Fresh Installation

This section describes how to install FortiSIEM for the current release.

• Pre-Installation Checklist
• All-in-one Installation
• Cluster Installation

Pre-Installation Checklist

Before you begin, check the following:

• Ensure that your system can connect to the network. You will be asked to provide a DNS Server and a host that can be resolved by the DNS Server and can respond to a ping. The host can either be an internal host or a public domain host like google.com.

• Choose deployment type – Enterprise or Service Provider. The Service Provider deployment provides multi-tenancy.
• Determine whether FIPS should be enabled.
• Choose Install type:
  • All-in-one with FortiSIEM Manager
    
  • Cluster with Manager, Supervisor and Workers
    
  • All-in-one with Supervisor only, or
  • Cluster with Supervisor and Workers
• Choose the storage type for Supervisor, Worker, and/or Collector
  • Online storage - There are 4 choices
    • ClickHouse - Recommended for most deployments. Please see ClickHouse Reference Architecture for more information.
      

      If you plan to use ClickHouse cluster, the Worker nodes will be defined as Keeper, Data or Query nodes. The Supervisor and Worker nodes can operate as a Keeper, Data or Query nodes. This is discussed in the ClickHouse Reference Architecture, Supervisor/Worker Nodes Running ClickHouse Functions and Configuring ClickHouse Topology.

    • EventDB on local disk
    • EventDB on NFS
    • Elasticsearch
  • Archive storage – There are 2 choices
    • EventDB on NFS
    • HDFS
• Fortinet recommends that you do not choose AWS Spot instances for Supervisor and Worker nodes. Such instances can go down at any time with short notice, causing instability and performance issues.
• Determine hardware requirements and choose AWS instance type accordingly:

Node	vCPU	RAM	Local Disks
Manager	Minimum – 16 Recommended - 32	Minimum 24GB Recommended 32GB	OS – 25GB OPT – 200GB CMDB – 100GB SVN – 60GB
Supervisor (All in one)	Minimum – 12 Recommended - 32	Minimum without UEBA – 24GB with UEBA - 32GB Recommended without UEBA – 32GB with UEBA - 64GB	OS – 25GB OPT – 100GB CMDB – 60GB SVN – 60GB Local Event database – based on need
Supervisor (Cluster)	Minimum – 12 Recommended - 32	Minimum without UEBA – 24GB with UEBA - 32GB Recommended without UEBA – 32GB with UEBA - 64GB	OS – 25GB OPT – 100GB CMDB – 60GB SVN – 60GB
Workers	Minimum – 8 Recommended - 16	Minimum – 16GB Recommended – 24GB	OS – 25GB OPT – 100GB
Collector	Minimum – 4 Recommended – 8 ( based on load)	Minimum – 4GB Recommended – 8GB	OS – 25GB OPT – 100GB

• If your Online event database is external (e.g. EventDB on NFS or Elasticsearch), then you must configure external storage before proceeding to FortiSIEM deployment.

  • For NFS deployment, see here.

  • For Elasticsearch deployment, see here.

• If your Online event database is internal, that is, inside Supervisor or Worker nodes, then you need to determine the size of the disks based on your EPS (events per second) and event retention needs.

  • For EventDB on local disk, see here.

  • For ClickHouse, see here.

• For OPT - 100GB, the 100GB disk for /opt will consist of a single disk that will split into 2 partitions, /OPT and swap. The partitions will be created and managed by FortiSIEM when configFSM.sh runs.

All-in-one Installation

This is the simplest installation with a single Virtual Appliance. If storage is external, then you must configure external storage before proceeding with installation.

• Launch an instance using FortiSIEM 7.2.1 AMI

• Configure FortiSIEM
• Upload the FortiSIEM License
• Configure an Event Database
• Final Check

Launch an Instance Using FortiSIEM 7.2.1 AMI

1. Navigate to the EC2 AMIs page, find FortiSIEM 7.2.1 AMI (or in AWS Marketplace after the GA release), and select it.

2. Launch FortiSIEM-7.2.1.0241 by clicking Launch instance from AMI.
   

   
   

3. From the Launch an instance page, take the following actions.

   1. From the Name and tags section, in the Name field, enter the name of your instance.
      

      
      

   2. From the Instance type drop-down list, select an appropriate instance type.

   3. From the Key pair (login) drop-down list, select a key pair (or create a key pair to use).

   4. From Network settings, click Edit. Configure instance network settings such as VPC, Subnet, IP, and Security Group.
      

      Notes:

      • If creating a Security Group, you will need ssh and https to begin with. Depending on whether this node will receive syslog or other inbound data, you may need to open additional protocols/ports.

