Investigating Slow RADIUS Proxy Request

[Vedsar Kushwaha]

I have a few questions about how the FreeRADIUS proxy handles requests:


  1.  When proxying a request, does a FreeRADIUS thread remain blocked until a response is received from the home server?
  2.  Does the max_outstanding flag apply only to health status checks, or does it affect all requests proxied to the home server?

     *   If a FreeRADIUS thread blocks while waiting for a response, does increasing max_outstanding to a high value like 65536 actually allow FreeRADIUS to send that many outstanding requests? Or is the number of threads the limiting factor?
     *   If a proxy request doesn't block a thread, how does FreeRADIUS listen for responses from the home server?

  1.  We have a request that took 40 minutes to proxy. The snippet below, along with the complete log, shows the details. We're trying to investigate the cause of this delay.

```
2024-09-17T13:05:52.495Z (44469) Starting proxy to home server 129.170.130.4 port 2083
2024-09-17T13:05:52.495Z (44469) server VS_P_790be815-af8f-4dc5-b8c2-e01b8a107db4 {
2024-09-17T13:05:52.495Z (44469) }
2024-09-17T13:46:55.967Z (44469) Proxying request to home server 129.170.130.4 port 2083 (TLS) timeout 30.000000
```



Juniper Business Use Only
-------------- next part --------------
A non-text attachment was scrubbed...
Name: r_44469
Type: application/octet-stream
Size: 10374 bytes
Desc: r_44469
URL: <http://lists.freeradius.org/pipermail/freeradius-users/attachments/20240925/b3f02597/attachment-0001.obj>