Hi good to see you are progressing.
Q1: you can shorten the timeout. That will help move stuff along. Or instead
of looking for agents, you can maybe incorporate a protocol like bonjour or
mdns? Google those and see. I forgot details on them
Q2: did you create an init_<module> and in there call the init function
created by mib2c ? You also might want to make that one a set request,
instead of get.
Good luck
David
On Nov 20, 2010 1:44 AM, "徐滕" <xut...@huaihuacity.com> wrote:
> Thanks David, and I still have to bother you with two questions.
>
> By now ,my practice is :
>
> I wrote my own mib which has two oids ,a vendorId (.1.3.6.1.4.1.8888.1, a
string type ,read-only ,for detecting device) ,a commandId
((.1.3.6.1.4.1.8888.2, a string type, read-only ,for execute command on the
agent) , add this mib to the agent
>
> question 1 : in detecting phase:
> I send the get-request for oid vendorID to the lan subnet , once anyone
return the vendorId I wanted , the manager recognise the responser . In my
test case, polling 60 IPs cost 6 minutes, it's a little slow, is anyother
more effective way to detecte devices I wanted?
>
> question 2: in execute phase
> I want the agent to execute a command such as shutdown when the manager
send a get-request for commandId, so ,I used the mib2c to generate .h and .c
from the mib I defined myself , in the get-request-handler for commandId, I
added a function system("shutdown -h now") , then compile these files to a
dynamic library mylib.so , loaded it in the snmpd.conf on the agent ,restart
snmpd service . but it doesn't work ,why ? How should I do ?
>
> Thanks
>
> Best Regards
>
> wade
>
> ------------------ 原始邮件 ------------------
> 发件人: "David A"<daaf...@gmail.com>;
> 发送时间: 2010年11月20日(星期六) 凌晨0:43
> 收件人: "徐滕"<xut...@huaihuacity.com>;
> 抄送: "net-snmp-users"<net-snmp-users@lists.sourceforge.net>;
> 主题: Re: Help using net-snmp
>
>
>
> Yes you would have to send requests to a range of IP addresses and see who
answers back. Or maybe you can use some bonjour like protocol.
>
> The manager can send requests to an agent that can trigger a command to be
executed yes.
> On Nov 18, 2010 6:48 PM, "徐滕" <xut...@huaihuacity.com> wrote:
>> Hi,all:
>>
>> I'm a fresh man to snmp, and now participating a project which has a
manger on the console server,and a agent on several
>>
>> terminals such as pc or tc via lan . I cannot get my head arount how the
manger detect the agent via lan ,should it brodcast a get-request to the lan
?
>>
>> and next question is ,could the manger send data to the agent or excute a
remote command on the agent via net-snmp ?
>>
>> Could anyone help me?
>>
>> Regards!
>>
>> wade
>>
>> 2010-11-19
------------------------------------------------------------------------------
Beautiful is writing same markup. Internet Explorer 9 supports
standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3.
Spend less time writing and rewriting code and more time creating great
experiences on the web. Be a part of the beta today
http://p.sf.net/sfu/msIE9-sfdev2dev
_______________________________________________
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/listinfo/net-snmp-users
