The configuration snippet descriptions and the associated GitHub repository link for each xml snippet.
Note
The template version is found in the template xml file as a tag attribute
Note
The set commands utilize the same configuration settings
Starting with IronSkillet 10.1, the xml snippets are included in sub-skillets in the ironskillet-components submodule. These sub-skillets are referenced in playlists through skillet includes, and allows for easy re-use of individual snippets. It also allows for subsets of the configuration to be run, as reflected in the new playlist options for panos and panorama configuration. An overview of the IronSkillet playlists available can be found in the playlist folder in the IronSkillet repo.
See the Playlist Includes tutorial in the SkilletBuilder documentation for more information on how this works.
This section provides templated configurations for general device settings.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Management configuration superuser access
- Administrative user name
- Password hash stored in the configuration file
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Administrative user password complexity profile
- Attributes including minimum length, characters, and history
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
General device settings that effect security posture. Found in Device > Setup in the GUI.
Wildfire: set optimal file size limits for Wildfire uploads and show verdict responses for grayware, malware and phishing
Session rematch: the firewall will go through all the existing sessions and apply the new security policy to any matching traffic
Notify User: user should be notified when web-application is blocked; enables the application response page
Log Suppression: disabled to ensure unique log entries even if similar session types
Prevent TCP and UDP buffer overflow and multi-part HTTP download evasions
- Disable 'tcp-bypass-exceed-queue'
- Disable 'udp-bypass-exceed-queue'
Enable high DP load logging
Prevent App-ID buffer overflow evasion
- set bypass-exceed-queue to 'no'
Prevent TCP and MPTCP evasions
- set urgent data to 'clear'
- set drop zero flag to 'yes'
- set bypass-exceed-oo-queue to 'no'
- set check-timestamp-option to 'yes'
- set strip-mptcp-option to yes
Set an API key lifetime instead of a permanent/static value
- default set to 525,600 minutes (1 year)
set export of csv log file to maximum of 1,048,576
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
View dns xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
View mgmt IP config xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
System configuration settings for dynamic updates and network services (eg. DNS, NTP).
Update schedule settings
- Check every 30 minutes for new threat signatures
- Hourly checks for new AV signatures
- Check realtime for new Wildfire signatures
- Recommended time delays and thresholds for checks and installs
Use SNMPv3
Set default DNS and NTP values
Set timezone to UTC
Provide a standard login banner warning for unauthorized users
Note
The management config types include static or dhcp-client.
This is specific to each deployment and can be selected as part of the tools to build `loadable_configs`
.
Since management interface is in the template config, this option must be included for deployment.
Logging best practice configurations for logging output and forwarding profiles.
Warning
Configure logging profiles before security rules The template creates a log forwarding profile call default. This profile is referenced in the template security rules and should be configured before the security rules.
Note
Logging can be deployment dependent The destination in the logging profile is templated to an unroutable syslog server address. This can vary based on actual deployment scenarios.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
View email xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Log forward profile referenced in security rules to determine where to forward log related events.
- Forward all log activity to syslog (see the reference syslog configuration in shared_log_settings.xml)
- Email malicious and phishing Wildfire verdicts to the address in the email profile (see shared_log_settings.xml)
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
View email profile xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
View email system critical xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Device event logging including sample profiles for email and syslog forwarding.
- Reference syslog profile that can be edited for a specific IP address and UDP/TCP port
- Reference email profile that can be edited for specific email domain and user information
- System, configuration, user, HIP, and correlation log forwarding to syslog
- Email critical system events to the email profile
Note
When to use email alerts The purpose of select email alert forwarding is ensure not to under alert or over alert yet provide critical messages for key events. Under alerting reduces visibility to key events while over alerting creates too much noise in the system. The templates are set with a median view to capture key events without too much 'log fatigue' noise
Address, External Dynamic List (EDL), and tag objects that are referenced in security rules by name.
View xml snippet: | 9.0 | 9.0 | 10.0 | 10.1 |
Tags used in security rules and related objects.
- Inbound - inbound (untrust to trust) elements
- Outbound - outbound (trust to untrust) elements
- Internal - internal (trust) segmentation elements
Tag showing IronSkillet loaded and the associated template version.
The key elements for security posture are security profiles and the security rules. The templates ensure best practice profiles and profile groups are available and can be referenced in any security rules. The template security rules focus on 'top of the list' block rules to reduce the attack surface.
Warning
Profiles and subscriptions All of the template security profiles other than file blocking require Threat Prevention, URL Filtering, and Wildfire subscriptions. Ensure that the device is properly licensed before applying these configurations.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Placeholder for custom url categories used in security rules and url profiles. Using these categories prevents the need to modify the default template.
- Block: placeholder to be used in block rules and objects to override default template behavior
- Allow: placeholder to be used in permit rules and objects to override default template behavior
- Custom-No-Decrypt: to be used in the decryption no-decrypt rule to specify URLs that should not be decrypted
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile for actions specific to file blocking (FB).
