Pretty stoked on this one guys!
Firmware upgrading and downgrading sounds pretty simple, anyone can do it, right? The mark of a professional is not that they can do something correctly, or even do it correctly over and over again. A professional works in such a way that, if anything goes wrong they are prepared and able to quickly get things back to normal. Firmware updates can go wrong just like anything else. So a real professional does things in a way that minimizes their risk and follows some best practices, as listed below.
Consider the following five points when performing firmware upgrades, not only in FortiOS but in general. This applies to pretty much any change you have to do in a production environment.
Before attempting any changes in production, first make sure you set up a laboratory where you can freely play with the new features, and understand them with enough time and no pressure. Read the Release Notes, Manuals, and other documentation like presentations, videos, or podcasts about the new version.
You are ready to explain the need for an upgrade once you understand:
l The differences and the enhancements between the new version and the previous version(s). l The impact of the upgrade on customers and the users of the operating platform. l The known limitations that might affect your environment. l The potential risks when performing the upgrade. l The licensing changes that may apply.
The reason can NOT be “Because I want to have the latest version”. The reason has to be explained in terms of business, technical, and/or operational improvement.
Affirmative answers to the following questions are valid reasons to upgrade:
If the best reason to upgrade is “Because the new features seem to be cool” or “Because I want to have the latest version”, a little more understanding and planning may be necessary.
If you choose to upgrade because you found a valid reason to do so, make sure you create a plan that covers business, technical, and operational aspects of the upgrade:
Business:
Proper planning and justification for an upgrade should be proportional to how critical the system is to the business.
The approval must be done in a formal (written or e-mail) form.
Technical and operational:
Execution of an upgrade is just as key as planning.
Once you are performing the upgrade, the pressure will rise and stress might peak. This is why you should stick to the plan you created with a cool head.
Resist the temptation to take decisions while performing the upgrade, as your judgment will be clouded by the stress of the moment, even if a new decision seems to be “obvious” at such time. If your plan says you should rollback, then execute the rollback despite the potential “We-can-fix-this-very-quickly” mentality.
While performing the upgrade, make sure all the involved components are permanently monitored before, during, and after the upgrade, either via monitoring systems, SNMP alerts, or at least with tools like a ping. Critical resources like CPU, memory, network, and/or disk utilization must also be constantly monitored.
To avoid misunderstandings, when performing the tests for each critical application defined on the planning, make sure there are formal notifications on the results for each user area, service, system, and/or application tested.
Regardless if you have to rollback or not, if a problem occurs, make sure you gather as much information about the problem as possible, so you can later place a Support ticket to find a solution.
Last but not least, document the upgrade:
Change Management and Change Control are huge knowledge areas in the field of Information Systems and Computer/Network Security.
This document is by no means a comprehensive list on what you should do when performing an upgrade, with either
Fortinet or any other technology. It is merely a list of important things you should take into consideration when
performing upgrades which are the result of years of experience dealing with changes on critical environments, as it is common that security devices are protecting critical applications and processes.
There are vast resources on the topic: books, public white papers, blog entries, etc. If you search the Internet for the “Change Control Best Practices” or “Change Management Best Practices” you will get many interesting documents.
Upgrading a firewall is something that should be compared to upgrading the operating system on your computer. It’s not to be taken lightly! You want to make sure everything is backed up and you have some options available if things go awry. Assuming it all seems to work you also want a list of things to do in order to confirm everything is working properly. Finally, you need enough time to do it. All really simple stuff, but what does this mean in relation to upgrading your FortiGate? It means, you follow these simple steps:
Digging into this a little, step 1 is easy to understand. Do a full backup of your old configuration. This is all part of your disaster recovery plan. If the upgrade fails in some way you need to make sure you can get the Firewall back up and running. The best way to do this is to get it back to a state where you know what the behavior was. For more information, refer to Performing a configuration backup on page 18.
Step 2, is also part of your disaster recovery. If the upgrade fails you might be able to switch the active partition. But as a Professional, you need to be prepared for the worst case scenario where you can’t do that. Which means you’ll need your old firmware.
