Virustotal update brings network traffic analysis support
It was not really clear in which direction Virustotal was heading after the acquisition of the service by Google was officially announced. Some feared that it would be integrated into core Google services and shut down, while others had hopes that the service would benefit from the parent company's vast resources.
The service improved afterwards, raising the maximum file size limit to 64 Megabytes in the process and eliminating nearly all of the wait time that users of the service experienced before the acquisition.
Virustotal announced the addition of a new feature two days ago on the official company blog. The service supports the analysis of so-called PCAP data now. PCAP - PAcket CApture - files contain captured network traffic. One use in this regard is the capturing of network traffic during software installations or while software is running that you want to analyze to find out if unauthorized connections are being made, and if this is the case, to which servers. Previously, you were able to scan the file on Virustotal, but that did not necessarily tell you anything about the connections that it made while it was running.
Here is one suggestion on how to make use of the new feature:
- Get a sandboxing application like Sandboxie to run software or files that you want to analyze in the sandbox.
- Use a network monitor like Wireshark to record the network traffic and save it as a PCAP file afterwards.
- Upload the PCAP file to Virustotal to have it analyzed.
Virustotal will scan the file with all scanners as usual, but use the intrusion detection systems Snort and Suricata afterwards to analyze the traffic. It performs a couple of operations that include:
- Extracts file metadata.
- Lists DNS resolutions.
- Lists HTTP communication.
- Extracts files that it recognizes in the network traffic and links to Virustotal reports.
The analysis of network traffic opens up additional possibilities in regards to Virustotal and the service that it makes available. It can be used for other purposes besides monitoring traffic of a sandboxed application. This may include logging the network traffic of a system on boot and shortly thereafter or recording browser exploitation traces.
The feature is a welcome addition to the Virustotal arsenal even though it may be used almost exclusively by security researchers. (via)
Advertisement
Thanks Martin for the warning and tips on how to prevent possible headaches.
Best regards
Thanks for the good tips.
I check my spam email to make sure their is no legitimate email and often find fake emails. So, when I hover the mouse over a link it goes to a different URL than the text.
If you are not sure of an email, type the URL in the web browser, don’t click the link in the email
Typing in the url manually is a good tip provided that you go to a legit site and not a fake one. So, if it shows paypal.com but the link is pointing to somesite.example.com, typing in paypal.com in your browser will do the trick. Then again, there is no need to do that if you have identified the email as fake.
Davin Peterson wrote, “… If you are not sure of an email, type the URL in the web browser, don’t click the link in the email”.
Rather than typing in the URL, why not simply right-click on the hyperlink in an email, choose to copy the actual or real link, paste it in the address bar of the Web browser that’s open, and then the user will immediately see what the real URL is? Pressing Return shouldn’t be done until having carefully examined the URL pasted in the address bar.
I wouldn’t bother doing that, either, for I simply hover the mouse pointer over hyperlinks in emails using Thunderbird and this permits being able to immediately see what the hyperlink really links to. But, copying and pasting would often be quicker than retyping whatever the real link is.
I begin by checking if the domain names are the same. If they aren’t and a reason isn’t provided in the email, then I surely won’t bother using the real url. Otherwise, I’ll check the rest of the real address shown when hovering the mouse pointer over the hyperlink. If there happened to be a redirection included in the full link, then we have a number of options:
1) We can just ignore the link altogether.
2) Copy the full link and paste it in a CTRL+D prompt for making bookmarks, but without saving the link as a bookmark; permitting us to scroll across the full length of the link to be able to carefully examine the whole thing.
3) If options 1 and 2 aren’t convenient, then we can alternatively paste the real url in the address bar for a new Web browser tab, f.e.
4) And if options 1, 2 and 3 aren’t convenient, due to the length of a url (being too long to be able to see the whole thing without needing to scroll or cursor over a lot of it, say), then we can paste the url in a temporary text file that provides wrapping of long lines.
When we don’t choose option 1, then we can view the whole url and then decide whether we want to use it, or not.
Retyping URLs is fine, if they’re not too long, but they often are. It would often be quicker to use one of the above 4 approaches, and maybe there’re other safe options that I’m not thinking or aware of. But retyping URLs isn’t going to be something I’d do, except for short ones; and I don’t mean links that’re just shorteners for the real URLs, for I won’t use these shorteners. It’s an interesting concept, but it’s unfortunately untrustworthy. Imo, no one should use services like TinyURL and likes. Nice little concept, but users need to know where they’re going, before going to wherever it is. Internauters, everyone for that matter, have an inherent right to know in what direction others try to lead them to.
A good middle-aged Catholic priest I momentarily met during the 1980s, and yes, there’re some good RCC clergy, it happens, he told me the first evening of a weekend retreat, a little secret, or something the overall RCC at least appears to want to keep secret. He said that I had two bosses. First, there is God. And, next in line, is …, well, guess? Yourself! In another manner of speaking, no one should “mess†with the inherent rights of another. And I detest services like TinyURL, et cetera, being used. If it’s done at websites that I know to be reliably/trustworthily managed, then I don’t mind the presence of these links, though still don’t like the idea of using them, so I do so only on very rare occasion. Having plenty of security for my PC, I can, on rare occasion, make use of such links; but, I always detest that they’re used. I’ve only used such links maybe twice over the past several years and it was only at websites that I know to be very trustworthy.
