I have an net-snmp agent (5.2.2) that crashes (segfault) when sending a
sighup to the agent, followed by an snmp access (eg. snmpget). It
doesn't crash on the sighup, but on the subsequent snmp access after the
sighup.

The agent loads one custom module. I can see it being inited and
de-inited. Everything looks ok.
The agent does not crash when access the built in mib vars, only on one
of the mib vars in the loaded module.

This happens on 5.2.2 on a powerpc host, and also with 5.4.1 on an Intel
host.

It also seems to be dependent on the build. I have some code that does
a floating point divide calculation in one of the mib routines. If that
is present (not called) the agent behaves correctly and never crashes.
If I replace the floating point divide with an integer divide then the
agent will crash as described above.

The code is legitimate and is not called, but it seems to have an impact
on the behaviour of the agent when it is sent a HUP. The only think I
can think of is some weird loader memory boundary or something.

Has anyone ever seen anything like this or know where the problem could
be ???

My colleague says it is something to do with registering and
unregistering the mibs and function pointers pointing to old memory
allocations, or something like that.

Any help to solve this would be welcomed.

Thanks, Brendan.



-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757...un.com/javaone
_______________________________________________
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/...net-snmp-users