      • If you are planning to also assign global IPv6 address to your instance, then from the Auto-assign IPv6 IP drop-down list, select Enable.

      

   5. From the Configure storage pane, add additional disks by clicking Add new volume for each disk you wish to add, and entering the size and selecting the type (gp3 should be used for most cases). These will be used for the additional partitions in the virtual appliance. An All In One deployment requires the following additional partitions.
      

      
      Note: If you plan to onboard greater than 500 devices, or 5000 EPS, please consider increasing IOPS and Throughput for the disk used to mount /cmdb in FortiSIEM.
      

      For instance, you can run the following command once FortiSIEM is initially deployed to determine which disk mounts the cmdb folder.

      [admin@6 data-definition]$ lsblk | grep cmdb

      └─sdc1 8:33 0 60G 0 part /cmdb

      In this case /dev/sdc.

      You can go into EBS volumes in AWS, and increase the IOPS to 5000, and Throughput to 400MB/s to be more in line with SSD performance.

      Use these partition values:
      

      Volume Name	Size	Disk Name
      EBS Volume 2	100GB	/opt For OPT - 100GB, the 100GB disk for /opt will consist of a single disk that will split into 2 partitions, /OPT and swap. The partitions will be created and managed by FortiSIEM when configFSM.sh runs.
      EBS Volume 3	60GB	/svn
      EBS Volume 4	60GB	/cmdb
      EBS Volume 5	60GB+	/data (see the following note)

      Note on EBS Volume 5:
      

      • Add the 5th EBS Volume only if using EventDB on local storage or ClickHouse. In all other cases, this disk is not required. ClickHouse is recommended for most deployments. Please see ClickHouse Reference Architecture for more information.

      • For EventDB on local disk, choose a disk based on your EPS and event retention policy. See EventDB Sizing Guide for guidance. 60GB is the minimum.

      • For ClickHouse, choose disks based on the number of Tiers and disks on each Tier. These depend on your EPS and event retention policy. See ClickHouse Sizing Guide for guidance. For example, you can choose 1 large disk for Hot Tier. Or you can choose 2 Tiers - Hot Tier comprised of one or more SSD disks and Warm Tier comprised of one or more magnetic hard disks.

      • Choose GP3 volume type for all volumes (GP3 is better than GP2 at a slightly lower cost). For the CMDB partition, you can choose to modify your volume type and IOPS based on your system workload if you see the consistently high IOPS requirement in your deployment.

   6. From the Summary pane, click Launch instance.

      

4. After receiving a Success message, navigate back to the Instances page.
   

   
   

   Note: You can click Instances from the top breadcrumb path.
   

   
   

5. Check the status of your instance by examining the Status check column.

6. When your instance shows that all checks have passed, select the instance that you just created and click Connect.
   

   
   

7. From the Connect to instance page, at "Example:", copy the ssh line.
   

   
   

8. Now ssh to the instance you created, using the command you just copied, but replace root user with ec2-user before executing the ssh command.

   Once logged in, you can execute the sudo su - command to become root user.
   

   Note: If you are going to assign and use a global IPv6 address to your instance in addition to IPv4, then you will need to perform additional steps and reboot the instance. Take the following steps:

   1. Add the following lines to the file /etc/sysconfig/network-scripts/ifcfg-eth0.
      

      IPV6INIT=yes

      IPV6_FAILURE_FATAL=no

   2. Run the following command to add service dhcpv6-client to the firewall rules (DHCP v6 works differently than v4)
      

      firewall-cmd --zone=fortisiem --permanent --add-service=dhcpv6-client

   3. Reboot the VM.
      

Configure FortiSIEM

At no stage of the installation process is it required that users manually format the disks. FortiSIEM will provision the file system on disks as needed.

Follow these steps to configure FortiSIEM by using a simple GUI.

