Radius Server Dead

[WMA Hell]

I get these periodically during the day on the router syslog. 

Feb 16 19:33:22.817 EDT: %RADIUS-4-RADIUS_DEAD: RADIUS server x.x.x.x:1645,1646 is not responding.
Feb 16 19:33:22.817 EDT: %RADIUS-4-RADIUS_ALIVE: RADIUS server x.x.x.x:1645,1646 is being marked alive.

We have radius on a Windows Network Policy Server on two servers. CPU/Memory/Bandwidth are never more than 50% utilization.

Router#sh run | inc radius
ip radius source-interface GigabitEthernet0/0/0.x
radius server Salt
radius server Pepper

What does this tell me? How do I fix it?

[marce1000]

  - Check if the router has a solid network  connection to the radius server.
     + Check the radius server's logs when that happens . 

 M.

[MHM Cisco World · Accepted solution]

Sorry I was busy last days 
Now the server is detect life dead 
if we merge two issue together we get same result there is interface congestion that drop control/mgmt traffic 
show policy-map interface <<- share this if you can 

MHM

[WMA Hell]

Team

As it turns out that second Radius server was decommissioned 6 months ago. I need to rebuilt it using the same line in the configuration and the IP. 

Currently they have a Windows Radius solution. Should I just build a second Windows Radius server or leverage ISE for this?

How do you integrate two different Radius server platforms?

Matt

[marce1000]

          >...Should I just build a second Windows Radius server or leverage ISE for this?
  - You may for instance compare performance rates as mentioned in https://www.cisco.com/c/en/us/td/docs/security/ise/performance_and_scalability/b_ise_perf_and_scale.html#reference_dfk_mjh_m5b   with other radius servers 

       >...How do you integrate two different Radius server platforms?
   - Try to keep them the same ; the authenticating policies will become more  complicated for instance and
      or playing with load balance or timeout settings will loose it's efficiency or disappear ,

 M.