Step 3, is your plan for what to do in the event of a critical failure. As we’re talking FortiGate this means that your firewall doesn’t come back after the upgrade. What this means is that you need to be able to get to the console port in order to find out why. Maybe it’s DHCP and the IP changed, maybe the OS is corrupt, who knows? Get to the console and find out.
There could be a simple fix. If there’s not, then be prepared for a format and TFTP reload.
Step 4, READ THE RELEASE NOTES. They contain all kinds of information, known bugs, fixed bugs even upgrade issues like lost configuration settings. Not all upgrade information is ever contained in any products release notes. That does not mean they are devoid of good/useful information. Read them, digest them, then a few days later read them again.
Step 5, do a double check of everything. Is your TFTP server working, does your console connection function, is there anything in the release notes that could impact your upgrade procedure, do you have your configuration backed up? Make sure you’ve done everything.
Step 6, do the upgrade. Doing an upgrade doesn’t take very long, a few minutes (less a lot of times) but make sure you schedule enough time for it. At the end of the day an upgrade can succeed or fail. If it succeeds you want some time to check/confirm that any important features you have are working (VPNs etc). If it fails you’ll need time to sort things out.
Just like upgrading, you need to make sure it’s done properly. While similar, the steps are somewhat different since there are other pitfalls in this case.
Step 1 is very important. This is why, when you upgrade you make a backup of your old configuration and save it. If you don’t, then you’ll need to rebuild manually.
Step 2 is fairly obvious. Even with devices that have multiple partitions and your downgrade process is simply going to be to switch the active partition, this could go wrong. In which case, you may be without Internet access. A professional has a plan for when things go wrong.
Step 3 is no different from before. Hopefully you don’t need to format the unit, but be prepared for that, just in case.
Step 4, once again, is to READ THE RELEASE NOTES. In this case, you will need to do this for the version you are on, and the version you are downgrading too, and everything in between (if you are going back multiple major releases or patches). Maybe the OS switched from 32 to 64 bits somewhere between the two firmware releases. In order to make sure you don’t get nailed by something like that you need to check the upgrade and downgrade information in every major release and patch, as it may have a direct impact on your options.
Step 5 and 6 are the same as before. Double check everything, then downgrade.
Step 7 is new. Obviously most settings are lost when you downgrade so in order to get back up and running you will need to restore your old configuration file.
Once you configure the FortiGate unit and it is working correctly, it is extremely important that you backup the configuration. In some cases, you may need to reset the FortiGate unit to factory defaults or perform a TFTP upload of the firmware, which will erase the existing configuration. In these instances, the configuration on the device will have to be recreated, unless a backup can be used to restore it.
It is also recommended that once any further changes are made that you backup the configuration immediately, to ensure you have the most current configuration available. Also, ensure you backup the configuration before upgrading the FortiGate unit’s firmware. Should anything happen during the upgrade that changes the configuration, you can easily restore the saved configuration.
Always backup the configuration and store it on the management computer or off-site. You have the option to save the configuration file to various locations including the local PC, USB key, FTP and TFTP site.The latter two are configurable through the CLI only.
If you have VDOMs, you can back up the configuration of the entire FortiGate unit or only a specific VDOM. Note that if you are using FortiManager or FortiCloud, full backups are performed and the option to backup individual VDOMs will not appear.
To back up the FortiGate configuration – GUI:
The USB Disk option will be grayed out if no USB drive is inserted in the USB port. You can also backup to the FortiManager using the CLI.
Encryption must be enabled on the backup file to back up VPN certificates.
To back up the FortiGate configuration – CLI:
execute backup config management-station <comment>
… or …
execute backup config usb <backup_filename> [<backup_password>] … or for FTP (note that port number, username are optional depending on the FTP site)… execute backup config ftp <backup_filename> <ftp_server> [<port>] [<user_name>] [<password>]
… or for TFTP … execute backup config tftp <backup_filename> <tftp_servers> <password> Use the same commands to backup a VDOM configuration by first entering the commands:
config vdom edit <vdom_name>
You can use secure copy protocol (SCP) to download the configuration file from the FortiGate unit as an alternative method of backing up the configuration file or an individual VDOM configuration file. This is done by enabling SCP for and administrator account and enabling SSH on a port used by the SCP client application to connect to the FortiGate unit. SCP is enabled using the CLI commands:
config system global set admin-scp enable end
Use the same commands to backup a VDOM configuration by first entering the commands:
config global
set admin-scp enable
end config vdom
edit <vdom_name>
By implementing the following best practices for system and performance, you will ensure maximum efficiency of your FortiGate device. Be sure to read everything carefully, particularly the section that concerns shutting down the FortiGate system, in order to avoid potential hardware issues.
