Skip to content

Commit

Permalink
New documentation version for the new release
Browse files Browse the repository at this point in the history
  • Loading branch information
@richard-julien
richard-julien committed Aug 4, 2019
1 parent 35c8155 commit 7f5efe9
Show file tree
Hide file tree
Showing 7 changed files with 511 additions and 2 deletions.
4 changes: 2 additions & 2 deletions opencti-documentation/website/package.json
View file
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
{
"version": "1.1.1",
"scripts": {
"examples": "docusaurus-examples",
"start": "docusaurus-start",
Expand All @@ -10,6 +11,5 @@
},
"devDependencies": {
"docusaurus": "^1.9.0"
},
"version": "1.1.0"
}
}
84 changes: 84 additions & 0 deletions ...umentation/website/versioned_docs/version-1.1.1/getting-started/requirements.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,84 @@
---
id: version-1.1.1-requirements
title: Infrastructure requirements
sidebar_label: Infrastructure requirements
original_id: requirements
---

Since OpenCTI has some dependencies, you can find below the minimum configuration and amount of resources needed to launch the OpenCTI platform.

## Total requirements

The minimal hardware requirements for all components of the platforms including the databases are:

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|------------------------------------|
| 6 cores | 16GB | SSD (recommanded) / Normal | Depending of your content (> 32GB) |

## Databases

### Grakn

Grakn is composed of 2 JAVA processes, one for Grakn itself and the other one for Cassandra. Each process requires a minimum of 4GB of memory. So Grakn needs:

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|------------------------------------|
| 2 cores | 8GB | SSD | Depending of your content (> 16GB) |

> In order to setup the JAVA memory allocation, you can use the environment variable `SERVER_JAVAOPTS` and `STORAGE_JAVAOPTS`. You can find more information in the [official Grakn documentation](https://dev.grakn.ai/docs).
### ElasticSearch

ElasticSearch is also a JAVA process that needs a minimal amount of memory and CPUs.

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|------------------------------------|
| 2 cores | 1GB | Normal | Depending of your content (> 16GB) |

> In order to setup the JAVA memory allocation, you can use the environment variable `ES_JAVA_OPTS`. You can find more information in the [official ElasticSearch documenation](ttps://www.elastic.co/guide/en/elasticsearch/reference/current/docker.html).
### Redis

Redis has a very small footprint and only needs a tiny configuration:

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|-----------------------------------|
| 1 core | 128MB | Normal | 128MB |

> You can use the option `--maxmemory` to limit the usage. You can find more information in the [Redis docker hub](https://hub.docker.com/r/bitnami/redis/).
### RabbitMQ

RabbitMQ has a very small footprint until and can store messages directly on the disk if it does not have enough memory.

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|-----------------------------------|
| 1 core | 128MB | Normal | 128MB |

> The RabbitMQ memory configuration can be find in the [RabbitMQ official documentation](https://www.rabbitmq.com/memory.html).
### Total

The requirements for the databases infrastructure of OpenCTI are:

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|------------------------------------|
| 4 cores | 12GB | SSD (recommanded) / Normal | Depending of your content (> 32GB) |

## Application

### Platform

OpenCTI platform is based on a NodeJS runtime, with a memory limit of **512MB by default**.

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|-----------------------------------|
| 1 core | 512MB | Normal | 256MB |

### Workers and connectors

OpenCTI workers and connectors are Python processes with a very small footprint. For each connector, requirements are:

| CPU | RAM | Disk type | Disk space |
| ------------- |---------------| ---------------------------|-----------------------------------|
| 1 core | 128MB | Normal | 128MB |
97 changes: 97 additions & 0 deletions ...i-documentation/website/versioned_docs/version-1.1.1/installation/connectors.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,97 @@
---
id: version-1.1.1-connectors
title: Connectors activation
sidebar_label: Enable connectors
original_id: connectors
---

## Introduction

Connectors are standalone processes that are independant of the rest of the platform. They are using RabbitMQ to push data to OpenCTI, through a dedicated queue for each instance of connector. Depending on your deployment, you can enable connectors by using the connectors Docker images or launch them manually.

## Connector configurations

All connectors have 2 mandatory configuration parameters, the `name` and the `confidence_level`. The `name` is the name of the instance of the connector. For instance, for the MISP connector, you can launch as many MISP connectors as you need, if you need to pull data from multiple MISP instances.

> The `name` of each instance of connector must be unique.
> The `confidence_level` of the connector will be used to set the `confidence_level` of the relationships created by the connector. If a connector needs to create a relationship that already exists, it will check the current `confidence_level` and if it is lower than its own, it will update the relationship with the new information. If it is higher, it will do nothing and keep the existing relationship.
## Docker activation

You can either directly run the Docker image of connectors or add them to your current `docker-compose.yml` file.

### Add a connector to your deployment

For instance, to enable the MISP connector, you can add a new service to your `docker-compose.yml` file:

```
connector-misp:
image: opencti/connector-misp:{RELEASE_VERSION}
environment:
- RABBITMQ_HOSTNAME=localhost
- RABBITMQ_PORT=5672
- RABBITMQ_USERNAME=guest
- RABBITMQ_PASSWORD=guest
- MISP_NAME=MISP\ Circle
- MISP_CONFIDENCE_LEVEL=3
- MISP_URL=http://localhost
- MISP_KEY=ChangeMe
- MISP_TAG=OpenCTI:\ Import
- MISP_UNTAG_EVENT=true
- MISP_IMPORTED_TAG=OpenCTI:\ Imported
- MISP_INTERVAL=1 # Minutes
- MISP_LOG_LEVEL=info
restart: always
```