Note
File blocking and file types The Block file type recommendation is based on common malicious file types with minimal impact in a Day 1 deployment. Although PE is considered the highest risk file type it is also used for legitimate purposes so blocking PE files will be deployment specific and not included in the template.
- Day 1 Block file types: 7z, bat, chm, class, cpl, dll, hlp, hta, jar, ocx, pif, scr, torrent, vbe, wsf
- The profiles will alert on all other file types for logging purposes
Profiles:
- Outbound-FB: For outbound (trust to untrust) security rules
- Inbound-FB: For inbound (untrust to trust) security rules
- Internal-FB: For internal network segmentation rules
- Alert-Only-FB: No file blocking, only alerts for logging purposes
- Exception-FB: For exception requirements in security rules to avoid modifying the default template profiles
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile for actions specific to anti-spyware (AS).
Note
Sinkhole addresses The profiles use IPv4 and IPv6 addresses for DNS sinkholes. IPv4 is currently provided by Palo Alto Networks. IPv6 is a bogon address. In 9.0 the IPv4 address is replaced by an FQDN
[9.x] Support for DNS Cloud subscription service
- In addition to the current malicious domain push to the device, also include domain lookups using the cloud service
[10.x] Support for DNS Cloud subscription domain categories and actions
- set malicious categories to sinkhole
Profiles:
Outbound-AS : For outbound (trust to untrust) security rules
- Block severity = Critical, High, Medium
- Default severity = Low, Informational
- DNS Sinkhole for IPv4 and IPv6
- Single packet capture for Critical, High, Medium severity
Inbound-AS : For inbound (untrust to trust) security rules
- Block severity = Critical, High, Medium
- Default severity = Low, Informational
- DNS Sinkhole for IPv4 and IPv6
- Single packet capture for Critical, High, Medium severity
Internal-AS : For internal network segmentation rules
- Block severity = Critical, High
- Default severity = Medium, Low, Informational
- DNS Sinkhole for IPv4 and IPv6
- Single packet capture for Critical, High, Medium severity
Alert-Only-AS : No blocking, only alerts for logging purposes
- Alert all severities and malicious domain events
- No packet capture
Exception-AS : For exception requirements in security rules to avoid modifying the default template profiles
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile for actions specific to URL filtering (URL).
Note
Only BLOCK
categories will be listed for each profile below.
All other URL categories will be set to ALERT
in the templates for logging
purposes. The complete list of categories can be found in the url filtering template.
[10.x] Support for local machine learning based on web content
- block malicious content using dynamic classification
Profiles:
Outbound-URL : For outbound (trust to untrust) security rules
- URL Categories
- Site Access: Block command-and-control, malware, phishing, Black List (custom URL category)
- User Credential Submission: Block all categories
- Alert category = includes White List (custom URL category)
- URL Filtering Settings: HTTP Header Logging (user agent, referer, X -Forwarded-For)
- dynamic classification to block malicious web conent
Alert-Only-URL : No blocking, only alerts for logging purposes
- Alert all categories including custom categories Black List and White List
Exception-URL : For exception requirements in security rules to avoid modifying the default template profiles
- URL Categories
- Site Access: Block command-and-control, malware, phishing, Black List (custom URL category)
- User Credential Submission: Block all categories
- Alert category = includes White List (custom URL category)
- URL Filtering Settings: HTTP Header Logging (user agent, referer, X -Forwarded-For)
- dynamic classification to block malicious web conent
Note
9.0 includes new URL categories for risk and newly created domains. In future best practices, these categories may be used to provide additional security protections when combined with existing URL categories. For now, these categories are only set to alert.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile for actions specific to AntiVirus (AV) and Wildfire signatures. All decoders using 'reset-both' as actions except for the Alert-Only profile.
[10.x] Support for WF-based local machine learning to block malicious content for exe and powershell files.
Profiles:
- Outbound-AV: For outbound (trust to untrust) security rules
- Inbound-AV: For inbound (untrust to trust) security rules
- Internal-AV: For internal network segmentation rules
- Alert-Only-AV: No blocking, only alerts for logging purposes
- Exception-AV: For exception requirements in security rules to avoid modifying the default template profiles
Note
Email response codes with SMTP not IMAP or POP3 Reset-both is used for SMTP, IMAP, and POP3. SMTP '541' response messages are returned to notify that the session was blocked. IMAP and POP3 do not have the same response model. In live deployments, instead of DoS concerns with retries, the endpoints typically stop resending after a small number of sends with timeouts.