Always shut down the FortiGate operating system properly before turning off the power switch to avoid potentially catastrophic hardware problems.
To power off the FortiGate unit – GUI:
To power off the FortiGate unit – CLI:
execute shutdown
Once this has been done, you can safely turn off the power switch or disconnect the power cables from the power supply.
Network administrators are often reluctant to change firewall vendors due to the perception that the migration process is difficult. Indeed, there is no point hiding the fact that moving to a new vendor requires careful consideration. But concern over the potential pain of migration should not stand in the way of adopting new security technologies. The purpose of this chapter is to describe the best practices for performing such migrations and ultimately to ease the migration process itself.
It is always best practice to perform a full network audit prior to any migration. This should include:
Understanding exactly where cables run in the network and verifying they are all correctly labeled is essential to avoid mistakes and unnecessary downtime during the upgrade. Don’t overlook simple things such as:
No matter how securely a FortiGate is configured in the network, it cannot help if it has been bypassed; visually checking where the device sits in the network in relation to other devices will ensure you are maintaining security and verify the network diagram is ‘as built’. Details of all networks including subnet masks should be documented at this point to ensure that the replacement device is configured with the correct information.
Whilst we have suggested some level of manual review is included in the policy migration, it can be useful to be able to automatically migrate simply between another vendor’s format and the FortiGate format. The FortiGate policy format is text based and can easily be cut and pasted into from other vendor formats however, responding to the high customer demand to migrate away from other vendors, Fortinet have released an automatic configuration migration tool at http://convert.fortinet.com to simplify this process. Supporting Cisco ACLs, PIX, ASA, Check Point, and Juniper, the Converter can securely upload and convert the policy into the Fortinet format.
This is an important process and should be tested offline first wherever possible i.e. configure the policy in the lab or on a test network and verify that the required access permissions are being implemented. To really test the solution out, the FortiGate can be implemented on the live network with a different gateway IP and the selected user pointed to the new gateway. This allows a staged approach to migrating the new platform into the network ensuring that the process does not interrupt day to day operations.
If testing and validation is successful at this point, you can migrate to the new firewall either by switching IP’s and removing the old devices or by changing the default gateway in DHCP. Once the firewall is in place, acceptance testing will of course need to be carried out and an iterative process of tuning undertaken to finalize the configuration.
The Fortinet solution will have a plethora of additional features compared to your previous vendor and it is very tempting to start switching them on but it is a good idea to wait and validate the new firewall as was previously configured before adding new functions as this simplifies testing and problem diagnosis. Finally complete the migration (don’t forget about the Plan Do Check Act Cycle) by adding any new services that were requested and learn about the multiple features you have available with the FortiGate appliance.
Little video discussing the basics of HA as well as configuring and cabling up redundant ISP connections (with failover)
This chapter provides an example of a FortiGate unit providing authenticated access to the Internet for both Windows network users and local users. The following topics are included in this section:
l Firewall authentication example l LDAP dial-in using member-attribute example l RADIUS SSO example l Troubleshooting
In this example, there is a Windows network connected to Port 2 on the FortiGate unit and another LAN, Network_1, connected to Port 3.
All Windows network users authenticate when they logon to their network. Members of the Engineering and Sales groups can access the Internet without entering their authentication credentials again. The example assumes that the Fortinet Single Sign On (FSSO) has already been installed and configured on the domain controller.
LAN users who belong to the Internet_users group can access the Internet after entering their username and password to authenticate. This example shows only two users, User1 is authenticated by a password stored on the FortiGate unit, User2 is authenticated on an external authentication server. Both of these users are referred to as local users because the user account is created on the FortiGate unit.
