Rules: Difference between revisions

From Edge Threat Management Wiki - Arista
Jump to navigationJump to search
No edit summary
(Undo revision 26017 by Cblaise (talk))
Line 1: Line 1:
[[Category:Applications]]
= Rules =
<span style="display:none" class="helpSource intrusion_prevention">Intrusion_Prevention</span>
<span style="display:none" class="helpSource intrusion_prevention_status">Intrusion_Prevention#Status</span>
<span style="display:none" class="helpSource intrusion_prevention_rules">Intrusion_Prevention#Rules</span>
<span style="display:none" class="helpSource intrusion_prevention_variables">Intrusion_Prevention#Variables</span>
<span style="display:none" class="helpSource intrusion_prevention_event_log">Intrusion_Prevention#Event_Log</span>


{| width='100%'
Rules are used frequently in Untangle and many other firewalls. Rules are very powerful, but can sometimes be difficult to configure. This documentation describes how rules work and gives some basic examples and some common mistakes to avoid. Many applications use rules like [[Firewall]], [[Captive Portal]], [[Application Control]], [[Bandwidth Control]], etc. All of these rules essentially share the same logic.
|-
| align="center" | [[Image:IntrusionPrevention.png|128px]] &nbsp; &nbsp; '''Intrusion Prevention'''
| align="center" |
{|
|-
| Other Links:
|-
|[http://www.untangle.com/store/intrusion-prevention.html Intrusion Prevention Description Page]
|-
|[http://demo.untangle.com/admin/index.do#service/intrusion-prevention Intrusion Prevention Demo]
|-
|[http://forums.untangle.com/intrusion-prevention/ Intrusion Prevention Forums]
|-
|[[Intrusion Prevention Reports]]
|-
|[[Intrusion Prevention FAQs]]
|}
|}
<br/>
----
 
== About Intrusion Prevention ==
 
Intrusion Prevention is an [http://en.wikipedia.org/wiki/Intrusion_detection_systems Intrusion Detection system] that detects malicious activity on your network.
 
To detect malicious activity, Intrusion Prevention uses ''signatures'', a method that draws upon a database of known attack patterns.
If a network [http://en.wikipedia.org/wiki/Session_%28computer_science%29 session] matches a signature, its enabled ''action'' directs Intrusion Prevention to
''Log'' (records the incident but '''does not stop''' the activity) or ''Block'' (records the incident and '''does stop''' the activity).
 
There is tremendous diversity between networks and it is possible for a signature to correctly identify malicious activity on one network and incorrectly match legitimate traffic on another.
Logging all matching signatures can make it difficult to effectively monitor Intrusion Prevention and blocking can disrupt legitimate traffic causing cause your network to appear to be broken. 
Therefore it is perfectly legitimate for there to be many signatures set as ''disabled'' or not active in Intrusion Prevention.
In fact, it is advised that you use to the ''Recommended'' actions as specified by the signature database providers.
 
The database contains over 26,000 signatures making it difficult to manage signatures directly.
''Rules'' are used to configure groups of signatures on matching various attributes.
A condition can match an attribute such as classtype. For all signatures that match, they are configured in Intrusion Prevention according to the rule action.
Any signature not matched by a rule is Disabled.
A default set of rules based on system memory are enabled by default.
 
The signature database is automatically updated several times a week.  New and updated rules will be configured as determined by rules.
 
All detected activity for enabled signatures is recorded to the Intrusion Prevention ''All Events'' log.  You should review this log on a daily basis.
 
''Note:'' Intrusion Prevention installs but is off by default.
 
''Note:'' Intrusion Prevention can be memory intensive and requires at least 2GB of RAM.  The amount used is a combination of the number of enabled signatures and the amount of traffic that goes through your system.
 
== Settings ==
 
=== Status ===
 
The Status tab shows the following information:
 
* Memory Usage.  The amount of system memory the IPS engine is using compared to your installed system memory.
 
* Metrics.  The number of blocked, logged, and scanned sessions.
 
* Overview.  Signatures and Signature Updates.
** Signatures.  Total number of signatures available and the number set for Log, Block, Disabled.
** Updates.  The last time the signature database was updated and the last time a check was performed.  Database updates do not occur on each check.


{{ServiceAppScreenshot|intrusion-prevention|status}}
= Basics =


Rules are configured by the user to categorize and act upon traffic. For example, [[Firewall]] uses rules to determine whether to block or pass traffic. [[Bandwidth Control]] uses rules to determine how to prioritize a session.


=== Rules ===
Rules are evaluated in order from top to bottom against sessions (not packets!). If a rule matches then the action from that rule is performed and no more rules are evaluated. If no matching rule is found the behavior is defined by the application, which is usually doing nothing.