Note
9.0 includes support for http/2. If you are upgrading from a previous version ensure that this decoder matches the actions for standard http.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Profiles:
Outbound-VP : For outbound (trust to untrust) security rules
- Block severity = Critical, High, Medium
- Alert severity = Low, Informational
- Single packet capture for Critical, High, Medium severity
Inbound-VP : For inbound (untrust to trust) security rules
- Block severity = Critical, High, Medium
- Alert severity = Low, Informational
- Single packet capture for Critical, High, Medium severity
Internal-VP : For internal network segmentation rules
- Block severity = Critical, High
- Alert severity = Medium, Low, Informational
- Single packet capture for Critical, High, Medium severity
Alert-Only-VP : No blocking, only alerts for logging purposes
- Alert all severities
- No packet capture
Exception-VP: For exception requirements in security rules to avoid modifying the default template profiles
Note
A separate branch is being used as a placeholder for Brute-Force-Exceptions. This provides a way to include Support recommended exceptions by ThreatID value. These can be loaded using console SET commands or using API-based tools
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile for actions specific to Wildfire upload and analysis (WF).
Note
Public Cloud
is the default
All template profiles are configured to upload all file types in any
direction to the public cloud for analysis.
Profiles:
- Outbound-WF: For outbound (trust to untrust) security rules
- Inbound-WF: For inbound (untrust to trust) security rules
- Internal-WF: For internal network segmentation rules
- Alert-Only-WF: No blocking, only alerts for logging purposes
- Exception-WF: For exception requirements in security rules to avoid modifying the default template profiles
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Security profile groups based on use case
- Inbound: For rules associated to inbound (untrust to trust) sessions
- Outbound: For rules associated to outbound (trust to untrust) sessions
- Internal: For rules associated to trust-domain network segmentation
- Alert Only: Provides visibility and logging without a blocking posture
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Recommended block rules for optimal security posture with associated default log-forwarding profile
- Outbound Block Rule: Block destination IP address match based on the Palo Alto Networks predefined externals dynamic lists
- Inbound Block Rule: Block source IP address match based on the Palo Alto Networks predefined externals dynamic lists
Note
Security rules in the template are block only The template only uses block rules. Allow rules are zone, direction and use case dependent. Additional templating work will provide recommended use case case security rules.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Configuration for the default interzone and intrazone default rules
Intrazone
- Enable logging at session-end using the default logging profile
- Use the Internal security profile-group
Interzone
- Explicit drop of traffic between zones
- Enable logging at session-end using the default logging profile
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Recommended_Decryption_Profile. Referenced by the default decryption rule.
SSL Forward Proxy
- Server Cert Verification : Block sessions with expired certs, Block sessions with untrusted issuers, Block sessions with unknown cert status
- Unsupported Mode Checks : Block sessions with unsupported versions, Blocks sessions with unsupported cipher suites
SSL No Proxy
- Server Cert Verification : Block sessions with expired certs, Block sessions with untrusted issuers
SSH Proxy
- Unsupported Mode Checks : Block sessions with unsupported versions, Block sessions with unsupported algorithms
SSL Protocol Settings:
- Minimum Version: TLSv1.2; Max version TLSv1.3; Any TLSv1.1 errors can help find outdated TLS endpoints
- Key Exchange Algorithms: RSA not recommended and unchecked
- Encryption Algorithms: 3DES and RC4 not recommended and unavailable when TLSv1.2 is the min version
- Authentication Algorithms:MD5 not recommended and unavailable when TLSv1.2 is the min version
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Recommended SSL decryption pre-rules for no-decryption.
- NO decrypt rule for select URL categories; Initially disabled in the Day 1 template until SSL decryption to be enabled
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Recommended_Zone_Protection profile for standard, non-volumetric best practices. This profile should be attached to all interfaces within the network.
Note
Recon Protection Default values enabled in alert-only mode; active blocking posture requires network tuning
Packet Based Attack Protection
- IP Drop: Spoofed IP Address, Malformed
- TCP Drop: Remove TCP timestamp, No TCP Fast Open, Multipath TCP (MPTCP) Options = Global
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Series of reports to look for traffic anomalies, where to apply or remove rules, etc. Reports are grouped by topic per the report group section below.
Note
Zones and Subnets in report queries The repo contains a separate folder for custom reports that use a placeholder zone called 'internet' for match conditions in reports. This value MUST be changed to match the actual public zone used in a live network. Additional zones and/or subnets to be used or excluded in the reports would be added in the query values.
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Report groups allow you to create sets of reports that the system can compile and send as a single aggregate PDF report with an optional title page and all the constituent reports included.
Template report groups include:
Simple (included in Day One template)
- Possible Compromise: malicious sites and verdicts, sinkhole sessions
Custom
- User Group Activity (eg. Employee, Student, Teacher): user-id centric reports grouped by user type
- Inbound/Outbound/Internal Rule Tuning: Used rules, app ports, unknown apps, geo information
- Inbound/Outbound/Internal Threat Tuning: Allowed threats traversing the device
- File Blocking Tuning: View of upload/download files and types with associated rule
- URL Tuning: Views by categories, especially questionable and unknown categories
- Inbound/Outbound/Internal Threats Blocked: Threat reports specific to blocking posture; complement to threat tuning
- Non-Working Traffic: View of dropped, incomplete, or insufficient data sessions
View xml snippet: | 9.0 | 9.1 | 10.0 | 10.1 |
Schedule and email recipients for each report group. The template uses a sample email profile configured in shared_log_settings.