User1 is authenticated by a password stored on the FortiGate unit. It is very simple to create this type of account.
To create a local user – web-based manager:
| User Type | Local User |
| User Name | User1 |
| Password | hardtoguess |
| Email Address
SMS |
(optional) |
| Enable | Select. |
To create a local user – CLI:
config user local edit user1 set type password set passwd hardtoguess
end
To authenticate users using an external authentication server, you must first configure the FortiGate unit to access the server.
To configure the remote authentication server – web-based manager:
| Name | OurRADIUSsrv |
| Primary Server Name/IP | 10.11.101.15 |
| Primary Server Secret | OurSecret |
| Authentication Scheme | Select Use Default Authentication Scheme. |
To configure the remote authentication server – CLI:
config user radius edit OurRADIUSsrv set server 10.11.102.15 set secret OurSecret
Firewall authentication
set auth-type auto
end
Creation of the user account is similar to the locally-authenticated account, except that you specify the RADIUS authentication server instead of the user’s password.
To configure a remote user – web-based manager:
| User Type | Remote RADIUS User |
| User Name | User2 |
| RADIUS server | OurRADIUSsrv |
| Email Address
SMS |
(optional) |
| Enable | Select |
To configure a remote user – CLI:
config user local edit User2 set name User2 set type radius
set radius-server OurRADIUSsrv
end
There are two user groups: an FSSO user group for FSSO users and a firewall user group for other users. It is not possible to combine these two types of users in the same user group.
For this example, assume that FSSO has already been set up on the Windows network and that it uses Advanced mode, meaning that it uses LDAP to access user group information. You need to
To configure LDAP for FSSO – web-based manager:
| Name | ADserver | |
| Server Name / IP | 10.11.101.160 | |
| Distinguished Name | dc=office,dc=example,dc=com | |
| Bind Type | Regular | |
| User DN | cn=FSSO_Admin,cn=users,dc=office,dc=example,dc=com | |
| Password | Enter a secure password. | |
To configure LDAP for FSSO – CLI”
config user ldap edit “ADserver” set server “10.11.101.160”
set dn “cn=users,dc=office,dc=example,dc=com”
set type regular
set username “cn=administrator,cn=users,dc=office,dc=example,dc=com” set password set_a_secure_password
next
end
To specify the collector agent for FSSO – web-based manager:
To specify the collector agent for FSSO – CLI:
config user fsso edit “WinGroups” set ldap-server “ADserver” set password ENC
G7GQV7NEqilCM9jKmVmJJFVvhQ2+wtNEe9T0iYA5Sa+EqT2J8zhOrbkJFDr0RmY3c4LaoXdsoBczA
1dONmcGfthTxxwGsigzGpbJdC71spFlQYtj set server “10.11.101.160” end
Firewall authentication
To create the FSSO_Internet-users user group – web-based manager:
| Name | FSSO_Internet_users |
| Type | Fortinet Single Sign-On (FSSO) |
| Members | Engineering, Sales |
To create the FSSO_Internet-users user group – CLI:
config user group edit FSSO_Internet_users set group-type fsso-service
set member CN=Engineering,cn=users,dc=office,dc=example,dc=com
CN=Sales,cn=users,dc=office,dc=example,dc=com end
The non-FSSO users need a user group too. In this example, only two users are shown, but additional members can be added easily.
To create the firewall user group – web-based manager:
| Name | Internet_users |
| Type | Firewall |
| Members | User1, User2 |
To create the firewall user group – CLI:
config user group edit Internet_users set group-type firewall set member User1 User2
end
| Address Name | Internal_net |
| Type | Subnet |
| Subnet / IP Range | 10.11.102.0/24 |
| Interface | Port 3 |
| Address Name | Windows_net |
| Type | Subnet |
| Subnet / IP Range | 10.11.101.0/24 |
| Interface | Port 2 |
Two security policies are needed: one for firewall group who connect through port3 and one for FSSO group who connect through port2.
