We could not leave for the week-end without issuing a minor release or two so here we go.
Starting from TheHive 3.0.1, an administrator has the ability to configure Cortex job polling by defining the time between two polls thanks to the cortex.refreshDelay parameter as well as the number of consecutive failures before giving up (via cortex.MaxRetryOnError). However, these settings prevent the service from starting correctly. TheHive 3.1.2 corrects this issue.
When running a job in Cortex with the exact same details, the function findSimilarJob is called. It should return results from any previous jobs, but in the latest versions (2.1.0, 2.1.1) it does not because of a change that went past our QA.
In a similar fashion, the GUI search function was broken. Cortex 2.1.2 fixes both issues.
Excuse my French but I Need Help
Keep calm. We speak French. So if you encounter any difficulty to update TheHive or Cortex, please join our user forum, contact us on Gitter, or send us an email at email@example.com. We are always ready to help as does our user community.
TheHive Project’s Master Cooks are happy to announce the immediate availability of TheHive 3.1.0. This is the first release of your favourite SIRP (Security Incident Response Platform) or, if you fancy new buzzwords, SOAR (Security Orchestration, Automation & Response) that we put out as a release candidate to give sufficient time for our ever growing user community to test it and report any outstanding bug before publishing a stable version.
Indeed, TheHive 3.1.0 brings significant new functionalities that we detailed in previous blog posts. One of the most prominent features of this new major version is the support of responders through Cortex 2.1, also released today as a stable version.
Responders are similar to analyzers but instead of analyzing stuff, they allow you to respond to stuff. Put otherwise, they give you the ability to implement specific actions by a simple click from different elements in TheHive: alerts, cases, tasks, task logs and observables.
For instance, imagine a user in your constituency reporting a suspicious email. Using Synapse or an alternative alert feeder, the email reported by the user will automatically show up as an alert in your alert pane. Before starting working on it as a case, you preview it only to realise it is a scam and it does not warrant your time & effort. Still, you’d like to reply to the user.
In such a case, you could implement a responder that will not only send an email back to the user asking them to ignore such a scam but that can mark the alert as read. Using. A. Simple. Click. C’est beau n’est-ce pas ?
Going through all 71 (yes, 71) issues that have been closed with this release and the 3 RCs we published since July 31, 2018 will be terribly boring but you can read the full changelog while dipping your croissant in your espresso cup.
We’d rather encourage you to install this new version, which is as usual, AI-free, machine learning free, cyberbullshit-free, gluten-free, organic (well as much as free, open source software can be anyway), vegan (if you can eat it), and most importantly made with huge love and care for the SOC, CSIRT & CERT communities and other fellow cybercrime fighters. So go ahead and try it out. It won’t cost you a dime (or a franc if you are a French old timer).
Paris, France. The Sun is shining on the city of lights and temperatures are quite high, even for the summer season. Life is good. As a matter of fact, life is excellent.
TheHive Master Cooks are about to go on vacation for a few weeks. But before they pack up their Patagonia bags and leave the sandy beaches for those who enjoy them, preferring mountains, trails, walking and breathing fresh air with family and friends, they would like to make a significant contribution to help TheHive and Cortex users fight cyberattacks even better than they already do. And maybe convince those who don’t that free, open source software is not a joke or a geek fad.
We’d like to welcome to the stage our latest babies, which we are really proud of: TheHive 3.1 and Cortex 2.1, the new versions of the power duo which make digital forensics, incident response and, to an extent, cyber threat intelligence, better, faster, happier, regular exercising (well you know the Radiohead song so we’ll let you continue singing along) since early 2017.
While our project might seem very young, it is not. We’ve been working steadily on TheHive, using it (i.e. eating our own dog food) since early 2014 before releasing it at the end of 2016 once we were satisfied with it, as a token of gratitude to a community that helped us due our jobs in various ways. We then extracted what has become Cortex from its core to ship it as a separate product in February 2017. And we kept improving them at a steady piece for the collective benefit of incident responders, forensicators and threat analysts. And adoption has been rather spectacular. Thanks to all of our users for their love and support!
We believe we are at a moment where people could not brush us off anymore as amateurs. Try TheHive and Cortex, preferably with MISP and get a taste of what professional, free and open source software can be.
