Thanks to the UrsaDB database, queries on large datasets can be extremely fast.

How does it work?
YARA is pretty fast, but searching through large dataset for given signature can take a lot of time. To countermeasure this, we have implemented a custom database called UrsaDB. It is able to pre-filter the results, so it is only necessary to run YARA against a small fraction of binaries:


Installation (Docker)
Recommended way of installing things is to build from sources using docker-compose:

git clone --recurse-submodules https://github.com/CERT-Polska/mquery.git
 docker-compose up --scale daemon=3

where --scale daemon=... refers to the number of workers which will simultaneously process select/index jobs.
Hint: Your docker-compose must support v3 syntax of docker-compose.yml. Update your software if you have any problems.

Installation (manual)

1. Run ursadb database (see ursadb project for further instructions on that topic).
2. Install redis-server and python2.
3. Install requirements: pip install -r requirements.txt
4. Copy config.example.py to config.py, remember to adjust the settings and set unique SECRET_KEY.
5. Setup a flask application originating from webapp.py in your favourite web server.
6. Run daemon.py - a standalone script which should work constantly, consider putting it in systemd.

How to use this thing

1. Start up the whole system (see "Installation").
2. Web interface (by default) should be available on http://localhost:80/
3. Upload files to be indexed to the mquery_samples volume. From the host it should be visible at /var/lib/docker/volumes/mquery_samples/_data. If in doubt, debug using docker image inspect mquery_samples command.
4. Open web interface, choose "admin" tab and click "Index /mnt/samples".
5. While indexing, the current progress will be displayed in the "backend" section of "admin" tab (no auto refresh), ursadb will also periodically report something on the console.
6. After successful indexing, your files should be searchable. Go to the main tab and upload some Yara, e.g.:

rule emotet4_basic: trojan
 {
     meta:
         author = "psrok1/mak"
         module = "emotet"
     strings:
         $emotet4_rsa_public = { 8d ?? ?? 5? 8d ?? ?? 5? 6a 00 68 00 80 00 00 ff 35 [4] ff 35 [4] 6a 13 68 01 00 01 00 ff 15 [4] 85 }
         $emotet4_cnc_list = { 39 ?? ?5 [4] 0f 44 ?? (FF | A3)}
     condition:
         all of them
 }