the website which triggers the vulnerability in Safari. Then this vulnerability is used to execute arbitrary code. In this case, it would use the limited access restrictions that Safari has available, which means that it is still running in the Sandbox (<a href="https://www.apple.com/lae/safari/">https://www.apple.com/lae/safari/</a>) and doesn’t run with full privileges. This is where the additional exploits come in that then lead to a Sandbox Escape and privilege escalation. Google has split this out into multiple blog posts and when into great detail on how these exploits work.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ai9JkGwikCj3r3KJyM40dA.jpeg" /></figure><p><a href="https://googleprojectzero.blogspot.com/2019/08/a-very-deep-dive-into-ios-exploit.html">https://googleprojectzero.blogspot.com/2019/08/a-very-deep-dive-into-ios-exploit.html</a></p><p><strong>Twitter CEO’s Account Hijacked</strong></p><p>From one targeted attack to another, more specific (and obvious), attack. An exploit that is becoming more and more common is known as the “Sim Card Swapping Exploit” this is more of a Social Engineering exploit than a technical one. It occurs when an attack is able to add a second phone to a user’s mobile phone accounts. If you use your phone for Multi-Factor Authentication via SMS, then the attacker will also receive these SMS messages and will be able to authenticate. This exploit occurred last week to Twitter’s CEO Jack Dorsey where his Twitter account was compromised using this technique. Twitter has relied on SMS and Caller ID for authentication since the service started, so it instantly becomes an obvious target for this attack. Unfortunately, there are no great solutions for this as of now. Many phone companies will allow you to set up PINs for your phone account to help deter this attack, however, this isn’t hugely helpful as the attack generally finds ways to convince employees to bypass this mechanism.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/898/1*fFEFGFAdvqRd3BvodvXSLQ.png" /></figure><p><a href="https://twitter.com/TwitterComms/status/1167528672523210752">https://twitter.com/TwitterComms/status/1167528672523210752</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=831721881823" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jakes-security-blog/ios-exploits-twitter-ceo-hacked-831721881823">iOS Exploits | Twitter CEO Hacked</a> was originally published in <a href="https://medium.com/jakes-security-blog">Jake’s Security Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Malware | Home Automation | Coinminer Botnet]]></title>
            <link>https://medium.com/jakes-security-blog/friday-30-08-48f3080b8029?source=rss-af78b4da01cb------2</link>
            <guid isPermaLink="false">https://medium.com/p/48f3080b8029</guid>
            <category><![CDATA[monero]]></category>
            <category><![CDATA[home-automation]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[malware]]></category>
            <dc:creator><![CDATA[Jake Williamson]]></dc:creator>
            <pubDate>Fri, 30 Aug 2019 21:20:39 GMT</pubDate>
            <atom:updated>2019-09-10T08:52:12.712Z</atom:updated>
            <content:encoded><![CDATA[<h4>30/08</h4><h3>Malware Compiling On-Premise</h3><p>Typically Windows environments don’t come with compilers by default, like on many Unix environments. As a result, malware often arrives precompiled as an executable. Xavier Mertens (<a href="https://xavier.mertens.consulting/">https://xavier.mertens.consulting</a>) found two exceptions to this behaviour. <a href="https://www.virustotal.com/gui/file/e5d58197b5d4465fe778bae8a63c5ab721a8c15b0f3c5e2aa6d20cbad3473b3e/detection">One</a> relies on jsc.exe, a JScript compiler that is part of the .NET Framework. Most Windows systems come with the .NET Framework installed and, as such, have jsc.exe installed. <a href="https://www.virustotal.com/gui/file/452722bf48499e772731e20d255ba2e634bba88347abcfb70a3b4ca4acaaa53d/detection">The other</a> relies on msbuild.exe, which is also part of the .NET Framework, that allows you to automatically build applications similar to the MAKE command in Unix. Whilst this doesn’t happen often, Xavier found two malware samples that took advantage of these tools. Most likely the purpose here is to obfuscate the attack and to evade antivirus scanners as they usually only look for the executables and not the source code, and, in this case, the code was delivered Base64 encoded. This is perhaps to make it easier to transfer, but could also be used to further obfuscate it.</p><p>Whilst jsc.exe and msbuild.exe are on most Windows machines because they all have the .NET Framework installed, they are not usually used so these two indicators that can be used to detect possible malicious behaviour if these executables are running on your system.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Uxzl6xsQAlmlqorAV3Xnjg.png" /><figcaption>Malware Searching for jsc.exe</figcaption></figure><blockquote><a href="https://isc.sans.edu/forums/diary/Malware+Samples+Compiling+Their+Next+Stage+on+Premise/25278/">https://isc.sans.edu/forums/diary/Malware+Samples+Compiling+Their+Next+Stage+on+Premise/25278/</a></blockquote><h3>Vulnerable Home Automation</h3><p>There are typically two ways to control a home automation system whilst away from the house, the first way relies on a cloud setup provided by the manufacturer where the system connects to the manufacturers APIs, then the mobile application connects to the same APIs. Whilst this relies on the manufacturer to take a strong approach with securing their APIs it usually beats the other option; opening up your firewall and directly exposing your home automation system to the outside world. Back in February, the German information security office, CERT-Bund, went and took a look at home automation systems exposed in Germany. They found 6000 HomeMatic systems that were exposed and not properly configured meaning that anybody could access these systems. As a result, they started a movement to notify these users of their exposed systems. After 6 months the result of this movement has come back, they were able to reach and correct about 3000 of these systems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oi8LF8gXsESNjrMPiCYSVA.png" /></figure><blockquote><a href="https://www.heise.de/security/meldung/CERT-Bund-warnt-vor-offenen-Smarthome-Systemen-4509977.html">https://www.heise.de/security/meldung/CERT-Bund-warnt-vor-offenen-Smarthome-Systemen-4509977.html</a></blockquote><h3>Coinminer Botnet Shutdown</h3><p>A collaboration between Avast and the French National Police targeted a Monero Bitcoin mining bot. A large part of the command and control infrastructure was located in France, the police went ahead and took down these servers and analysed the content of these servers, as would be expected. They then took it a step further, after getting approval from prosecutors, by replacing the infrastructure with servers that would remove the malware from infected systems. This isn’t usually done as part of these takedowns as there is always the fear that removing the software from the systems would cause damage. Also interesting is that most of the systems infected were outside of France and located in South America and Mexico.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*cZX9YAurudbhZZ10.jpg" /></figure><blockquote><a href="https://decoded.avast.io/janvojtesek/putting-an-end-to-retadup-a-malicious-worm-that-infected-hundreds-of-thousands/">https://decoded.avast.io/janvojtesek/putting-an-end-to-retadup-a-malicious-worm-that-infected-hundreds-of-thousands/</a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=48f3080b8029" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jakes-security-blog/friday-30-08-48f3080b8029">Malware | Home Automation | Coinminer Botnet</a> was originally published in <a href="https://medium.com/jakes-security-blog">Jake’s Security Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>