The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
This document describes the use and configuration of redirectionless posture flow and troubleshooting tips.
Cisco recommends that you have knowledge of these topics:
For a better understanding of the concepts described later, it's recommended to go through:
Compare Earlier ISE Versions to ISE Posture Flow in ISE 2.2
ISE Session Management and Posture
The information in this document is based on these software and hardware versions:
The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, ensure that you understand the potential impact of any command.
ISE Posture flow consists of the following steps:
0. Authentication/Authorization. Generally performed right before posture flow is initiated but it can be bypassed for certain use cases such as Posture Reassessment (PRA). As authentication itself does not trigger posture discovery this is not considered essential for every posture flow.
This document focuses on the Discovery process of ISE Posture flow.
Cisco recommends to use redirection for the discovery process, however, there are certain cases where redirection is not possible to implement such as the use of third party Network Devices where redirection is not supported. This document aims to provide a general guidance and best practices to implement and troubleshoot redirectionless posture in such environments.
Full description of redirectionless flow is described in Compare Earlier ISE Versions to ISE Posture Flow in ISE 2.2.
There are two types of posture discovery probes that do not use redirection:
Connectiondata.xml is a file created and maintained automatically by Cisco Secure Client. It consists of a list of PSNs that the client has previously successfully connected to for posture, hence, this is only a local file and its content is not persistent across all endpoints.
The main purpose of connectiondata.xml is to work as a backup mechanism for both Stage 1 and Stage 2 discovery probes. In case that redirection or Call Home List probes are unable to find a PSN with an active session, Cisco Secure Client sends a direct request to each of the servers listed in connectiondata.xml.
A common problem caused by the use of connectiondata.xml probes is an overload of the ISE deployment due to a large number of HTTPS requests sent by the endpoints. It is important to consider that while connectiondata.xml is effective as a backup mechanism to avoid full outages for both redirection and redirectionless posture mechanisms, it is not a sustainable solution for a posture environment, therefore, it is necessary to diagnose and resolve the design and configuration problems that cause the failure of the main discovery probes and that result in discovery issues.
Call Home List is a section of the posture profile where a list of PSNs is specified to be used for posture. Unlike connectiondata.xml, this is created and maintained by an ISE administrator and might require a design phase for optimal configuration. The list of PSNs in Call Home List should match the list of authentication and accounting servers that is configured in the network device or load balancer for RADIUS.
Call Home List probes enable the use of an MnT lookup during active session search in case of a local lookup failure in a PSN. The same functionality extends to connectiondata.xml probes only when they are used during Stage 2 discovery. For this reason, all Stage 2 probes are also referred to as New Generation probes.
As a redirectionless discovery process often entails a more complex flow and a larger amount of processing on PSNs and MnT compared to a redirection flow, there are two common challenges that can arise during implementation:
In order to cope with these challenges, it is recommended to design the Call Home List to limit the number of PSNs that a given endpoint can use for posture. For medium and large deployments, it is necessary to distribute the deployment in order to create multiple Call Home Lists with reduced number of PSNs, in consequence the list of PSNs that are used for RADIUS authentication for a given Network Device should be limited in the same way to match the corresponding Call Home List.
The following aspects can be taken into consideration while developing the PSN distribution strategy to determine the maximum number of PSNs in each Call Home List:
Tip: Use Network Device Groups to classify the network devices according to the design.
Network Device Groups can be used to identify and match network devices with their corresponding RADIUS server list and Call Home List. In the case of hybrid environments, they can also be used to identify devices that support redirection from devices that don’t.
If the distribution strategy developed during design phase relies on Network Device Groups, follow the next steps to configure them on ISE:
In the examples used throughout this guide, Location Device Group is used to identify the RADIUS servers list and Call Home List, and a custom Posture Device Group is used to identify Redirection from Redirectionless posture devices.
There are two ways to provision the client with the right software and profile to perform posture in a redirectionless environment:
Note: Refer to step 4 of the Client provisioning policy section for instructions on how to verify the Client Provisioning Portal port if necessary.