To create a security policy for FSSO authentication – web-based manager:
| Incoming Interface | Port2 |
| Source Address | Windows_net |
| Source User(s) | FSSO_Internet_users |
| Outgoing Interface | Port1 |
| Destination Address | all |
| Schedule | always |
| Service | ALL |
| NAT | ON |
| Security Profiles | Optionally, enable security profiles. |
To create a security policy for FSSO authentication – CLI:
config firewall policy edit 0 set srcintf port2 set dstintf port1 set srcaddr Windows_net set dstaddr all
LDAP dial-in using member-attribute
set action accept set groups FSSO_Internet_users set schedule always set service ANY set nat enable
end
To create a security policy for local user authentication – web-based manager:
| Incoming Interface | Port3 |
| Source Address | Internal_net |
| Source User(s) | Internet_users |
| Outgoing Interface | Port1 |
| Destination Address | all |
| Schedule | always |
| Service | ALL |
| NAT | ON |
| Security Profiles | Optionally, enable security profiles. |
To create a security policy for local user authentication – CLI:
config firewall policy edit 0 set srcintf port3 set dstintf port1 set srcaddr internal_net set dstaddr all set action accept set schedule always set groups Internet_users set service ANY set nat enable
end
In this example, users defined in MicroSoft Windows Active Directory (AD) are allowed to set up a VPN connection simply based on an attribute that is set to TRUE, instead of based on their user group. In AD the “Allow Dialin” property is activated in the user properties, and this sets the msNPAllowDialin attribute to “TRUE”.
LDAP dial-in using member-attribute example
This same procedure can be used for other member attributes, as your system requires.
To accomplish this with a FortiGate unit, member-attribute must be set. This can only be accomplished through the CLI – the option is not available through the web-based manager.
Before configuring the FortiGate unit, ensure the AD server has the msNPAllowDialin attribute set to “TRUE” for the users in question. If not, those users will not be able to authenticate.
To configure user LDAP member-attribute settings – CLI:
config user ldap edit “ldap_server” set server “192.168.201.3” set cnid “sAMAccountName” set dn “DC=fortilabanz,DC=com,DC=au” set type regular
set username “fortigate@sample.com” set password ****** set member-attr “msNPAllowDialin”
next
end
To configure LDAP group settings – CLI:
config user group edit “ldap_grp” set member “ldap” config match edit 1 set server-name “ldap” set group-name “TRUE”
next
end
next
end
Once these settings are in place, users that are a member of the ldap user group will be able to authenticate.
To ensure your settings are correct, here is the sample output from a diag debug command that shows the authentication process.
When the “Allow Dial-in” attribute is set to “TRUE” the following will likely be in the output:
get_member_of_groups-Get the memberOf groups. get_member_of_groups- attr=’msNPAllowDialin’, found 1 values
get_member_of_groups-val[0]=’TRUE’ fnbamd_ldap_get_result-Auth accepted fnbamd_ldap_get_result-Going to DONE state res=0
fnbamd_auth_poll_ldap-Result for ldap svr 192.168.201.3 is SUCCESS fnbamd_auth_poll_ldap-Passed group matching
If the attribute is not set but it is expected, the following will likely be in the output:
get_member_of_groups-Get the memberOf groups. get_member_of_groups- attr=’msNPAllowDialin’, found 1 values
get_member_of_groups-val[0]=’FALSE’ fnbamd_ldap_get_result-Auth accepted fnbamd_ldap_get_result-Going to DONE state res=0
fnbamd_auth_poll_ldap-Result for ldap svr 192.168.201.3 is SUCCESS fnbamd_auth_poll_ldap-Failed group matching
The only difference between these two outputs is the last line which is either passed or failed based on if the member-attribute is set to the expected value or not.
A common RADIUS SSO topology involves a medium sized company network of users connecting to the Internet through the FortiGate unit, and authenticating with a RADIUS server. RADIUS SSO authentication was selected because it is fast and relatively easy to configure.