TheHive 3.1 and Cortex 2.1 are feature-packed and we won’t be able to cover them all in detail in a single blog post. Rather, we’d like to concentrate on a few important ones.
Stable, Pre-release Channels and New Repositories
We’d like to point out that, in order to improve our release process and given the number of features that we added, TheHive 3.1 and Cortex 2.1 are release candidates at this stage. So we encourage you to test them and report back any bugs or issues you encounter so we can address them and make the final releases as rock-solid as possible.
We have now two release channels: a stable one that should be used for production systems and a pre-release channel that should be used to try the release candidates such as TheHive 3.1-RC1 and Cortex 2.1-RC1, the subjects of this blog post, and help us iron out bugs before adding them to the stable channel. Those who love living on the bleeding edge may be tempted by running the release candidates on their production environment given all the candy and icing we added. They are at liberty of doing so but we don’t want to hear anyone one whining about an RC that broke everything and asking for their money back grin.
In previous releases of TheHive, whenever you configured a MISP instance, it was used to import events from and export cases to. Starting from TheHive 3.1.0, we added a purpose to the configuration file. By default, any added MISP instance will be used for import and export (ImportAndExport). However you can configure it to be used for importing events only (ImportOnly) or exporting cases only (ExportOnly).
When an analyst attempts to update a MISP event on which the account used by TheHive to connect to the MISP instance is not part of the original author’s organization, previous versions of TheHive will display a you do not have permission to do that error produced by MISP. Starting from TheHive 3.1, analysts have the ability to create a MISP extended event.
Case tasks can now be associated with task groups. For example, you could create groups called Identification and Malware Analysis, Containment and Communication and add tasks to them. Of course, this new feature can be used when designing case templates as well.
Import Observables from Analyzer Output
If analyzers produce a set of artifacts in their output (which is the case of several existing ones), TheHive will give you the ability to select those artifacts very easily and add them to your case as observables.
ZIP File Upload
Austin Haigh contributed an important feature which will allow analysts to directly import password-protected ZIP files into a case. The code uses the supplied password when adding the archive to extract its contents and add them one by one to the existing set of observables. This is highly practical when you want to add suspicious files without risking an accidental click which would compromise your endpoint or having to unzip archives containing such files first then add them one by one to TheHive.
Revamped Search Page
The search page has been completely revamped as shown in the screenshot below:
You can now select your search scope (cases, tasks, observables, alerts, analyser reports a.k.a. jobs or even the audit logs), apply filters and search TheHive without having to resort to complex, mind numbing Lucene syntax.
Responders and PAP
Last but not least, TheHive and Cortex offer you response capabilities (i.e. perform an action depending on the context) thanks to a new breed of programs called … wait for it … wait for it … responders. TADA!
Responders are very similar to analyzers. In fact we’ve taken the concept and extended it to apply to different elements in TheHive: alerts, cases, tasks, task logs, and observables of course.
You can reuse almost the same principles that apply to analyzers to write your own responders and if you are feeling generous, contribute them to the community. To give you a head start, we published a sample Mailer responder which, when customized for your environment, should allow you to send emails to inform your fellow analysts that a case has been created and that their help is required. Another example could be the ability to respond to a suspicious email report from a user, which is displayed as an alert, that they can safely ignore the corresponding email.
Like an analyzer, a responder can have two or more service interaction files (or flavors) to allow it to perform different actions. For example, a Mailer responder can send messages using several body templates.
Thanks to our long-time friend Andras Iklody from MISP Project who brought that to our attention, responders (and analyzers starting from Cortex 2.1) support PAP, the Permissible Actions Protocol.
Running into Troubles?
Something does not work as expected? You have troubles installing or upgrading? Spotted new bugs? No worries, please open issues on GitHub or comment on existing ones, join our user forum, contact us on Gitter, or send us an email at firstname.lastname@example.org. We are here to help.
When we’re not busy cooking new features, we go back to the trenches and face incidents like many of our fellow analysts who read our publications and use our tools. To do so, we swap our chef toques for firefighter helmets, not only because such shiny headwear is cool, but mainly because incident response (IR) is, at its very heart, firefighting (minus all the dangerous stuff).