1. At the root command prompt, go to /usr/local/bin and enter configFSM.sh, for example:
   

   # sudo -s; cd /usr/local/bin;
   

   # configFSM.sh
   

2. In VM console, select 1 Set Timezone and then press Next.
   

   

3. Select your Location, and press Next.
   

   

4. Select your Continent, and press Next.
   

   
   

5. Select the Country and City for your timezone, and press Next.
   

   
   

6. If installing Supervisor, select 1 Supervisor, and press Next.
   If installing a Worker, select 2 Worker, and press Next.
   If installing a Collector, select 3 Collector, and press Next.
   If Installing FortiSIEM Manager, select 4 FortiSIEM Manager, and press Next.
   If Installing FortiSIEM Supervisor Follower, select 5 Supervisor Follower and press Next.
   Note: The appliance type cannot be changed once it is deployed, so ensure you have selected the correct option.

   
   

   Regardless of whether you select FortiSIEM Manager, Supervisor, Supervisor Follower, Worker, or Collector, you will see the same series of screens with only the header changed to reflect your target installation, unless noted otherwise. A dedicated ClickHouse Keeper uses a Worker, so first install a Worker and then in later steps configure the Worker as a ClickHouse Keeper.

7. Select the Network Interface you wish to use, and press Next.

   Note: If a bond interface is configured, it will appear in the Select Network Interface window.
   

   
   

8. If you want to enable FIPS, then choose 2 install_with_fips. Otherwise, choose 1 install_without_fips. You have the option of enabling FIPS (option 3) or disabling FIPS (option 4) later.
   Note: After Installation, a 5th option to change your network configuration (5 change_network_config) is available. This allows you to change your network settings and/or host name.
   

   

9. Determine whether your network supports IPv4-only, or Both IPv4 and IPv6 (Dual Stack). Choose 1 for IPv4-only, or choose 3 for Both IPv4 and IPv6. Press Next.

   Note: In AWS, do not choose option 2 (IPv-6 only), because you will end up with a status check failure. AWS infrastructure currently requires that the VM has an IPv4 address for its monitoring purpose.
   

   
   

10. First you will configure the IPv4 network by entering the following fields, then press Next.
    

    Option	Description
    IPv4 Address	The Manager/Supervisor/Worker/Collector's IPv4 address
    NetMask	The Manager/Supervisor/Worker/Collector's IPv4 subnet
    Gateway	IPv4 Network gateway address
    DNS1, DNS2	Addresses of the DNS servers

    

11. If you chose 1 in step 9, then you will need to skip to step 12. If you chose 3 in step 9, then you will also configure the IPv6 network by entering the following fields, then press Next.
    

    Option	Description
    IPv6 Address	The Manager/Supervisor/Worker/Collector's IPv6 address
    prefix (Netmask)	The Manager/Supervisor/Worker/Collector's IPv6 prefix
    Gateway ipv6	IPv6 Network gateway address
    DNS1 IPv6, DNS2 IPv6	Addresses of IPv6 DNS server 1 and 2

    

    Note: If you chose option 3 in step 9 for both IPv4 and IPv6, then even if you configure 2 DNS servers for IPv4 and IPv6, the system will only use the first DNS server from IPv4 and the first DNS server from IPv6 configuration.
    Note: In AWS dual stack networks, IPv4 DNS server(s) can resolve names to both IPv4 and IPv6. In such environments, if you do not have an IPv6 DNS server, then you can use public IPv6 DNS servers such as Google DNS.

12. Configure Hostname for the FortiSIEM Manager, Supervisor ,Worker, or Collector, then press Next.
    

    

13. Test network connectivity by entering a host name that can be resolved by your DNS Server (entered in the previous step) and can respond to a ping. The host can either be an internal host or a public domain host like google.com. Press Next.
    

    

14. The final configuration confirmation is displayed. Verify that the parameters are correct. If they are not, then press Back to return to previous dialog boxes to correct any errors. If everything is OK, then press Run.
    

    

    The options are described in the following table.