This section includes:
l Assumptions l Topology l Configuring RADIUS l Configuring FortiGate regular and RADIUS SSO security policies l Testing
Example.com has an office with 20 users on the internal network. These users need access to the Internet to do their jobs. The office network is protected by a FortiGate-60C unit with access to the Internet through the wan1 interface, the user network on the internal interface, and all the servers are on the DMZ interface. This includes an Ubuntu Linux server running FreeRADIUS. For this example only two users will be configured — Pat Lee with an account name plee, or plee@example.com, and Kelly Green with an account name kgreen, or kgreen@example.com.
RADIUS SSO topology
Configuring RADIUS includes configuring the RADIUS server such as FreeRADIUS, a radius client on user’s computers, and configuring users in the system. For this example the two users will be Pat Lee, and Kelly Green. They belong to a group called exampledotcom_employees. When it is all configured, the RADIUS daemon needs to started.
The users have a RADIUS client installed on their PCs that allows them to authenticate through the RADIUS server.
FreeRADIUS can be found on the freeradius.org website. For any problems installing FreeRADIUS, see the FreeRADIUS documentation.
Before configuring the RADIUS SSO security policy, configure FortiGate interfaces. This includes defining a DHCP server for the internal network as this type of network typically uses DHCP. The wan1 and dmz interfaces are assigned static IP addresses and do not need a DHCP server.
FortiGate interfaces used in this example
| Interface | Subnet | Act as DHCP Server | Devices |
| wan1 | 172.20.120.141 | No | Internet Service Provider |
| dmz | 10.11.101.100 | No | Servers, including RADIUS server |
| internal | 10.11.102.100 | Yes: x.x.x.110-.250 | Internal user network |
To configure FortiGate interfaces – web-based manager:
| Alias | Internet |
| Addressing Mode | Manual |
| IP/Network Mask | 172.20.120.141/255.255.255.0 |
| Administrative Access | HTTPS, SSH |
| Enable DHCP Server | Not selected |
| Comments | Internet |
| Administrative Status | Up |
| Alias | Servers |
| Addressing Mode | Manual |
| IP/Network Mask | 10.11.101.100/255.255.255.0 |
| Administrative Access | HTTPS, SSH, PING, SNMP |
| Enable DHCP Server | Not selected |
| Listen for RADIUS
Accounting Messages |
Select |
| Comments | Servers |
| Administrative Status | Up |
| Alias | Internal network |
| Addressing Mode | Manual |
| IP/Network Mask | 10.11.102.100/255.255.255.0 |
| Administrative Access | HTTPS, SSH, PING |
| Enable DHCP Server | Select |
| Address Range | 10.11.102.110 – 10.11.102.250 |
| Netmask | 255.255.255.0 |
| Default Gateway | Same as Interface IP |
| DNS Server | Same as System DNS |
| Comments | Internal network |
| Administrative Status | Up |
To create a RADIUS SSO agent:
To define a local user group for RADIUS SSO:
With the RADIUS server and FortiGate interfaces configured, security policies can be configured. This includes both RADIUS SSO and regular policies, as well as addresses and address groups. All policies require NAT to be enabled.
Security policies required for RADIUS SSO
| Seq. No. | From -> To | Type | Schedule | Description | ||
| 1 | internal -> wan1 | RADIUS SSO | business hours | Authenticate outgoing user traffic. | ||
| 2 | internal -> wan1 | regular | always | Allow essential network services and VoIP. | ||
| 3 | dmz -> wan1 | regular | always | Allow servers to access Internet. | ||
| Seq. No. | From -> To | Type | Schedule | Description | ||
| 4 | internal ->
dmz |
regular | always | Allow users to access servers. | ||
| 5 | any -> any | deny | always | Implicit policy denying all traffic that hasn’t been matched. | ||
The RADIUS SSO policy must be placed at the top of the policy list so it is matched first. The only exception to this is if you have a policy to deny access to a list of banned users. In this case, that policy must go at the top so the RADIUS SSO does not mistakenly match a banned user or IP address.
This section includes:
l Schedules, address groups, and services groups l Configuring regular security policies l Configuring RADIUS SSO security policy
This section lists the lists that need to be configured before security policies are created. Creating these lists is straight forward, so the essential information has been provided here but not step by step instructions. For more information on firewall related details, see
Schedules
Only one schedule needs to be configured — business_hours. This is a fairly standard Monday to Friday 8am to 5pm schedule, or whatever days and hours covers standard work hours at the company.