Thankfully, there’re add-ons for Firefox and these extensions will show the real URLs of short-cut, say, links; but, I don’t know if there’s such an add-on for Thunderbird and I won’t use Microsoft Outlook, Live, and other email services. I use other Web browsers, first Opera and then MSIE, but only when I’m trying to troubleshoot a problem. Otherwise, it’s always Firefox and Thunderbird. My main Firefox profile (having several of them in one Windows user account and using the -P option to select which profile to open with Firefox, and similar for Thunderbird) has plenty of security. GRC.com’s ShieldsUp! tests this morning reported that all of my ports, the first 1,085 or so, are all stealth-protected and I have basic Windows XP Firewall, Avira Antivir, Threatfire (from PC Tools) and WinPatrol always running. So I’m not worried; but, I still detest the usage of TinyURL and such services. It’s of no service to me, for what I want to see is the real url.
Preferable to using such services is to just use a short name or title for a hyperlink, but a meaningful title; preferably. If the title attracts a reader’s attention, then the person can just hover their mouse over the hyperlink’s text in order to see what the real link is. With services like TinyURL, et cetera, the real links are completely masked, unless we use, f.e., Firefox with the Long URL Please extension, or the Long URL Please Mod one.
Quote: “It is important to verify links before you click on them.”
Definitely, but as aware as I am about this, I sometimes forget and just blindly trust the text shown for a link. There haven’t been any nasty surprises for me, not yet, but it could happen. People should try to never forget to check what links lead to, before using the links.
That’s when I used Firefox or another Web browser. With email, which I use Thunderbird for, I never open links directly from the bodies of the emails. Instead, I always use right-click on a link, select the copy option, and then paste the URL in the Web browser’s address bar or field. That guarantees always seeing what the URL is, before pressing the Return key, so it’s less likely that I’ll get “skunked”, say, when it’s a link from an email. It’s links in Web browsers that I need to try to never forget to properly check before opening whatever the corresponding pages are. Using Firefox over 99% of the time and Having Long URL Please Mod installed for Firefox extension is often useful, but I’ve recently found that this add-on didn’t seem to be working for some shortened URLs. I think that that’s now corrected, but it’ll take a while to be able to see if it is, or not.
It may’ve possibly been for shortening links that the add-on wasn’t yet programmed for, but I’m pretty sure that this wasn’t the case. And some other things were no longer working correctly in Firefox over the past couple or few weeks. It all seems to now be corrected, after I shut down and restarted Firefox yesterday. And I just updated Firefox to v16.0.1 this morning.
Your example using google.com (for hyperlink text) and bing.com for the actual URL is a good one, and I wonder if I would’ve checked what the real URL is, if I wasn’t reading the text of your tutorial or advice. By far most websites I use are very reliable, so I can easily navigate through them without double-checking what hyperlinks actually link to, and this can be a problem, for then we can easily come to develop the habit of opening links before hovering the mouse point over them in order to see what the real linked pages are. And habits are sometimes a little difficult to cease, until we “burn” ourselves one or two times; if not three.
I think that I’m very safe, anyway. For the past couple of hours, I’ve been running GRC.com ShieldsUp! tests and the very initial page finds a “machine name” for my PC or connection, though also says that this may or may not be permanent, so I’ll have to check again after rebooting the PC, eventually. Other GRC ShieldsUp! tests say that all of my ports are stealthed though. Apparently, people from the WWW or WAN world can get access to my PC and maybe that’s because a router was added last week. I have only one PC connected to the Internet, but learned last year that passing through a router, rather than directly through a hi-speed modem, provides a considerable amount of security. So, and having learned that from both experienced users as well as college teachers in computer network administration, I definitely wanted a router for my personal connection.
Nonetheless, we should always double-check what hyperlinks actually link to, as a matter of personal Internet usage policy, say. What would be possibly good to have is an add-on that would automatically run a verification of every URL users try to open; but, that would surely not be fail-proof. How could a developer do that in a fully fail-proof way? And there’d surely be a fairly serious performance hit/impact, I imagine anyway.
Correction:
Wherein, in the 2nd-to-last paragraph, I wrote, “Apparently, people from the WWW or WAN world can get access to my PC and maybe that’s because a router was added last week.”, there’s a slight typo. error.
It should be, “Apparently, people from the WWW or WAN world can’t get access to my PC …”.
There’re some other typo. errors in my prior post, but I think readers will quickly realize what the correct text would be, so I won’t bother explaining these errors.
What about server side redirects? Is there a way to know the actual destination your browser will land on before you click a link, when a server has been set up to redirect traffic straight away?
It may be that the displayed link and the “a href” tag are both giving one url, but if you click it it instantly redirects you somewhere else because the server is set up to do this. (This just recently happened to me).
Is the redirection detectable before you follow a link, so you don’t end up somewhere you wouldn’t want to be?