### Launch a standalone connector

To launch standalone connector, you can use the `docker-compose.yml` file of the connector itself. Just download the [release](https://github.com/OpenCTI-Platform/connectors/archive/{RELEASE_VERSION}.zip) and start the connector:

```
$ wget https://github.com/OpenCTI-Platform/connectors/archive/{RELEASE_VERSION}.zip
$ unzip {RELEASE_VERSION}.zip
$ cd connectors-{RELEASE_VERSION}/misp/
```

Change the configuration in the `docker-compose.yml` according to the parameters of the platform and of the targeted service. RabbitMQ credentials are the only parameters that the connector need to send data to OpenCTI. Then launch the connector:

```
$ docker-compose up
```

## Manual activation

If you want to manually launch connector, you just have to install Python 3 and pip3 for dependencies:

```
$ apt install python3 python3-pip
```

Download the [release](https://github.com/OpenCTI-Platform/connectors/archive/{RELEASE_VERSION}.zip) of the connectors:

```
$ wget https://github.com/OpenCTI-Platform/connectors/archive/{RELEASE_VERSION}.zip
$ unzip {RELEASE_VERSION}.zip
$ cd connectors-{RELEASE_VERSION}/misp/src/
```

Install dependencies and initialize the configuration:

```
$ pip3 install -r requirements.txt
$ cp config.yml.sample config.yml
```

Change the `config.yml` content according to the parameters of the platform and of the targeted service and launch the connector:

```
$ python3 misp.py
```

## Connectors status

The connector status can be displayed in the dedicated section. You will be able to see the statistics of the RabbitMQ queue of the connector:

![Connectors status](assets/installation/connectors_status.png "Connectors status")
114 changes: 114 additions & 0 deletions opencti-documentation/website/versioned_docs/version-1.1.1/installation/docker.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
---
id: version-1.1.1-docker
title: Docker installation
sidebar_label: Using Docker
original_id: docker
---

OpenCTI could be deployed using the *docker-compose* command.

## Clone the repository

```bash
$ mkdir /path/to/your/app && cd /path/to/your/app
$ git clone https://github.com/OpenCTI-Platform/docker.git
$ cd doker
```

### Configure the environment

Before running the docker-compose command, please change the admin token (this token must be a [valid UUID](https://www.uuidgenerator.net/)) and password of the application in the file `docker-compose.yml`:

```yaml
- APP__ADMIN__PASSWORD=ChangeMe
- APP__ADMIN__TOKEN=ChangeMe
```
And change the variable `OPENCTI_TOKEN` (for `worker-import` and `worker-export`) according to the value of `APP__ADMIN__TOKEN`

```yaml
- OPENCTI_TOKEN=ChangeMe
```

As OpenCTI has a dependency to ElasticSearch, you have to set the `vm.max_map_count` before running the containers, as mentioned in the [ElasticSearch documentation](https://www.elastic.co/guide/en/elasticsearch/reference/current/docker.html#docker-cli-run-prod-mode).

```bash
$ sysctl -w vm.max_map_count=262144
```

## Run

In order to have the best experience with Docker, we recommend to use the Docker stack feature. In this mode we will have the capacity to easily scale your deployment.

### In Swarm or Kubernetes

```bash
$ docker stack deploy -c docker-compose.yml opencti
```

### In standard Docker
```bash
$ docker-compose --compatibility up
```

You can now go to http://localhost:8080 and log in with the credentials configured in your environment variables.

## Data persistence

If you wish your OpenCTI data to be persistent in production, you should be aware of the `volumes` section for both `Grakn` and `ElasticSearch` services in the `docker-compose.yml`.

Here is an example of volumes configuration:

```yaml
volumes:
grakndata:
driver: local
driver_opts:
o: bind
type: none
esdata:
driver: local
driver_opts:
o: bind
type: none
```

## Memory configuration

OpenCTI default `docker-compose.yml` file does not provide any specific memory configuration. But if you want to adapt some dependencies configuration, you can find some links below.

### OpenCTI - Platform

OpenCTI platform is based on a NodeJS runtime, with a memory limit of **512MB by default**. We do not provide any option to change this limit today. If you encounter any `OutOfMemory` exception, please open a [Github issue](https://github.com/OpenCTI-Platform/opencti/issues/new?assignees=&labels=&template=bug_report.md&title=).

### OpenCTI - Workers and connectors

OpenCTI workers and connectors are Python processes. If you want to limit the memory of the process we recommend to directly use Docker to do that. You can find more information in the [official Docker documentation](https://docs.docker.com/compose/compose-file/).

> If you do not use Docker stack, think about `--compatibility` option.

### Grakn

Grakn is a JAVA process that rely on Cassandra (also a JAVA process). In order to setup the JAVA memory allocation, you can use the environment variable `SERVER_JAVAOPTS` and `STORAGE_JAVAOPTS`.

> The current recommendation is `-Xms4G` for both options.

You can find more information in the [official Grakn documentation](https://dev.grakn.ai/docs).

### ElasticSearch

ElasticSearch is also a JAVA process. In order to setup the JAVA memory allocation, you can use the environment variable `ES_JAVA_OPTS`.

> The minimal recommended option today is `-Xms512M -Xmx512M`.

You can find more information in the [official ElasticSearch documentation](https://www.elastic.co/guide/en/elasticsearch/reference/current/docker.html).

### Redis

Redis has a very small footprint and only provides an option to limit the maximum amount of memory that can be used by the process. You can use the option `--maxmemory` to limit the usage.

You can find more information in the [Redis docker hub](https://hub.docker.com/r/bitnami/redis/).

### RabbitMQ

The RabbitMQ memory configuration can be find in the [RabbitMQ official documentation](https://www.rabbitmq.com/memory.html). Basically RabbitMQ will consumed memory until a specific threshold. So it should be configure along with the Docker memory limitation.
Loading

0 comments on commit 7f5efe9

Please sign in to comment.