Hi MF,
Thanks for your help.
That's shame I don't really understand how to use your example.
I guess your main point is taht I can get index values in handle_table().
static int
handle_table(netsnmp_mib_handler* handler,
netsnmp_handler_registration* reginfo,
netsnmp_ag
Aha, I hadn't changed --default-mibs. That's it. Thanks kindly!
-Dan
2009/6/9 Wes Hardaker :
>> On Tue, 9 Jun 2009 15:57:58 -0700, dan anderson
>> said:
>
> da> A few places in the docs it is mentioned that MIBS=ALL is needed to
> da> denote that all mib files (normally in /usr/local/s
> On Tue, 9 Jun 2009 15:57:58 -0700, dan anderson
> said:
da> A few places in the docs it is mentioned that MIBS=ALL is needed to
da> denote that all mib files (normally in /usr/local/share/snmp/mibs/)
da> should be loaded. Alternately, individual mibs can be specified
da> instead of the
A few places in the docs it is mentioned that MIBS=ALL is needed to
denote that all mib files (normally in /usr/local/share/snmp/mibs/)
should be loaded. Alternately, individual mibs can be specified
instead of the ALL keyword. In my testing, library calls to
snmp_parse_oid parse the mibs in that d
> On Wed, 03 Jun 2009 17:05:37 +0200 (CEST), Havard Eidnes
> said:
HE> This allows the user to implement trap-directed polling, which I think
HE> was part of the original management model of SNMP. However, I could
HE> not find anything which would immediately allow me to implement the
H
> On Fri, 5 Jun 2009 11:35:14 +0530, said:
pk> ld.so.1: snmpwalk: fatal: libcrypto.so.0.9.8: open failed: No such
pk> file or directory
The application was compiled on a system that had openssl installed.
You don't have openssl on the machine you're trying to run it on, so
there is an issue.
> On Tue, 9 Jun 2009 17:35:54 -0400, Geoffrey Bays
> said:
GB> I do not seem to be able to test the return of the IPADDRESS type,
GB> and am wondering which netsnmp_vardata field (this is a union) it
GB> returns into.
Many of them will technically work. The easiest one is probably the
Net-Snmpers:
I do not seem to be able to test the return of the IPADDRESS type,
and am wondering which netsnmp_vardata field (this is a union) it returns into.
My surmise would be the objid field as it is an unsigned long rather
than the integer field.
That way a four-part IP v4 address would be
On Mon, 2009-06-08 at 19:10 +0800, Tewen Hsieh wrote:
>
> Hi,
>
> Thanks for your help.
>
> But I can't understand your suggestion. Could you explain more
> clearly? Thanks.
>
> I want to get the the index *values*, not to know the indices a table
> used.
> For example, in XXX table there
Hi Dave,
Thanks for your help.
I have tried to generate the MIB code on another system(5.4.1), and copy
them to the 5.1.1 system.
I also referenced the code at the 'agent/mibgroup/hardware/cpu'(especially
cpu.c), but I can't understand them well.
Even I don't think they were generated from "mib
2009/6/9 Tewen Hsieh :
> does 5.1.1 also support "mib2c -Scache=1 -c mib2c.iterate.conf "?
No - I don't believe it does.
Though you should be able to generate the MIB code
on another (later) system, and copy this to the 5.1.1 system.
The cache helper is certainly present in the 5.1.x line,
i
Hi Dave,
Thanks for your help. I got it.
One more question, does 5.1.1 also support "mib2c -Scache=1 -c
mib2c.iterate.conf "?
Even I plan to update my agent in the near future.
Thanks,
永鈦鑫 iTAS Corporation
新竹市水源街75號
NO. 75,Shuiyuan St., Hsinchu City, Taiwan, R.O.C.
Tel:03-5716099 Ext.
2009/6/9 Tewen Hsieh :
> My key question is how to *not* get *all* external data one time.
You don't need to get all the external data - just the index information.
That's all that the iterate hook routines actually require.
If you can retrieve information about what rows are valid,
then that's s
13 matches
Mail list logo
