Re: Implement Get-Range

2007-01-24 Thread Juergen Schoenwaelder
On Wed, Jan 24, 2007 at 07:02:01PM +0200, Yaniv S. wrote: > Hello all, > > I'm Intrested to add an Implementation of get-range method to an snmp > client for academic reasons (resarch and such). > Get Range was defined in Internet Draft but wasn't approved as RFC. I personally would very much app

RE: USM DH Key change process

2007-01-24 Thread santhosh.sundarasamy
Thanks for the reply. But I need notification filtering. Thanks & Regards, Santhosh -Original Message- From: Wes Hardaker [mailto:[EMAIL PROTECTED] Sent: Thursday, January 25, 2007 10:36 AM To: SANTHOSH S (WT01 - Broadband Networks) Cc: [EMAIL PROTECTED]; net-snmp-coders@lists.sourceforg

Re: USM DH Key change process

2007-01-24 Thread Wes Hardaker
> "ss" == santhosh sundarasamy <[EMAIL PROTECTED]> writes: ss> The other process sends V3 notifications to manager. This ss> process uses Net-SNMP APIs to send the same. Since the trigger for ss> notifications are received as HTML (from one server) requests, sending ss> notification to Manager

Re: rfc: library layering

2007-01-24 Thread Wes Hardaker
> "TA" == Thomas Anders <[EMAIL PROTECTED]> writes: >> On Mac OS X, when we fold one library into another, we create a symlink >> of the old library to the new. For example, libpoll and libpthread now >> point to libSystem. Why not do the same here? Most linkers would ignore >> or at least see

Re: Re: Re: Re: Porting net-snmp to arm platform questions

2007-01-24 Thread liu.renfeng
I tried "--with-mibs=mibII ", and I get the summary as follow. obvious it does not contain all the mib II groups. As I konw there is 11 groups in mib II. Since CMOT is not used any more, so at least I should get 10 groups. Also I find out there were no "mibII/sysORTable" or "mibII/vacm_vars " in

Help please, need to access a row column value from inside the agent

2007-01-24 Thread Vinicius Pacheco
Hello all, I need help accessing an especific row column value using for that a given row. For example, please consider the following code: CODE --- int index; netsnmp_table_data_set *table_set; netsnmp_table_row *row; oid my_registration_oid[] = { 1, 3,

Re: Implement Get-Range

2007-01-24 Thread Dave Shield
On 24/01/07, Yaniv S. <[EMAIL PROTECTED]> wrote: > I'm Intrested to add an Implementation of get-range method to an snmp > client for academic reasons (resarch and such). > Get Range was defined in Internet Draft but wasn't approved as RFC. > My question is whether I'm able to implement it as a dif

Visual C++ Express and net-snmp

2007-01-24 Thread Daniel Lacey
I finally got the Visual C++ Express and PSDK on my XP machine... It does have NMAKE!!! That's the end of the good news... I tried to compile net-snmp as in section "Win32\build.bat script" and using the Visual C++ application. The BAT method gives me an error caused by: if (! (-d $ENV{MSVCD

Re: rfc: library layering

2007-01-24 Thread Thomas Anders
chris jalbert wrote: > On Mac OS X, when we fold one library into another, we create a symlink > of the old library to the new. For example, libpoll and libpthread now > point to libSystem. Why not do the same here? Most linkers would ignore > or at least see nothing useful in the symlink library.

Re: rfc: library layering

2007-01-24 Thread chris jalbert
On 23-Jan-2007, at 1:37 PM, Thomas Anders wrote: Wes Hardaker wrote: "TA" == Thomas Anders <[EMAIL PROTECTED]> writes: TA> Are you proposing to move libnetsnmphelpers content to TA> libnetsnmpagent and not ship libnetsnmphelpers with 5.5 at all? If TA> so, are we fine with requiring develo

Implement Get-Range

2007-01-24 Thread Yaniv S.
Hello all, I'm Intrested to add an Implementation of get-range method to an snmp client for academic reasons (resarch and such). Get Range was defined in Internet Draft but wasn't approved as RFC. My question is whether I'm able to implement it as a different module on your code? I need to retriev

Re: RowStatus type mib2c template problem

2007-01-24 Thread Emi . Yanagi
Hi Robert, Robert Story <[EMAIL PROTECTED] com>

Re: Questions

2007-01-24 Thread Dave Shield
On 22/01/07, Dave Shield <[EMAIL PROTECTED]> wrote: On 21/01/07, Magnus Fromreide <[EMAIL PROTECTED]> wrote: > The current net-snmp handling of [a short trap OID] seems > to be SIGSEGV, so I suppose something have to be done. Crashing out is not a good idea, no! At the very least

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Thomas Anders
Dave Shield wrote: > Good news: > EXE_PATH=`dirname $0` > *does* work, and gives the same result. > > Is it worth trying with that instead, and see how things go? I've already made this change on AIX, but I'm running into more issues [than time permits to work on for the moment]. +Thomas -