Rules allow you to control which signatures are enabled (and their actions) or disabled.
This is similar to other firewalls. Untangle rules are "quick" rules which means the first match is always used. Unlike some other firewalls, Untangle evaluates against sessions, not packets. This means that the "Source Address" will be the initiator of the session and the "Destination Address" will be the server address the client is connecting to, and the same is true for "Source Port" and "Destination Port."
Any signature not matched by a rule is disabled.


The [[Rules|Rules documentation]] describes how rules generally work and how they are configured.  The major difference for Intrusion Prevention is the Conditions List.
Each rule has several properties:
* An enable checkbox
* A name/description
* A set of conditions
* An action or set of actions


{{AppScreenshot|intrusion-prevention|rules}}
[[Image:Rule_anatomy.png|frame|center|Rule Anatomy]]


==== Condition List ====
The enable checkbox determines if the rule is evaluated. If the enable checkbox is not checked, the rule is simply skipped.


Conditions define which signatures will match the rule. If and only if all of the conditions match, the rule is considered a match.  
The description is for the user to document what the rule does. It is ''highly'' suggested to give the rule a meaningful name. Trying to troubleshoot a set of rules all named "[no description]" can be extremely difficult.


Conditions can be compared in one of two ways:
The set of conditions is the description of the traffic that should match the rule. If ''all'' of the conditions are true for the given session then the rules matches. This is discussed in more details in the next section.


The action or set of actions configures what action is performed if the rule matches. This is dependent on the application. For example, in [[Firewall]] it determines whether to block or pass the session and additionally if it should be flagged.


= Conditions =


can also be inverted by selecting "is NOT" in the dropdown, effectively inverting when it matches. ''Destination Port'' "is NOT" "80" matches on all ports except port 80.
Conditions define which sessions will match the rule. If and only if all of the conditions match, the rule is considered a match.
Conditions can also be inverted by selecting "is NOT" in the dropdown, effectively inverting when it matches. ''Destination Port'' "is NOT" "80" matches on all ports except port 80.


Lets take a simple example: We want to block TCP traffic to port 80 on a server. There is a web service running on that server that you don't want to allow access to. First, create a rule and check the ''enable'' checkbox and give it a reasonably descriptive name like "blocking TCP port 80 to serverX." Now you will need to add some conditions that match only the traffic you want to block.
Lets take a simple example: We want to block TCP traffic to port 80 on a server. There is a web service running on that server that you don't want to allow access to. First, create a rule and check the ''enable'' checkbox and give it a reasonably descriptive name like "blocking TCP port 80 to serverX." Now you will need to add some conditions that match only the traffic you want to block.
Line 109: Line 50:
* Reliability - The ''Reliability'' is a "reliability" of that condition. ''True'' means it is 100% reliable. ''False'' means it is 99% or less reliable. For example, some conditions, like ''Destination Port'', are always known and thus matching on ''Destination Port'' is 100% reliable. Other conditions, like ''Client Hostname'', only match if the hostname for the client hostname is known. Hostname is determined through many ways, including DNS and DHCP. If all of these methods fail then it is entirely possible that the hostname of a server is "foo" but Untangle has not been able to determine this and as such a ''Client Hostname'' is "foo" condition will not match. This column is there purely as a warning that users in these cases must be aware of when conditions might deliver false negatives.  
* Reliability - The ''Reliability'' is a "reliability" of that condition. ''True'' means it is 100% reliable. ''False'' means it is 99% or less reliable. For example, some conditions, like ''Destination Port'', are always known and thus matching on ''Destination Port'' is 100% reliable. Other conditions, like ''Client Hostname'', only match if the hostname for the client hostname is known. Hostname is determined through many ways, including DNS and DHCP. If all of these methods fail then it is entirely possible that the hostname of a server is "foo" but Untangle has not been able to determine this and as such a ''Client Hostname'' is "foo" condition will not match. This column is there purely as a warning that users in these cases must be aware of when conditions might deliver false negatives.  