Address groups
The following address groups need to be configured before the security policies.
| Address Group Name | Interface | Address range included |
| internal_network | internal | 10.11.102.110 to 10.11.102.250 |
| company_servers | dmz | 10.11.101.110 to 10.11.101.250 |
Service groups
The following service groups need to be configured before the security policies. Note that the services listed are suggestions and may include more or less as required.
| Service Group Name | Interface | Description of services to be included |
| essential_network_services | internal | Any network protocols required for normal network operation such as DNS, NTP, BGP. |
| essential_server_services | dmz | All the protocols required by the company servers such as BGP, HTTP, HTTPS, FTP, IMAP, POP3, SMTP, IKE, SQL, MYSQL, NTP, TRACEROUTE, SOCKs, and SNMP. |
| user_services | internal | Any protocols required by users HTTP, HTTP, FTP, |
The following security policy configurations are basic and only include logging, and default AV and IPS.
Regular security policies allow or deny access for non-RADIUS SSO traffic. This is essential as there are network services—such as DNS, NTP, and FortiGuard—that require access to the Internet.
To configure regular security policies – web-based manager:
| Incoming Interface | Internal |
| Source Address | internal_network |
| Outgoing Interface | wan1 |
| Destination Address | all |
| Schedule | always |
| Service | essential_network_services |
| Action | ACCEPT |
| NAT | ON |
| Security Profiles | ON: AntiVirus, IPS |
| Log Allowed Traffic | ON |
| Comments | Essential network services |
| Incoming Interface | dmz | |
| Source Address | company_servers | |
| Outgoing Interface | wan1 | |
| Destination Address | all | |
| Schedule | always | |
| Service | essential_server_services | |
| Action | ACCEPT | |
| NAT | ON | |
| Security Profiles | ON: AntiVirus, IPS | |
| Log Allowed Traffic | enable | |
| Comments | Company servers accessing the Internet | |
| Incoming Interface | Internal |
| Source Address | internal_network |
| Outgoing Interface | dmz |
| Destination Address | company_servers |
| Schedule | always |
| Service | all |
| Action | ACCEPT |
| NAT | ON |
| Security Profiles | ON: AntiVirus, IPS |
| Log Allowed Traffic | enable |
| Comments | Access company servers |
The RADIUS SSO policy allows access for members of specific RADIUS groups.
To configure RADIUS SSO security policy:
| Incoming Interface | Internal |
| Source Address | internal_network |
| Source User(s) | Select the user groups you created for RSSO. |
| Outgoing Interface | wan1 |
| Destination Address | all |
| Schedule | business_hours |
| Service | ALL |
| Action | ACCEPT |
| NAT | ON |
| Security Profiles | ON: AntiVirus, Web Filter, IPS, and Email Filter. In each case, select the default profile. |
Once configured, a user only needs to log on to their PC using their RADIUS account. After that when they attempt to access an Internet website, the FortiGate unit will use their session information to get their RADIUS information. Once the user is verified, they are allowed access to the website.
To test the configuration perform the following steps:
Each step generates log entries that enable you to verify that each step was successful.
Troubleshooting
RADIUS SSO test
In the web-based manager, a good tool for troubleshooting is the packet counter column on the security policy page at Policy & Objects > IPv4 Policy. This column displays the number of packets that have passed through this security policy. Its value when you are troubleshooting is that when you are testing your configuration (end to end connectivity, user authentication, policy use) watching the packet count for an increase confirms any other methods you may be using for troubleshooting. It provides the key of which policy is allowing the traffic, useful information if you expect a user to require authentication and it never happens.
This section addresses how to get more information from the CLI about users and user authentication attempts to help troubleshoot failed authentication attempts. diag firewall iprope list
Shows the IP that the computer connected from. This is useful to confirm authorization and VPN settings.
diag firewall iprope clear
Clear all authorized users from the current list. Useful to force users to re-authenticate after system or group changes. However, this command may easily result in many users having to re-authenticate, so use carefully.
diag rsso query ip diag rsso query rsso-key
Queries the RSSO database.