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Dave Shield
On 24/01/07, Dave Shield <[EMAIL PROTECTED]> wrote: > On 24/01/07, Thomas Anders <[EMAIL PROTECTED]> wrote: > > Random sample from nsb-nightly: > > > > EXE_PATH=${0%nsb-nightly} > > EXE_PATH=${EXE_PATH%/} > > > > won't work with generic Bourne shell, AFAIK. > > That ought to work with Posix-compati

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Dave Shield
On 24/01/07, Thomas Anders <[EMAIL PROTECTED]> wrote: > Random sample from nsb-nightly: > > EXE_PATH=${0%nsb-nightly} > EXE_PATH=${EXE_PATH%/} > > won't work with generic Bourne shell, AFAIK. That ought to work with Posix-compatible Bourne shells, AFAIK. It's certainly OK on HP-UX. But perhaps not

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Thomas Anders
Dave Shield wrote: > I haven't looked through the longest script (nsb-functions) > in the same detail, but I wouldn't expect anything untoward > there either. Random sample from nsb-nightly: EXE_PATH=${0%nsb-nightly} EXE_PATH=${EXE_PATH%/} won't work with generic Bourne shell, AFAIK. > If there

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Dave Shield
On 24/01/07, Thomas Anders <[EMAIL PROTECTED]> wrote: > While the nsb* scripts probably *could* be made to work with > generic /bin/sh, I don't think they do as-is. Have you any specific problems in mind? Four out of the five scripts are relatively trivial, and I've just been through them (again).

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Thomas Anders
Dave Shield wrote: > On 24/01/07, Thomas Anders <[EMAIL PROTECTED]> wrote: >> Have you actually *double-checked* whether all of these scripts work >> fine with basic Bourne shell on (early) *nix systems? > > No. > >>I don't think it'd >> be beneficial to prefer '#!/bin/sh' ove

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Dave Shield
On 24/01/07, Thomas Anders <[EMAIL PROTECTED]> wrote: > Have you actually *double-checked* whether all of these scripts work > fine with basic Bourne shell on (early) *nix systems? No. >I don't think it'd > be beneficial to prefer '#!/bin/sh' over '#!/usr/bin/env bash' if ther

Re: CVS: net-snmp/dist cvsshow, 1.2 cvsup, 1.12 nsb-config, 1.4 nsb-functions, 1.73 nsb-nightly, 1.21 nsb-package, 1.15 nsb-platform, 1.3

2007-01-24 Thread Thomas Anders
Dave Shield wrote: > Update of /cvsroot/net-snmp/net-snmp/dist > In directory sc8-pr-cvs7.sourceforge.net:/tmp/cvs-serv20299/dist > > Modified Files: > cvsshow cvsup nsb-config nsb-functions nsb-nightly nsb-package > nsb-platform > Log Message: > Improve shell-script portability. (Ev

Re: configure warning

2007-01-24 Thread Thomas Anders
Karyn Williams wrote: > Got this message running configure on freebsd 6.1-RELEASE today. > > checking sys/conf.h presence... yes > configure: WARNING: sys/conf.h: present but cannot be compiled This (known) warning can be safely ignored. I'll see if it can be suppressed for 5.4.1. Can you send

Re: Re: Re: Porting net-snmp to arm platform questions

2007-01-24 Thread Dave Shield
On 24/01/07, liu.renfeng <[EMAIL PROTECTED]> wrote: > Thanks very much for your kindly reply. > You really understand my problem well. That's what you're paying me for :-) > Agent MIB code: default_modules => snmpv3mibs mibII/snmp_mib > mibII/system_mib mibII/sysORTable mibII

configure warning

2007-01-24 Thread Karyn Williams
Got this message running configure on freebsd 6.1-RELEASE today. checking sys/conf.h presence... yes configure: WARNING: sys/conf.h: present but cannot be compiled configure: WARNING: sys/conf.h: check for missing prerequisite headers? configure: WARNING: sys/conf.h: see the Autoconf documen

Re: Re: Re: Porting net-snmp to arm platform questions

2007-01-24 Thread liu.renfeng
Thanks very much for your kindly reply. You really understand my problem well. I searched the Internet and used following options: - [EMAIL PROTECTED] net-snmp-5.4]# ./configure --enable-mini-agent --with-mibs="IP-MIB:IF-MIB:TCP-MIB:UDP-M

Re: Re: Porting net-snmp to arm platform questions

2007-01-24 Thread Dave Shield
On 24/01/07, liu.renfeng <[EMAIL PROTECTED]> wrote: > Is there any thing I need to do to fix this problem ? You need to simplify your snmpd.conf file! In particular, you have *two* active access control sections: > com2sec notConfigUser default public > group notConfigGroup v1