== Condition List ==


{| border="1" cellpadding="2" font="sans-serif" style="border-collapse: collapse;"
{| border="1" cellpadding="2"
|+
|+
! style="text-align: left" | Name
! Name !! Syntax !! Function !! Availability !! Reliability
! style="text-align: left" | Syntax
|-
! style="text-align: left" | Function
| style="width: 20%;"|Destination Address
| [[IP Matcher]]
| Matches if value matches the Destination/Server Address of the session. (after NAT/port forwarding)
| all
| True
|-
| Destination Port
| [[Int Matcher]]
| Matches if value matches the Destination/Server Port of the session. (after NAT/port forwarding)
| all
| True
|-
| Destination Interface
| checkboxes
| Matches if value matches the Destination/Server Interface of the session. (after routing/port forwarding)
| all
| True
|-
| Destined Local
| boolean
| Matches if the session is destined to one of Untangle's IP addresses.
| all
| True
|-
| Source Address
| [[IP Matcher]]
| Matches if value matches the Source/Client Address of the session. (before NAT/port forwarding)
| all
| True
|-
| Source Port
| [[Port Matcher]]
| Matches if value matches the Source/Client Port of the session. (before NAT/port forwarding)
| hidden
| True
|-
| Source Interface
| checkboxes
| Matches if value matches the Source/Client Interface of the session. (before routing/port forwarding)
| all
| True
|-
| Protocol
| checkboxes
| Matches if value matches the Protocol of the session.
| all
| True
|-
| Tagged
| [[Glob Matcher]]
| Matches if session is tagged with matching tag
| all
| True
|-
| Client Tagged
| [[Glob Matcher]]
| Matches if client of the session is tagged with matching tag
| all
| True
|-
| Server Tagged
| [[Glob Matcher]]
| Matches if server of the session is tagged with matching tag
| all
| True
|-
| Username
| [[User Matcher]]
| Matches if value matches the username associated with the Client IP in the Host table.
| all
| False
|-
| Host Hostname
| [[Glob Matcher]]
| Matches if value matches the hostname associated with the Local Address in the Host table.
| all
| False
|-
| Client Hostname
| [[Glob Matcher]]
| Matches if value matches the hostname associated with the Client IP in the Host table.
| all
| False
|-
| Server Hostname
| [[Glob Matcher]]
| Matches if value matches the hostname associated with the Server IP in the Host table.
| all
| False
|-
| Client MAC Address
| [[Glob Matcher]]
| Matches if value matches the MAC address associated with the Client IP in the ARP table.
| all
| False
|-
| Server MAC Address
| [[Glob Matcher]]
| Matches if value matches the MAC address associated with the Server IP in the ARP table.
| all
| False
|-
| Client MAC Vendor
| [[Glob Matcher]]
| Matches if value matches the device manufacturer. This is identified using the Client IP's MAC address in the ARP table and [https://standards.ieee.org/develop/regauth/oui/public.html OUI Lookup].
| all
| False
|-  
| Server MAC Vendor
| [[Glob Matcher]]
| Matches if value matches the device manufacturer. This is identified using the Server IP's MAC address in the ARP table and [https://standards.ieee.org/develop/regauth/oui/public.html OUI Lookup].
| all
| False
|-
| Host has no Quota
| boolean
| Matches if the local address has no quota
| all
| True
|-
| User has no Quota
| boolean
| Matches if the username has no quota
| all
| True
|-
| Client has no Quota
| boolean
| Matches if the client has no quota
| hidden
| True
|-
| Server has no Quota
| boolean
| Matches if the server has no quota
| hidden
| True
|-
| Host has Exceeded Quota
| boolean
| Matches if the local address has exceeded their quota
| all
| True
|-
| User has Exceeded Quota
| boolean
| Matches if the user has exceeded their quota
| all
| True
|-
| Client has Exceeded Quota
| boolean
| Matches if the client has exceeded their quota
| hidden
| True
|-
| Server has Exceeded Quota
| boolean
| Matches if the server has exceeded their quota
| hidden
| True
|-
| Host Quota Attainment
| [[Int Matcher]]
| Matches the local address quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc)
| all
| True
|-
| User Quota Attainment
| [[Int Matcher]]
| Matches the username quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc)
| all
| True
|-  
| Client Quota Attainment
| [[Int Matcher]]
| Matches the client quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc)
| hidden
| True
|-
| Server Quota Attainment
| [[Int Matcher]]
| Matches the server quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used. etc)
| hidden
| True
|-
| HTTP: Hostname
| [[Glob Matcher]]
| Matches if the value matches the hostname specified in the HTTP session
| deep sessions
| False
|-
| HTTP: Referer
| [[Glob Matcher]]
| Matches if the value matches the referer string specified in the HTTP header
| deep sessions
| False
|-
| HTTP: URI
| [[Glob Matcher]]
| Matches if the value matches the latest URI specified in the HTTP session
| deep sessions
| False
|-
| HTTP: URL
| [[URL Matcher]]
| Matches if the value matches the latest URL (hostname+URI) specified in the HTTP session
| deep sessions
| False
|-
| HTTP: Content Type
| [[Glob Matcher]]
| Matches the content-type of the latest content in the HTTP session
| deep sessions
| False
|-
| HTTP: Content Length
| [[Int Matcher]]
| Matches if [[Int Matcher]] matches the content length specified in the latest HTTP response
| deep sessions
| False
|-
| HTTP: Request Method
| [[Glob Matcher]]
| Matches if the value matches the HTTP request method of the requested URL. Standard request methods include GET, POST, HEAD, OPTIONS, PUT, DELETE, TRACE, and CONNECT.
| deep sessions
| False
|-
| HTTP: Request File Path
| [[Glob Matcher]]
| Matches if the value matches the entire file path of the requested URL. This is everything including and after the first slash character following the host name or address. (e.g. /some/location/mypage.html)
| deep sessions
| False
|-
| HTTP: Request File Name
