Zhouhuai Shen wrote:
> Hi Jan,
>
> Thanks a lot for your help.
>
> I have put the '[snmp] clientaddr' option to set source address of UDP
> packet with the trap as in the following in my snmpd.conf file:
>
> [snmp]
> clientaddr 192.168.1.1
> [snmpd]
> .....(The rest of commands in my snmpd.conf file)
>
> But when I capture the trap packet, it still shows the IP address of the
> outgoing interface. Did i do something wrong in my config file?

clientaddr option affects only UDP source address of the traps. It
should be 192.168.1.1, right? The agent address inside the UDP payload
will still be wrong. You need to apply the patch I mentioned before to
have possibility to set also this address.

Jan

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.p...r_id=100&url=/
_______________________________________________
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/...et-snmp-coders

Zhouhuai Shen wrote:
> Hi Jan,
>
> I checked the patch for 'v1trapaddress', it used the following to
> register the v1trapaddress string in the snmpd.conf:
>
> netsnmp_ds_register_config(ASN_OCTET_STR, app, "v1trapaddress",
> NETSNMP_DS_APPLICATION_ID, NETSNMP_DS_AGENT_TRAP_ADDR);
>
> Once the snmpd is re-read as it receives a SIGHUP signal, the string
> v1trapaddress will not be refreshed. For example, if I manually
> remove the 'v1trapaddress 10.0.0.1' from the snmpd.conf, then a
> SIGHUP is sent, the snmp daemon still thinks that the string defined
> for v1trapaddress is there

v1trapaddress does react to SIGHUP, it you change it, then the agent's
address in outgoing traps will be different. On the other side, you are
right with the removal - the old value is still used even if it's
removed from the config file. That's the usual behavior of snmpd, it
does not restore default value if a config option gets deleted. I am not
sure if it is bug or feature.

Jan

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.p...r_id=100&url=/
_______________________________________________
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/...et-snmp-coders

On Thu, 2008-09-18 at 14:47 +0200, Jan Safranek wrote:
>
> v1trapaddress does react to SIGHUP, it you change it, then the agent's
> address in outgoing traps will be different. On the other side, you are
> right with the removal - the old value is still used even if it's
> removed from the config file. That's the usual behavior of snmpd, it
> does not restore default value if a config option gets deleted. I am not
> sure if it is bug or feature.

I think that is a bug in the general case.

I see a couple of different value sources that I think should have
different semantics:

* Command line values - should be preserved for all
reconfigurations
* Config file values - should be rechecked at each SIGHUP, should
be used unless the command line overrides it
* Default values - should be used unless any of the other ones
override it

If a removal from the config file followed by a recheck fails to remove
the value from the actual config then I wonder how one is supposed to
tell the daemon that the value should be turned off?

/MF


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.p...r_id=100&url=/
_______________________________________________
Net-snmp-coders mailing list
Net-snmp-coders@lists.sourceforge.net
https://lists.sourceforge.net/lists/...et-snmp-coders