Security

last person joined: 21 hours ago 

Forum to discuss Enterprise security using HPE Aruba Networking NAC solutions (ClearPass), Introspect, VIA, 360 Security Exchange, Extensions, and Policy Enforcement Firewall (PEF).
Expand all | Collapse all

Session timeout issues in CPPM 6.8.2

This thread has been viewed 12 times
  • 1.  Session timeout issues in CPPM 6.8.2

    Posted Sep 09, 2019 05:42 AM
      |   view attached

    Seem to be having Session-Timeout issues on CPPM 6.8.2.... more specifically whats displayed in Access Tracker.

    I tend to send back an access accept packet for real breathing users with a Session Timeout of 3600 secs and its been like that for a long time.. Works just fine in 6.8.0 and I think in 6.8.1 ( TBH never checked).  However, I recently upgraded our Building ( i.e. my dev server) to 6.8.2 and now I'm seeing Session-Timeout values in access tracker that are NOT 3600, even though the configs say thats what I set up.

    As an example, my EAP-TLS session for my iPhone is set to 7200 seconds ... indicated by the Enforcement profile shown in the attached image. My phone hasn't reauth'd so am thinking it might be an Access-Tracker isssue rather than actually auth'ing every 120 seconds

     

    Anyone else seen this ?

     

    A

     

     

     



  • 2.  RE: Session timeout issues in CPPM 6.8.2

    EMPLOYEE
    Posted Sep 09, 2019 06:12 AM

    Are you saying ClearPass is sending incorrect session timeout? 

     

    If so please share sshots of your Enforcement Profile. For faster resolution work with TAC.



  • 3.  RE: Session timeout issues in CPPM 6.8.2

    Posted Sep 09, 2019 07:10 AM

    I'm saying clearpas is *showing* a Session timeout value in the Access Tracker output tab that doesn't match up with what is defined in the enforcement profile.

    Also, in terms of reauths, things seem to be o.k



  • 4.  RE: Session timeout issues in CPPM 6.8.2

    EMPLOYEE
    Posted Sep 09, 2019 08:43 AM

    Sorry but i think there is a config issue here if you are saying that the Enforcment Profile sent = X and X has a session timeout of Y seconds but you are seeing the value being sent as Z. It is highly unlikely. 

     

    Please work with TAC or share screenshots of your policies and profiles for troubleshooting.



  • 5.  RE: Session timeout issues in CPPM 6.8.2

    Posted Sep 09, 2019 10:51 AM

    Our eduroam service (for my EAP-TLS iPhone) uses the following Enforcement profiles

     

     

    UoY Wireless Allow Access Profile - 130318

     

    1.

    Radius:IETF

    Acct-Interim-Interval

    =

    900

    2.

    Radius:IETF

    Termination-Action

    =

    RADIUS-Request (1)

     

     

    UoY Session Timeout - 7200

     

    Radius:IETF

    Session-Timeout

    =

    7200

     

    [Update Endpoint Known]

     

    1.

    Status-Update

    Endpoint

    =

    Known

     

     

    Which gives the following in the Access-Tracker output

     

    RADIUS Response

    Radius:IETF:Acct-Interim-Interval

    900

    Radius:IETF:Filter-Id

    airgroup_devices

    Radius:IETF:Session-Timeout

    180

    Radius:IETF:Termination-Action

    1

    Status-Update:Endpoint

    Known

    Note that the Session-Timeout  swtting is 180 and not 7200 as defined in the profile. On our production 6.8.0 setup with the same config, the Value displayed  for Session-Timeout is a correct 7200

     

    Rgds

    Alex

     

     



  • 6.  RE: Session timeout issues in CPPM 6.8.2

    Posted Sep 09, 2019 11:11 AM

    However, Some services do display the correct session-timeout ... :-(



  • 7.  RE: Session timeout issues in CPPM 6.8.2

    EMPLOYEE
    Posted Sep 10, 2019 05:26 AM

    Have never come across something like this. It is not possible to troubleshoot this here. Please work with TAC.

     

    Ensure the Enforcement Profile being sent within the service has a static value and no getting an expiration time using SQL based source.



  • 8.  RE: Session timeout issues in CPPM 6.8.2

    Posted Sep 11, 2019 05:48 AM

    Having just upgrqadedf our production cppm system to 6.8.2 from 6.8.0 everything is working on that one just fine. Methinks I'll try the old switch it off and on approach for our building/dev cppm cluster :-)