| [[Glob Matcher]]
| Matches if the value matches the file name of the requested URL. This is everything after the final slash character of the request. (e.g. mypage.html)
| deep sessions
| False
|-
| HTTP: Request File Extension
| [[Glob Matcher]]
| Matches if the value matches the file extension of the requested URL. This is everything after following the dot after the final slash character of the request. (e.g. html)
| deep sessions
| False
|-
| HTTP: Response Content Type
| [[Glob Matcher]]
| Matches if the value matches the content or MIME type reported in the server response.
| deep sessions
| False
|-
| HTTP: Response File Name
| [[Glob Matcher]]
| Matches if the value matches the file name returned in the server response.
| deep sessions
| False
|-
| HTTP: Response File Extension
| [[Glob Matcher]]
| Matches if the value matches the file extension returned in the server response. This is everything after (but not including) the final dot of the file name.
| deep sessions
| False
|-
| HTTP: Client User Agent
| [[Glob Matcher]]
| Matches if the value matches the User Agent string for the client in the Host table
| all
| False
|-  
|-  
| Signature identifier
| Application Control: Application
| Numeric
| [[Glob Matcher]]
| Matches if value matches the exact or partial signature identifier.
| Matches if the value matches the Application determined by Application Control
| deep sessions
| False
|-  
|-  
| Group identifier
| Application Control: Category
| Numeric
| [[Glob Matcher]]
| Matches if value matches the exact or partial group identifier.
| Matches if the value matches the Category of the Application determined by Application Control
| deep sessions
| False
|-  
|-  
| Category
| Application Control: Protochain
| checkbox
| [[Glob Matcher]]
| Matches if value is in one of the checked categories.
| Matches if the value matches the Protochain determined by Application Control
| deep sessions
| False
|-  
|-  
| Classtype
| Application Control: Detail
| checkbox
| [[Glob Matcher]]
| Matches if value is in one of the checked classtypes.
| Matches if the value matches the Detail of the Application determined by Application Control
| deep sessions
| False
|-  
|-  
| Message
| Application Control: Confidence
| Text
| [[Int Matcher]]
| Matches if value matches the exact or partial signature subject message.
| Matches if [[Int Matcher]] matches the confidence rating determined by Application Control
| deep sessions
| False
|-  
|-  
| Protocol
| Application Control: Productivity
| checkbox
| [[Int Matcher]]
| Matches if value is in one of the checked protocols.
| Matches if [[Int Matcher]] matches the productivity rating determined by Application Control
| deep sessions
| False
|-  
|-  
| Source Address
| Application Control: Risk
| Text
| [[Int Matcher]]
| Matches if value matches the exact or partial source address.
| Matches if [[Int Matcher]] matches the risk rating determined by Application Control
| deep sessions
| False
|-  
|-  
| Source Port
| Application Control Lite: Signature
| Text
| [[Glob Matcher]]
| Matches if value matches the exact or partial source port.
| Matches if the value matches the Signature determined by Application Control Lite
| deep sessions
| False
|-  
|-  
| Destination Address
| Application Control Lite: Category
| Text
| [[Glob Matcher]]
| Matches if value matches the exact or partial destination address.
| Matches if the value matches the Category of the Signature determined by Application Control Lite
| deep sessions
| False
|-  
|-  
| Destination Port
| Application Control Lite: Description
| Text
| [[Glob Matcher]]
| Matches if value matches the exact or partial destination port.
| Matches if the value matches the Description of the Signature determined by Application Control Lite
| deep sessions
| False
|-  
|-  
| Signature
| Web Filter: Category
| Text
| [[Glob Matcher]]
| Matches if value matches the exact or any part of the entire signature.
| Matches if the value matches the Category determined by Web Filter
| deep sessions
| False
|-  
|-  
| Custom
| Web Filter: Category Description
| Boolean
| [[Glob Matcher]]
| Matches if value is a custom signature.
| Matches if the value matches the Description of the Category determined by Web Filter
| deep sessions
| False
|-  
|-  
| Recommended Action
| Web Filter: Site is Flagged
| Select
| boolean
| Matches if value is a signature's recommended action.
| Matches if the latest request in this session was flagged by Web Filter
| deep sessions
| False
|-
| Directory Connector: User in Group
| [[Group Matcher]]
| Matches if the username associated with the client in the Host Table is in the specified group(s)
| all
| False
|-
| SSL Inspector: SNI Host Name
| [[Glob Matcher]]
| Matches if the value matches the Server Name Indication (SNI) host name included by the client in the initial session request.
| deep sessions
| False
|-
| SSL Inspector: Certificate Subject
| [[Glob Matcher]]
| Matches if the value matches the Subject DN in the SSL certificate received from the external server.
| all
| False
|-
| SSL Inspector: Certificate Issuer
| [[Glob Matcher]]
| Matches if the value matches the Issuer DN in the SSL certificate received from the external server.
| all
| False
|-
| Time of Day
| [[Time of Day Matcher]]
| Matches times of day.
| all
| True
|-
| Day of Week
| [[Day of Week Matcher]]
| Matches days of the week.
| all
| True
|-
| Remote Host Country
| [[Glob Matcher]]
| Matches if value matches the country associated with the remote IP.
| all
| False
|-
| Client Country
| [[Glob Matcher]]
| Matches if value matches the country associated with the Client IP.
| all
| False
|-
| Server Country
| [[Glob Matcher]]
| Matches if value matches the country associated with the Server IP.
| all
| False
|-  
|-  
| System Memory
| Numeric
| Matches if system memory matches this value.
|}
|}