    Option	Description
    -r	The FortiSIEM component being configured
    -z	The time zone being configured
    -i	IPv4-formatted address
    -m	Address of the subnet mask
    -g	Address of the gateway server used
    --host	Host name
    -f	FQDN address: fully-qualified domain name
    -t	The IP type. The values can be either 4 (for IPv4) or 6 (for IPv6) or 64 (for both IPv4 and IPv6)
    --dns1, --dns2	Addresses of the DNS server 1 and DNS server 2.
    --i6	IPv6-formatted address
    --m6	IPv6 prefix
    --g6	IPv6 gateway
    -o	Installation option (install_without_fips, install_with_fips, enable_fips, disable_fips, change_network_config*) *Option only available after installation.
    --testpinghost	The URL used to test connectivity

15. It will take some time for this process to finish. When it is done, proceed to Upload the FortiSIEM License. If the VM fails, you can inspect the ansible.log file located at /usr/local/fresh-install/logs to try and identify the problem.

Upload the FortiSIEM License

Before proceeding, make sure that you have obtained valid FortiSIEM license from Forticare. For more information, see the Licensing Guide.

You will now be asked to input a license.

1. Open a Web browser and log in to the FortiSIEM UI. Please note that if you are logging into FortiSIEM with an IPv6 address, you should input https://[IPv6 address] on the browser tab.
2. The License Upload dialog box will open.

   

3. Click Browse and upload the license file.

   Make sure that the Hardware ID shown in the License Upload page matches the license.

4. For User ID and Password, choose any Full Admin credentials.

   For the first time installation, enter admin as the user and admin*1 as the password. You will then be asked to create a new password for GUI access.

5. For Supervisor, Worker, or Collector, choose License type as Enterprise or Service Provider. The following option will be available for first time installations. Once the database is configured, this option will not be available.
   For FortiSIEM Manager, License Type is not an available option, and will not appear. At this point, FortiSIEM Manager installation is complete. You will not be taken the Event Database Storage page, so you can skip Configure an Event Database.
   Note: The FortiSIEM Manager license allows a certain number of instances that can be registered to FortiSIEM Manager.

6. Proceed to Configure an Event Database.
   

Configure an Event Database

Choose the event database.

If the Event Database is one of the following options, additional disk configuration is required.

• ClickHouse: See Case 2 in Creating ClickHouse Online Storage.
  

  Recommended for most deployments. Please see ClickHouse Reference Architecture for more information.
  

• EventDB on Local Disk: See Case 2 in Creating EventDB Online Storage.

Final Check

FortiSIEM installation is complete. If the installation is successful, the VM will reboot automatically. Otherwise, the VM will stop at the failed task.

You can inspect the ansible.log file located at /usr/local/fresh-install/logs if you encounter any issues during FortiSIEM installation.

After installation completes, ensure that the phMonitor is up and running, for example:

# phstatus

For the Supervisor, Supervisor Follower, Worker and Collector, the response should be similar to the following.

For FortiSIEM Manager, the response should look similar to the following.

Cluster Installation

For larger installations, you can choose Worker nodes, Collector nodes, and external storage (NFS, ClickHouse or Elasticsearch).

• Install Supervisor
• Install Workers
• Register Workers
• Create ClickHouse Topology (Optional)
• Install Collectors
• Register Collectors
• Install Manager
• Register Instances to Manager

Install Supervisor

Follow the steps in All-in-one Installation, except with the following differences.

1. Event Database choices are EventDB on NFS, ClickHouse, or Elasticsearch.

2. If you choose EventDB on NFS

   1. EBS Volume 5 is not required (From Launch an instance using FortiSIEM 7.2.1 AMI Step 3e).

   2. You need to configure NFS after license upload.
      

      
      
      

3. If you choose ClickHouse

   1. You need to create disks during Launch an instance using FortiSIEM 7.2.1 AMI Step 3e based on the role of the Supervisor node in the ClickHouse cluster. See the ClickHouse Sizing Guide for details.

   2. You need to configure disks after license upload.
      

      

4. If you choose Elasticsearch, define Elasticsearch endpoints after license upload. See the Elasticsearch Sizing Guide for details.

   

Install Workers

Once the Supervisor is installed, take the same steps in All-in-one Installation to install a Worker with the following differences.

1. Choose appropriate CPU and memory for the Worker nodes based on Sizing guide.

2. Two hard disks for Operating Systems and FortiSIEM Application:

   • OS – 25GB

   • OPT – 100GB
     For OPT - 100GB, the 100GB disk for /opt will consist of a single disk that will split into 2 partitions, /OPT and swap. The partitions will be created and managed by FortiSIEM when configFSM.sh runs.
     