Warning: Make sure that the same Cisco Secure Client files are also on the headends you plan to connect to: Secure Firewall ASA, ISE, and so on. Even when manual provisioning is used, ISE must be configured for client provisioning with the corresponding software version. Refer to the Client provisioning policy configuration section for detailed instructions.
Tip: Install the Diagnostic and Reporting Tool to be used for troubleshooting purposes.
ISE Client Provisioning Portal can be used to install Cisco Secure Client ISE Posture module and the posture profile from ISE, it can also be used to push the posture profile alone if ISE Posture module is already installed on the client.
Note: To make use of the portal FQDN, clients must have the PSN Admin certificate chain as well as the Portal certificate chain installed in the trusted store, and the Admin certificate must contain the portal FQDN in the SAN field.
Client provisioning must be configured on ISE regardless of the type of provisioning (pre-deploy or web deploy) that is used to install Cisco Secure Client on the endpoints.
To find the port that should be used in Call Home List, navigate to Work Centers > Posture > Client Provisioning > Client Provisioning Portal, select the portal in use and expand Portal Settings.
Warning: If Cisco Secure Client has been pre deployed to the clients, make sure that the version on ISE matches the version on the endpoints. If ASA or FTD is used for web deploy, the version on this device should match as well.
Note: In the case of multiple Call Home Lists, use the Other Conditions field to push the right profile to the corresponding clients. In the example, Device Location Group is used to identify the posture profile that is pushed in the policy.
Tip: If multiple client provisioning policies are configured for the same OS, it is recommended to make them mutually exclusive, that is, a given client should only be able to hit one policy at a time. RADIUS attributes can be used under Other Conditions column to differentiate one policy from another.
permit udp any any eq domain
permit udp any any eq bootps
permit ip any host <PSN 1>
permit ip any host <PSN 2>
deny ip any any
Caution: Some third party devices might not support DACLs, in such cases it is necessary to use a Filter-ID or other vendor specific attributes. Refer to the vendor documentation for more information. If DACLs are not used, make sure to configure the corresponding ACL in the network device.
Note: If DACLs are not used, use Filter-ID from Common Tasks or the Advanced Attribute Settings to push the corresponding ACL name.
The presence of stale or phantom sessions in the deployment can generate intermittent and apparently random failures with redirectionless posture discovery which result in users being stuck in a posture unknown/not applicable access on ISE while Cisco Secure Client UI shows Compliant access.
Stale sessions are old sessions that are no longer active. They are created by an authentication request and accounting start but no accounting stop is received on the PSN to clear the session.
Phantom sessions are sessions that was never actually active in a particular PSN. They are created by an accounting interim update but no accounting stop is received on the PSN to clear the session.
To identify a stale/phantom session issue, verify the PSN used in system scan on the client and compare with the PSN performing the authentication:
ISE versions above ISE 2.6 patch 6 and 2.7 patch 3 implement RADIUS Session Directory as a solution for stale/phantom session scenario in re-directionless posture flow.
Note: This service refers to the communication method that is used for RSD between PSNs and should be running regardless of the status of the ISE Messaging Service setting for syslog that can be set from ISE UI.
Note: It is expected to observe Queue Link Error alarms with cause Unknown CA or Econnrefused while the certificates are regenerated, monitor the alarms after certificate generation to confirm that the issue is resolved.
Performance issues such as high CPU utilization and high load average related to redirectionless posture can impact PSN as well as MnT nodes and are often accompanied or preceded by the following events:
If performance of the deployment is impacted by redirectionless posture, this is often indicative of an ineffective implementation. It is recommended to revise the following aspects:
To mitigate the impact:
RADIUS accounting is essential for session management on ISE. Since posture relies on an active session to be performed, incorrect or lack of accounting configuration can also impact posture discovery and ISE performance. It is important to verify that accounting is correctly configured on the network device to send authentications requests, accounting start, accounting stop and accouting updates to a single PSN for each session.
To verify accounting packets received on ISE, navigate to Operations > Reports > Reports > Endpoints and Users > RADIUS Accounting.
Revision | Publish Date | Comments |
---|---|---|
1.0 |
24-Jul-2023 |
Initial Release |