= Order =


* Recommended
As discussed above, the order of rules is very important. Often users want to do very complex tasks that can be difficult or impossible with a single rule.
* Enable Log
In these cases, multiple rules is useful. For example, assume you want to use Firewall to block all traffic to a server (1.2.3.4) except port 80 traffic or from the admin IP (192.168.1.100) to RDP (port 3389).
* Enable Block if Recommended is Log
* Emable Block
* Disable
 
 
 
 
Simply uncheck '''Block''' (and '''Log''' if you wish) and the the traffic will no longer be blocked.
 
 
 
 
=== Signatures ===
 
Intrusion Prevention provides a list of [[#Learning More About Signature ID Rules|signatures]] that you can have Untangle '''Log''' or '''Block''' when traffic matches them. The rules are grouped by classtype and can be searched using the search field at the bottom of the page.  
 
In most cases, you do not need to change the recommended settings. You should only need to disable a signature if that rule blocks traffic from a unique software application that you must use. CREATE RULES.
 
The signatures are automatically updated using the latest Suricata signatures.
 
*SID: The signature's identifier.
*Classtype: Suricata classtype (grouping) of the signature.  
*Category: Suricata category (grouping) for the signature.
*Msg: Name of the signature.
*Reference: Links to reference information on the attack the signature will detect (if available).
*Log/Block: Enable these to log or block traffic matching the signature.
 
*Edit: Modify a a custom signature from the system.  
*Copy: Copy a signature. Copied signatures become part of the custom set.
*Delete: Delete a custom signature from the system.
 
Using the Add button, you can also add your own custom signatures to the system. This should only be attempted by advanced users with a strong knowledge of Suricata signature creation. Adding invalid or poorly written rules will negatively impact network performance.
 
''Signatures'' match malicious network traffic patterns and perform actions such as logging without affecting the session or blocking the session.
 
 
{{AppScreenshot|intrusion-prevention|signatures}}
 
=== Variables ===
 
This tab provides administrators access to Suricata variables. These variables are used in rules to specify criteria for the source and destination of a packet.
 
Suricata's most important variable is $HOME_NET. $HOME_NET defines the network or networks you are trying to protect - it is computer automatically based on your network configuration - it includes all local networks (including aliases).
 
Using the Add button, custom variables can be added. Adding variables may be used by users adding their own rules.This should only be attempted by advanced users with a strong knowledge of Suricata signature creation.
 
{{AppScreenshot|intrusion-prevention|variables}}
 
 
== Updates ==
 
Signatures are automatically updated every night.  Any rule modifications the administrator has made will remain.  New signatures are added with recommended actions.
 
== Reports ==


{{:Intrusion Prevention Reports}}
Doing this in one rule would be difficult (actually impossible in this case), but it is quite easy using several rules:
First, create a rule to just block all traffic to the server IP, 1.2.3.4. Above that rule create a rule that passes all traffic from the admin IP, 192.168.1.100, to the server IP where ''Destination Port'' == "3389", and another rule that passes all traffic to 1.2.3.4 with ''Destination Port'' == "80."
This set of three rules does exactly what we describe above and effects no other traffic.


[[Image:Rule_order.png|760px|center|Rule Anatomy]]


== All Events ==
= Common Mistakes =


This is a list of common mistakes to avoid.