3. If you are running ClickHouse, then create additional data disks based on the role of the Worker in ClickHouse topology. If it is a Keeper node, then a smaller disk is needed. If it is a data node, then a bigger disk is needed based on your EPS and retention policy. See ClickHouse Sizing Guide for details.

Sizing Guide References:

• ClickHouse Sizing Guide

• EventDB Sizing Guide

• Elasticsearch Sizing Guide

Register Workers

Once the Worker is up and running, add the Worker to the Supervisor node.

1. Go to ADMIN > License > Nodes.
2. Select Worker from the Mode drop-down list and enter the following information:
   1. In the Host Name field, enter the Worker's host name.
   2. In the IP Address field, enter the Worker's IP address.

   3. If you are running ClickHouse, then select the number for Storage Tiers from the Storage Tiers drop-down list, and input disk paths for disks in each Tier in the Disk Path fields.
      

      For Disk Path, use one of the following CLI commands to find the disk names.

      fdisk -l

      or

      lsblk

      When using lsblk to find the disk name, please note that the path will be /dev/<disk>, for example, /dev/vdc.

   4. Click Test.
      

      

   5. If the test succeeds, then click Save.
3. See ADMIN > Health > Cloud Health to ensure that the Workers are up, healthy, and properly added to the system.
   
   

Create ClickHouse Topology (Optional)

If you are running ClickHouse, you need to configure ClickHouse topology by specifying which nodes belong to ClickHouse Keeper and Data Clusters. Follow the steps in Configuring ClickHouse Topology.

Install Collectors

Once Supervisor and Workers are installed, follow the same steps in All-in-one Install to install a Collector except in Edit FortiSIEM Hardware Settings, you need to only choose OS and OPT disks. The recommended settings for Collector node are:

• CPU = 4
• Memory = 8GB
• Two hard disks:
  • OS – 25GB
  • OPT – 100GB
    For OPT - 100GB, the 100GB disk for /opt will consist of a single disk that will split into 2 partitions, /OPT and swap. The partitions will be created and managed by FortiSIEM when configFSM.sh runs.
    

Register Collectors

Collectors can be deployed in Enterprise or Service Provider environments.

• Enterprise Deployments
• Service Provider Deployments

Enterprise Deployments

For Enterprise deployments, follow these steps.

1. Log in to Supervisor with 'Admin' privileges.
2. Go to ADMIN > Settings > System > Cluster Config.
   1. Under Event Upload Workers, enter the IP of the Worker node. If a Supervisor node is only used, then enter the IP of the Supervisor node. Multiple IP addresses can be entered on separate lines. In this case, the Collectors will load balance the upload of events to the listed Event Workers.
      Note: Rather than using IP addresses, a DNS name is recommended. The reasoning is, should the IP addressing change, it becomes a matter of updating the DNS rather than modifying the Event Worker IP addresses in FortiSIEM.
   2. Click OK.
3. Go to ADMIN > Settings > System > Event Worker and enter the IP of the Supervisor node. Click Save.
4. Go to ADMIN > Setup > Collectors and add a Collector by entering:
   1. Name – Collector Name
   2. Guaranteed EPS – this is the EPS that Collector will always be able to send. It could send more if there is excess EPS available.
   3. Start Time and End Time – set to Unlimited.
5. SSH to the Collector and run following script to register Collectors:

   phProvisionCollector --add <user> '<password>' <Super IP or Host> <Organization> <CollectorName>

   The password should be enclosed in single quotes to ensure that any non-alphanumeric characters are escaped.

   1. Set user and password using the admin user name and password for the Supervisor.
   2. Set Super IP or Host as the Supervisor's IP address.
   3. Set Organization. For Enterprise deployments, the default name is Super.
   4. Set CollectorName from Step 2a.

      The Collector will reboot during the Registration.

6. Go to ADMIN > Health > Collector Health for the status.

   

Service Provider Deployments

For Service Provider deployments, follow these steps.

1. Log in to Supervisor with 'Admin' privileges.
2. Go to ADMIN > Settings > System > Cluster Config.
   1. Under Event Upload Workers, enter the IP of the Worker node. If a Supervisor node is only used, then enter the IP of the Supervisor node. Multiple IP addresses can be entered on separate lines. In this case, the Collectors will load balance the upload of events to the listed Event Workers.
      Note: Rather than using IP addresses, a DNS name is recommended. The reasoning is, should the IP addressing change, it becomes a matter of updating the DNS rather than modifying the Event Worker IP addresses in FortiSIEM.
   2. Click OK.

