A little more info. Looks like maybe an SNMP issue with checkrad?
I modified checkrad to log what it was called as from radiusd, so that I
could recreate what happened when it crashed the server. When I say
"(etc)" below, it's shorthand for the arguments that radiusd called
checkrad with.
Running 'checkrad cisco (etc)' on a certain of our ciscoes came back with
"Timeout: No Response from IP address". When called from radiusd, this
killed the radius daemon completely.
Running 'checkrad livingston (etc)' on a portmaster came back with:
SNMP Error:
no response received
SNMPv1_Session (remote host: "10.x.x.x" [10.x.x.x].161)
community: "public"
request ID: -361687079
PDU bufsize: 8000 bytes
timeout: 2s
retries: 5
backoff: 1)
at /usr/local/freeradius-0.8.1/sbin/checkrad line 217
checkrad: No SNMP answer from livingston.
When called from radiusd, this also killed the server with "Terminated."
Can't find any core files.
If this is already documented, please let me know and I'll look it up.
Otherwise, it seems to be fairly easily reproducible, so if anyone would
like more info, I'd be glad to give it up. :-)
K.
-
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