== Conditions must ALL match ==


== Related Topics ==
In order for a rule to match, ALL conditions must match. In some cases you may want to add a rule that just passes all traffic to and from an IP.
Users sometimes will create a pass rule with two conditions:
* ''Destination Address'' is "1.2.3.4"
* ''Source Address'' is "1.2.3.4"


[http://en.wikipedia.org/wiki/Intrusion_prevention_system Intrusion Prevention Systems]
This rule will never match anything because it will only match traffic where the destination address is 1.2.3.4 AND the source address is 1.2.3.4 (a session destined to itself). If you want to match when the destination address is 1.2.3.4 OR the source address is 1.2.3.4, then you must create two separate rules, one for the destination address and one for the source address.


[https://suricata.readthedocs.io/en/suricata-3.2.1/rules/index.html Suricata - Writing Suricata Signatures]
== Rule Order ==


== Intrusion Prevention FAQs ==
Occasionally, you will add a rule and not understand why it is not working as intended. Often we find that there was a rule above that was matching the session before the desired rule. For example, just adding a rule to the bottom often isn't sufficient, you must find the appropriate place in your ruleset to place the new rule.


{{:Intrusion Prevention FAQs}}
To do so simply logically evaluate the rules in your head starting at the top to find the appropriate place for your rule. Also use the event log to run several tests to see how the session was handled in the event log and adjust your ruleset accordingly.

Revision as of 19:59, 13 November 2018

Rules

Rules are used frequently in Untangle and many other firewalls. Rules are very powerful, but can sometimes be difficult to configure. This documentation describes how rules work and gives some basic examples and some common mistakes to avoid. Many applications use rules like Firewall, Captive Portal, Application Control, Bandwidth Control, etc. All of these rules essentially share the same logic.

Basics

Rules are configured by the user to categorize and act upon traffic. For example, Firewall uses rules to determine whether to block or pass traffic. Bandwidth Control uses rules to determine how to prioritize a session.

Rules are evaluated in order from top to bottom against sessions (not packets!). If a rule matches then the action from that rule is performed and no more rules are evaluated. If no matching rule is found the behavior is defined by the application, which is usually doing nothing.

This is similar to other firewalls. Untangle rules are "quick" rules which means the first match is always used. Unlike some other firewalls, Untangle evaluates against sessions, not packets. This means that the "Source Address" will be the initiator of the session and the "Destination Address" will be the server address the client is connecting to, and the same is true for "Source Port" and "Destination Port."

Each rule has several properties:

  • An enable checkbox
  • A name/description
  • A set of conditions
  • An action or set of actions
Rule Anatomy

The enable checkbox determines if the rule is evaluated. If the enable checkbox is not checked, the rule is simply skipped.

The description is for the user to document what the rule does. It is highly suggested to give the rule a meaningful name. Trying to troubleshoot a set of rules all named "[no description]" can be extremely difficult.

The set of conditions is the description of the traffic that should match the rule. If all of the conditions are true for the given session then the rules matches. This is discussed in more details in the next section.

The action or set of actions configures what action is performed if the rule matches. This is dependent on the application. For example, in Firewall it determines whether to block or pass the session and additionally if it should be flagged.

Conditions

Conditions define which sessions will match the rule. If and only if all of the conditions match, the rule is considered a match. Conditions can also be inverted by selecting "is NOT" in the dropdown, effectively inverting when it matches. Destination Port "is NOT" "80" matches on all ports except port 80.

Lets take a simple example: We want to block TCP traffic to port 80 on a server. There is a web service running on that server that you don't want to allow access to. First, create a rule and check the enable checkbox and give it a reasonably descriptive name like "blocking TCP port 80 to serverX." Now you will need to add some conditions that match only the traffic you want to block. So in this example will will add:

  • Protocol is TCP
  • Destination Address is 1.2.3.4 (The IP address of the server)
  • Destination Port is 80

Finally, set the action to block and flag, and click "Done" and "Apply". Since all conditions must be true this rule will block TCP traffic to 1.2.3.4 port 80 only and nothing else.

Rule Conditions

There are many conditions available to carefully define precise sets of traffic. The following Table defines the list of conditions.

Each condition has several properties:

  • Name - The Name of the condition
  • Syntax - The accepted Syntax of the condition. If editing through the UI, some conditions have custom editors to help you craft conditions, others are just a text field.
  • Availability - The Availability is the availability of that matcher at various times. Some conditions, like Destination Address, are always known and can always be used to match. Other conditions, like Application Control: Application, are only known after the session is created and some data flows and Application Control is able to identify the application. These type conditions are 'deep session' conditions because they are not able to be evaluated at session creation time, only after some data flows. As such they are not available in rules that are evaluated at session creation time, like Firewall and Captive Portal.
  • Reliability - The Reliability is a "reliability" of that condition. True means it is 100% reliable. False means it is 99% or less reliable. For example, some conditions, like Destination Port, are always known and thus matching on Destination Port is 100% reliable. Other conditions, like Client Hostname, only match if the hostname for the client hostname is known. Hostname is determined through many ways, including DNS and DHCP. If all of these methods fail then it is entirely possible that the hostname of a server is "foo" but Untangle has not been able to determine this and as such a Client Hostname is "foo" condition will not match. This column is there purely as a warning that users in these cases must be aware of when conditions might deliver false negatives.