      

3. Go to ADMIN > Setup > Organizations and click New to add an Organization.

   

4. Enter the Organization Name, Admin User, Admin Password, and Admin Email.
5. Under Collectors, click New.
6. Enter the Collector Name, Guaranteed EPS, Start Time, and End Time.

   The last two values could be set as Unlimited. Guaranteed EPS is the EPS that the Collector will always be able to send. It could send more if there is excess EPS available.

   

7. SSH to the Collector and run following script to register Collectors:

   phProvisionCollector --add <user> '<password>' <Super IP or Host> <Organization> <CollectorName>

   The password should be enclosed in single quotes to ensure that any non-alphanumeric characters are escaped.

   1. Set user and password using the admin user name and password for the Organization that the Collector is going to be registered to.
   2. Set Super IP or Host as the Supervisor's IP address.
   3. Set Organization as the name of an organization created on the Supervisor.
   4. Set CollectorName from Step 4a.

      

      The Collector will reboot during the Registration.

8. Go to ADMIN > Health > Collector Health and check the status.

   

Install Manager

Starting with release 6.5.0, you can install FortiSIEM Manager to monitor and manage multiple FortiSIEM instances. An instance includes a Supervisor and optionally, Workers and Collectors. The FortiSIEM Manager needs to be installed on a separate Virtual Machine and requires a separate license. FortiSIEM Supervisors must be on 6.5.0 or later versions.

Follow the steps in All-in-one Install to install Manager. After any Supervisor, Workers, and Collectors are installed, you add the Supervisor instance to Manager, then Register the instance to Manager. See Register Instances to Manager.

Register Instances to Manager

To register your Supervisor instance with Manager, you will need to do two things in the following order.

• First, add the instance to Manager

• Then register the instance itself to Manager

Note that Communication between FortiSIEM Manager and instances is via REST APIs over HTTP(S).

Adding Instance to Manager

You can add an instance to Manager by taking the following steps.
Note: Make sure to record the FortiSIEM Instance Name, Admin User and Admin Password, as this is needed when you register your instance.

1. Login to FortiSIEM Manager.

2. Navigate to ADMIN > Setup.

3. Click New.

4. In the FortiSIEM Instance field, enter the name of the Supervisor instance you wish to add.

5. In the Admin User field, enter the Account name you wish to use to access Manager.

6. In the Admin Password field, enter the Password that will be associated with the Admin User account.

7. In the Confirm Admin Password field, re-enter the Password.

8. (Optional) In the Description field, enter any information you wish to provide about the instance.

9. Click Save.
   
   

10. Repeat steps 1-9 to add any additional instances to Manager.
    Now, follow the instructions in Register the Instance Itself to Manager for each instance.
    

Register the Instance Itself to Manager

To register your instance with Manager, take the following steps.

1. From your FortiSIEM Supervisor/Instance, navigate to ADMIN > Setup > FortiSIEM Manager, and take the following steps.

   1. In the FortiSIEM Manager FQDN/IP field, enter the FortiSIEM Manager Fully Qualified Domain Name (FQDN) or IP address.

   2. If the Supervisor is under a Supervisor Cluster environment, in the FortiSIEM super cluster FQDN/IP field, enter the Supervisor Cluster Fully Qualified Domain Name (FQDN) or IP address.

   3. In the FortiSIEM Instance Name field, enter the instance name used when adding the instance to Manager.

   4. In the Account field, enter the Admin User name used when adding the instance to Manager.

   5. In the Password field, enter your password to be associated with the Admin User name.

   6. In the Confirm Password field, re-enter your password.

   7. Click Test to verify the configuration.

   8. Click Register.
      A dialog box displaying "Registered successfully" should appear if everything is valid.
      

   9. Login to Manager, and navigate to any one of the following pages to verify registration.

      • ADMIN > Setup and check that the box is marked in the Registered column for your instance.

      • ADMIN > Health, look for your instance under FortiSIEM Instances.

      • ADMIN > License, look for your instance under FortiSIEM Instances.

Install Log

The install ansible log file is located here: /usr/local/fresh-install/logs/ansible.log.

Errors can be found at the end of the file.