[prev in list] [next in list] [prev in thread] [next in thread] 

List:       net-snmp-support
Subject:    [ net-snmp-Support Requests-428760 ] extending net-snmp: timeouts!
From:       noreply () sourceforge ! net
Date:       2001-06-01 11:46:17
[Download RAW message or body]

Support Requests item #428760, was updated on 2001-05-30 11:35
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=212694&aid=428760&group_id=12694

Category: None
Group: None
Status: Open
>Priority: 1
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: extending net-snmp: timeouts!

Initial Comment:
Hi there!
	I'll first explain what I'm doing and after that, I'll
be very
simple to explain what problem I'm having. I'm writing
an attached agent to the main agent, using the AgentX
protocolol, following the UCD-SNMP tutorial. But the
variables I'm interested are not resident in the
computer running the agent programs, they are stored in
another equipment. The whole idea is to provide SNMP
management to an equipment without native SNMP support.
The communication between this equipment and the
computer running the agents is made through a serial
line using a proprietary protocol. So the agent gets a
request (GET, SET) and then "translates" to this other
protocol, gets the value and returns (or writes) the
correct value.
	Everything is working fine, but this: the whole CASE
routine for the variable is executed several times.
Lets say there's a printf command right before the
return in the variable routine( just like in the
ustScalarSet.c file) , well, I see the same message
printed several times. What I think is happening is
that due to the delay of the serial communication, an
timeout occurs somewhere and then the agent just call
for the variable's routine again and again. It's very
sensible in a SET command. I managed to reduce this
delay, so now SETs are working fine. I made an
experience to create a integer variable just like in
the UCD-SNMP tutorial and (as before) put a printf in
the write_method of this variable and it executes it
four times, for a variable that is in the machine's
memory.
	To correct the problem I tried do edit the
snmp_select_info calls in the agent_check_and_process()
function (snmp_agent.c), but that hasn't solved it. I
then started reading several files (snmpd.c,
master_admin.c, agentX .c files) but I couldn't find
where to block the requests or where to disable some
timeouts.
	I'd be very (really) grateful if anyone could give me
ANY help or indicate a FAQ, man page, literature so I
could solve this problem. Thank you very much for your
time and sorry for the long massage.
	
Best Regards.

Leonardo Santos
------------------------------------------------

"All things are ready if our minds be so"
The Life of King Henry V - W. Shakespeare

Leonardo Pereira Santos
Engenharia de Desenvolvimento
Divisão de Fibras Óticas
Parks SA Comunicações Digitais


----------------------------------------------------------------------

>Comment By: John Naylon (jbpn)
Date: 2001-06-01 04:46

Message:
Logged In: YES 
user_id=93926

This is normal behaviour for a SET request.

See
http://net-snmp.sourceforge.net/tutorial/toolkit/mib_module/index.html

----------------------------------------------------------------------

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=212694&aid=428760&group_id=12694

_______________________________________________
Net-snmp-support mailing list
Net-snmp-support@lists.sourceforge.net
http://lists.sourceforge.net/lists/listinfo/net-snmp-support

[prev in list] [next in list] [prev in thread] [next in thread]