Condition List

Name Syntax Function Availability Reliability
Destination Address IP Matcher Matches if value matches the Destination/Server Address of the session. (after NAT/port forwarding) all True
Destination Port Int Matcher Matches if value matches the Destination/Server Port of the session. (after NAT/port forwarding) all True
Destination Interface checkboxes Matches if value matches the Destination/Server Interface of the session. (after routing/port forwarding) all True
Destined Local boolean Matches if the session is destined to one of Untangle's IP addresses. all True
Source Address IP Matcher Matches if value matches the Source/Client Address of the session. (before NAT/port forwarding) all True
Source Port Port Matcher Matches if value matches the Source/Client Port of the session. (before NAT/port forwarding) hidden True
Source Interface checkboxes Matches if value matches the Source/Client Interface of the session. (before routing/port forwarding) all True
Protocol checkboxes Matches if value matches the Protocol of the session. all True
Tagged Glob Matcher Matches if session is tagged with matching tag all True
Client Tagged Glob Matcher Matches if client of the session is tagged with matching tag all True
Server Tagged Glob Matcher Matches if server of the session is tagged with matching tag all True
Username User Matcher Matches if value matches the username associated with the Client IP in the Host table. all False
Host Hostname Glob Matcher Matches if value matches the hostname associated with the Local Address in the Host table. all False
Client Hostname Glob Matcher Matches if value matches the hostname associated with the Client IP in the Host table. all False
Server Hostname Glob Matcher Matches if value matches the hostname associated with the Server IP in the Host table. all False
Client MAC Address Glob Matcher Matches if value matches the MAC address associated with the Client IP in the ARP table. all False
Server MAC Address Glob Matcher Matches if value matches the MAC address associated with the Server IP in the ARP table. all False
Client MAC Vendor Glob Matcher Matches if value matches the device manufacturer. This is identified using the Client IP's MAC address in the ARP table and OUI Lookup. all False
Server MAC Vendor Glob Matcher Matches if value matches the device manufacturer. This is identified using the Server IP's MAC address in the ARP table and OUI Lookup. all False
Host has no Quota boolean Matches if the local address has no quota all True
User has no Quota boolean Matches if the username has no quota all True
Client has no Quota boolean Matches if the client has no quota hidden True
Server has no Quota boolean Matches if the server has no quota hidden True
Host has Exceeded Quota boolean Matches if the local address has exceeded their quota all True
User has Exceeded Quota boolean Matches if the user has exceeded their quota all True
Client has Exceeded Quota boolean Matches if the client has exceeded their quota hidden True
Server has Exceeded Quota boolean Matches if the server has exceeded their quota hidden True
Host Quota Attainment Int Matcher Matches the local address quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc) all True
User Quota Attainment Int Matcher Matches the username quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc) all True
Client Quota Attainment Int Matcher Matches the client quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used, etc) hidden True
Server Quota Attainment Int Matcher Matches the server quota used to quota size ratio (">1" means over quota, ">.5" means over 50% used. etc) hidden True
HTTP: Hostname Glob Matcher Matches if the value matches the hostname specified in the HTTP session deep sessions False
HTTP: Referer Glob Matcher Matches if the value matches the referer string specified in the HTTP header deep sessions False
HTTP: URI Glob Matcher Matches if the value matches the latest URI specified in the HTTP session deep sessions False
HTTP: URL URL Matcher Matches if the value matches the latest URL (hostname+URI) specified in the HTTP session deep sessions False
HTTP: Content Type Glob Matcher Matches the content-type of the latest content in the HTTP session deep sessions False
HTTP: Content Length Int Matcher Matches if Int Matcher matches the content length specified in the latest HTTP response deep sessions False
HTTP: Request Method Glob Matcher Matches if the value matches the HTTP request method of the requested URL. Standard request methods include GET, POST, HEAD, OPTIONS, PUT, DELETE, TRACE, and CONNECT. deep sessions False
HTTP: Request File Path Glob Matcher Matches if the value matches the entire file path of the requested URL. This is everything including and after the first slash character following the host name or address. (e.g. /some/location/mypage.html) deep sessions False
HTTP: Request File Name Glob Matcher Matches if the value matches the file name of the requested URL. This is everything after the final slash character of the request. (e.g. mypage.html) deep sessions False
HTTP: Request File Extension Glob Matcher Matches if the value matches the file extension of the requested URL. This is everything after following the dot after the final slash character of the request. (e.g. html) deep sessions False
HTTP: Response Content Type Glob Matcher Matches if the value matches the content or MIME type reported in the server response. deep sessions False
HTTP: Response File Name Glob Matcher Matches if the value matches the file name returned in the server response. deep sessions False
HTTP: Response File Extension Glob Matcher Matches if the value matches the file extension returned in the server response. This is everything after (but not including) the final dot of the file name. deep sessions False
HTTP: Client User Agent Glob Matcher Matches if the value matches the User Agent string for the client in the Host table all False
Application Control: Application Glob Matcher Matches if the value matches the Application determined by Application Control deep sessions False
Application Control: Category Glob Matcher Matches if the value matches the Category of the Application determined by Application Control deep sessions False
Application Control: Protochain Glob Matcher Matches if the value matches the Protochain determined by Application Control deep sessions False
Application Control: Detail Glob Matcher Matches if the value matches the Detail of the Application determined by Application Control deep sessions False
Application Control: Confidence Int Matcher Matches if Int Matcher matches the confidence rating determined by Application Control deep sessions False
Application Control: Productivity Int Matcher Matches if Int Matcher matches the productivity rating determined by Application Control deep sessions False
Application Control: Risk Int Matcher Matches if Int Matcher matches the risk rating determined by Application Control deep sessions False
Application Control Lite: Signature Glob Matcher Matches if the value matches the Signature determined by Application Control Lite deep sessions False
Application Control Lite: Category Glob Matcher Matches if the value matches the Category of the Signature determined by Application Control Lite deep sessions False
Application Control Lite: Description Glob Matcher Matches if the value matches the Description of the Signature determined by Application Control Lite deep sessions False
Web Filter: Category Glob Matcher Matches if the value matches the Category determined by Web Filter deep sessions False
Web Filter: Category Description Glob Matcher Matches if the value matches the Description of the Category determined by Web Filter deep sessions False
Web Filter: Site is Flagged boolean Matches if the latest request in this session was flagged by Web Filter deep sessions False
Directory Connector: User in Group Group Matcher Matches if the username associated with the client in the Host Table is in the specified group(s) all False
SSL Inspector: SNI Host Name Glob Matcher Matches if the value matches the Server Name Indication (SNI) host name included by the client in the initial session request. deep sessions False
SSL Inspector: Certificate Subject Glob Matcher Matches if the value matches the Subject DN in the SSL certificate received from the external server. all False
SSL Inspector: Certificate Issuer Glob Matcher Matches if the value matches the Issuer DN in the SSL certificate received from the external server. all False
Time of Day Time of Day Matcher Matches times of day. all True
Day of Week Day of Week Matcher Matches days of the week. all True
Remote Host Country Glob Matcher Matches if value matches the country associated with the remote IP. all False
Client Country Glob Matcher Matches if value matches the country associated with the Client IP. all False
Server Country Glob Matcher Matches if value matches the country associated with the Server IP. all False

