On Nov 20, 2020, at 12:07 PM, Nicolás Ciuffolotti <nicolas.ciuffolotti@cotesma.com.ar> wrote:
Sorry my mistake, I did not pass all the config
We generally don't need to see the configuration. We need to see the debug output. We also need people to pay careful attention to what they're doing. In this case, you were proxying to a realm that didn't exist. Instead of checking proxy.conf, you just gave up and asked the list what was wrong. That means you want *us* to do work you're not willing to do. This is not polite.
copy-acct-to-home-server file:
update control { &Proxy-To-Realm :="elektra" }
Yes... proxying to a KNOWN REALM is a good idea.
Just I need to configure that a server (primary, secondary) to send copies of the accounting information to each other. I understand the copy-acct-to-home-server does this, but I have issues
Proxy: Marking home server 192.168.9.20 port 1812 as zombie (it has not responded in 30.000000 seconds). ERROR: (11) Failing proxied request for user "xxx", due to lack of any response from home server 192.168.9.20 port 1812 ERROR: (11) Failed to find live home server: Cancelling proxy Debug: (11) detail (/var/log/radius/radacct/copy/detail-*): No response to request. Will retry in 30 seconds
What part of that is unclear? The home server isn't responding. How do you expect *us* to fix that issue? What can *we* do to make the home server respond? What needs to happen is that YOU need to track down why the home server isn't responding. This involves you doing work, and understanding the pieces involved. i.e. where packets go, what the RADIUS server does, etc. It could even involve looking at the DEBUG OUTPUT on the home server to see if it receives packets, and what it does with them. Alan DeKok.