<?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:media="http://search.yahoo.com/mrss/"><channel><title><![CDATA[mongo - Shodan Blog]]></title><description><![CDATA[The latest news and developments for Shodan.]]></description><link>https://blog.shodan.io/</link><generator>Ghost 0.7</generator><lastBuildDate>Thu, 09 Apr 2026 17:55:49 GMT</lastBuildDate><atom:link href="https://blog.shodan.io/tag/mongo/rss/" rel="self" type="application/rss+xml"/><ttl>60</ttl><item><title><![CDATA[Memory As A Service]]></title><description><![CDATA[<p>I've written and presented on the topic of insecure databases for nearly 2 years now. The example I use the most to demonstrate the problem is MongoDB because it's popular and had <a href="https://blog.shodan.io/its-still-the-data-stupid/">terrible defaults</a>. Invariably though the focus of the conversation ends up on MongoDB and not that there are</p>]]></description><link>https://blog.shodan.io/memory-as-a-service/</link><guid isPermaLink="false">7d5429f8-0bd1-4dfc-a025-4e0e32a69d8f</guid><category><![CDATA[research]]></category><category><![CDATA[mongo]]></category><category><![CDATA[Memcached]]></category><dc:creator><![CDATA[John Matherly]]></dc:creator><pubDate>Thu, 17 Dec 2015 07:13:47 GMT</pubDate><media:content url="http://blog.shodan.io/content/images/2015/12/artificial-engine_00229391.jpg" medium="image"/><content:encoded><![CDATA[<img src="http://blog.shodan.io/content/images/2015/12/artificial-engine_00229391.jpg" alt="Memory As A Service"><p>I've written and presented on the topic of insecure databases for nearly 2 years now. The example I use the most to demonstrate the problem is MongoDB because it's popular and had <a href="https://blog.shodan.io/its-still-the-data-stupid/">terrible defaults</a>. Invariably though the focus of the conversation ends up on MongoDB and not that there are hundreds of thousands of databases on the Internet without any authentication.</p>

<p>So for today I decided to take a look at something else: <a href="http://memcached.org/">Memcached</a>. Their website explains it best:</p>

<blockquote>
  <p>Memcached is an in-memory key-value store for small chunks of arbitrary data (strings, objects) from results of database calls, API calls, or page rendering.</p>
</blockquote>

<p>Do you operate a website? Does it get a lot of traffic? Then memcached is what you need to speed up response times by caching database lookups, web responses or anything else that takes more than a second to accomplish.</p>

<p>Shodan shows there are <a href="https://www.shodan.io/report/IsKj8RXU">more than 130,000 Memcached servers</a> running on the Internet. And they also return a lot of detailed information about their status:</p>

<p><img src="https://blog.shodan.io/content/images/2015/12/Firefox_Screenshot_2015-12-17T06-22-15-983Z.png" alt="Memory As A Service"></p>

<p>Memcached provides its uptime, version, current number of connections, how much is being stored and much more. For now, I just took a look at the amount of data stored and how much memory is made available. Aggregating all the information from the publicly-available Memcached instances here are some stats:</p>

<ul>
<li><strong>8 TB</strong> of data stored</li>
<li><strong>49,153 PB</strong> of memory collectively available</li>
</ul>

<p>Since Memcached is a caching layer we wouldn't expect to see a lot of data stored in it on a permanent basis (records also usually have an expiration attached). And it doesn't offer advanced querying as a regular database would, which makes navigating the 8 TB of data more difficult than with MongoDB. That being said, there is still a lot of sensitive information that is temporarily stored on these instances. However, there is also a ridiculously giant amount of memory available on public Memcached servers. For people not familiar with petabytes, the total amount of memory advertised is <strong>49,153,000 TB</strong>.</p>

<p>The organizations that are hosting the most instances are:</p>

<p><img src="https://blog.shodan.io/content/images/2015/12/Firefox_Screenshot_2015-12-17T06-35-04-260Z.png" alt="Memory As A Service"></p>

<ol>
<li><strong>ColoCrossing</strong>  </li>
<li><strong>GoDaddy</strong>  </li>
<li><strong>Enzu</strong>  </li>
<li><strong>Aliyun</strong>  </li>
<li><strong>Alibaba Advertising</strong></li>
</ol>

<p>One of the reason for all these publicly accessible instances is the same as with MongoDB: the official, default configuration of Memcached listens on all interfaces. Curiously, the Linux distributions I looked at that are offering Memcached packages provided secure defaults; i.e. only listen on <em>localhost</em>. This means that most likely the above organizations installed Memcached from source.</p>

<p>I hope this has provided some evidence that it's not just MongoDB facing insecure-by-default issues when it comes to data storage services. I could've performed the same analysis as above for <a href="https://www.shodan.io/search?query=product%3Aredis">Redis</a>, <a href="https://www.shodan.io/search?query=product%3Acassandra">Cassandra</a>, <a href="https://www.shodan.io/search?query=product%3Acouchdb">CouchDB</a> or <a href="https://www.shodan.io/search?query=port%3A8098+mochiweb">Riak</a>.</p>]]></content:encoded></item></channel></rss>