Order

As discussed above, the order of rules is very important. Often users want to do very complex tasks that can be difficult or impossible with a single rule. In these cases, multiple rules is useful. For example, assume you want to use Firewall to block all traffic to a server (1.2.3.4) except port 80 traffic or from the admin IP (192.168.1.100) to RDP (port 3389).

Doing this in one rule would be difficult (actually impossible in this case), but it is quite easy using several rules: First, create a rule to just block all traffic to the server IP, 1.2.3.4. Above that rule create a rule that passes all traffic from the admin IP, 192.168.1.100, to the server IP where Destination Port == "3389", and another rule that passes all traffic to 1.2.3.4 with Destination Port == "80." This set of three rules does exactly what we describe above and effects no other traffic.

Rule Anatomy
Rule Anatomy

Common Mistakes

This is a list of common mistakes to avoid.

Conditions must ALL match

In order for a rule to match, ALL conditions must match. In some cases you may want to add a rule that just passes all traffic to and from an IP. Users sometimes will create a pass rule with two conditions:

  • Destination Address is "1.2.3.4"
  • Source Address is "1.2.3.4"

This rule will never match anything because it will only match traffic where the destination address is 1.2.3.4 AND the source address is 1.2.3.4 (a session destined to itself). If you want to match when the destination address is 1.2.3.4 OR the source address is 1.2.3.4, then you must create two separate rules, one for the destination address and one for the source address.

Rule Order

Occasionally, you will add a rule and not understand why it is not working as intended. Often we find that there was a rule above that was matching the session before the desired rule. For example, just adding a rule to the bottom often isn't sufficient, you must find the appropriate place in your ruleset to place the new rule.

To do so simply logically evaluate the rules in your head starting at the top to find the appropriate place for your rule. Also use the event log to run several tests to see how the session was handled in the event log and adjust your ruleset accordingly.