If you think about it, when handling incidents you can see everything from cats in trees (spam) to major fire (APT). Thankfully, there are more cats to bring down than fire to extinguish. That being said, a big herd of cats could be a serious threat to your organization, to your mental health or both.
We tend to forget that incident handlers are humans, not robots. Unlike our metal cyberfriends, we need diversity. We can’t risk insanity like Charlie Chaplin in Modern Times if we can avoid it. Unfortunately IR can be highly repetitive, especially if you only have cats to deal with.
Some could say ‘Nah, this is minor, nothing critical here’ but at some point, an analyst brainwashed by the same tasks again and again will be led to fault. In the worst case scenario, one could see an alert and immediately categorize it as false positive without any further consideration. Because ‘this alert is always a false positive’, until the day it is not…
Automation, a Solution?
Intuitively, we look in the direction of automation in order to minimize what we call ‘zombie’ tasks: highly repetitive and brainless tasks that need to be done. We believe that doing so will allow incident handlers to focus on the analysis and not on the tedious side of IR. Ultimately, we hope it will keep analysts stimulated and in a state of alert. Also, it should reduce time and effort spent on the low-hanging fruits.
One of the most dreary tasks in our opinion is to record the context around an incident.
What is the problem? When did it happen? What’s the origin? Who are the victims? How many are there? Answers to these questions let you have an overview of what is happening and are valuable to correlate incidents. So it is worth taking some minutes to add this information to your case. Sadly, most of the time it will look like a succession of ‘Ctrl+C; Alt+Tab; Ctrl+V’ from your incident source to TheHive. Exactly the kind of tasks we want to forego.
Having identified the threat that apathetic analysts pose, the root cause (highly repetitive tasks) and a solution (automate the recording of incident context), the question of the implementation has been raised.
The first challenge to solve is the number of incident sources. Almost everything can trigger an incident: a firewall, an IDS, antivirus, SIEM, users, etc… So the application must be designed to accept several sources and must permit to easily integrate new ones. And instead of having to configure multiple alert feeders to supply alerts to TheHive, we would have only one. To some extent, it can be assimilated to a meta feeder.
And if the application works as intended, we still have a second challenge. Let’s say you, dear reader, and ourselves use the galaxy renowned Stargazer IDS. Maybe you’d like to include the full packet capture in the case but we wouldn’t. Using the same product doesn’t mean using it the same way. So we have a variety of sources and for each source, we have a variety of configurations and workflows. Hence any app we design needs to accept multiple configurations and workflows for any given source.
Finally: the third challenge. We want to make the most out of TheHive. Creating cases, creating alerts, assigning cases, adding logs, adding observables … all those actions are not an option.
After several trials and failures, we came up with Synapse. Basically it is a Python 3app which sits between TheHive and your incident sources:
To solve the first and third challenge, we rely on connectors. A connector is a Python object dedicated to interact with a security device. In the picture above, you can see the Exchange Connector and TheHive Connector. To extend the number of sources, you just have to develop the connector that corresponds to your device.
Regarding the second challenge, we rely on workflows. Workflows are python scripts who use connectors to automate repetitive tasks when tracking a case. Not happy with the current workflow? Develop your own using the connectors.
At this point, you probably wonder why there’s an API in the picture above. Well, the API is the link between the user and the workflows. By hitting a specific endpoint of the Synapse’s API, the corresponding workflow will be launched. That way the user can choose what to launch, especially if they are only interested in a particular workflow. Moreover, using an API allows us to listen to TheHive’s real-time stream and initiate some actions like closing a QRadar offense when the related case is solved.
At the moment, Synapse includes the Exchange connector and the associated Ews2Case workflow. The workflow features:
Case creation from emails
Adding email bodies to task logs
Adding email replies to the case
Adding email attachments as observables
And of course, everything is done to minimize the number of clicks! Check the workflow documentation to understand how it works under the hood.
We’re still working on the QRadar connector and the associated workflows but if you can’t wait, have a look at the work done by the community like pierrebarlet’s script.
Check it Out
As usual, Synapse is an open source and free software released under the AGPL (Affero General Public License).