net-snmp-coders — Discussion of internal project coding (Please use this OR -users, NOT both!)

Re: bug 123815: Perl getnext failing to split instances in 4.2pre2

[G. S. M. <ma...@ti...>]

I took it...ahh...now I have to figure how to turn on prEntry in the
agent ;) anyone want to give me a hint.

-GSM

(no subject)

[wk l. <le...@ho...>]

hi,
i'm curently working on a project which requires semding SNMP commands in 
Tcl. i was wondering if you could help me on this. i've loaded in the Tnm 
package and that comes with Scotty. when i issuing the "snmp session" 
command, it returns me with the following.

% snmp session
bad option "session": must be alias, delta, expand, find, generator, info, 
listener, notifier, oid, responder, type, value, wait, or watch

i understand that there're snmp agents around such as CMU SNMP, WIlMA and 
MIB compilers such as mosy and snacc. do i need to load them in as well ??


regards
weikit

_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com

Perl module tests 100% OK

[G. S. M. <ma...@ti...>]

Well at least on my linux box. I think it is good to go...I may get a
chance to test it on Solaris and win32 but that may wait until Monday.

regards, GSM

Re: the strange cores related to wide characters are back?

[G. S. M. <ma...@ti...>]

no LAN or LC_CTYPE set. It is comforting to hear I am not the only one who gets
repeatable problems with efence...I thought it was something to do with my
environment.  I am pretty wide-char illiterate - I just saw that ISO stuff and
thought it must have something to do with wide characters.

any way all is well (except that it would be nice to run efence some times).

I see I have version 2.0.5
 Electric Fence 2.0.5 Copyright (C) 1987-1998 Bruce Perens.

Maybe there is a fix.

regards, GSM

ni...@ba... wrote:

> On Sat, 02 Dec 2000 12:05:48 -0500 "G. S. Marzot" <ma...@ti...> wrote:
> >Just did another make distclean and rebuild on my RH 6.1 linux box and
> >the strange cores that seem related to wide character support have
> >resurfaced. Maybe my linux box is just screwed up? Is anyone else using
> >RH6.1?
> >
> >I don't know why it thinks I am using wide chars at all...this line
> >seems strange to me as well.
> >
> >Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.
>
> Well, 8859-1 is my usual character set, and it is no wider than 8 bit, but I
> have long since had to always configure --without-efence, as it would never
> work. SO there is probably something bad in that shared library. Did you set any
> LANG or LC_CTYPE environment variables?
>
> /Niels
>
> --
> Get your firstname@lastname email for FREE at http://Nameplanet.com/?su

Re: the strange cores related to wide characters are back?

[<ni...@ba...>]

On Sat, 02 Dec 2000 12:05:48 -0500 "G. S. Marzot" <ma...@ti...> wrote:
>Just did another make distclean and rebuild on my RH 6.1 linux box and
>the strange cores that seem related to wide character support have
>resurfaced. Maybe my linux box is just screwed up? Is anyone else using
>RH6.1?
>
>I don't know why it thinks I am using wide chars at all...this line
>seems strange to me as well.
>
>Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.

Well, 8859-1 is my usual character set, and it is no wider than 8 bit, but I
have long since had to always configure --without-efence, as it would never
work. SO there is probably something bad in that shared library. Did you set any
LANG or LC_CTYPE environment variables?

/Niels


-- 
Get your firstname@lastname email for FREE at http://Nameplanet.com/?su

Datagram Berkeley Sockets

[Daniel L. N. <dne...@jp...>]

Hello,
  This is slightly off topic but I couldn't find any references =
elsewhere.  I'm trying to write a SNMP broker, a process that handles =
all the UCD requests. I got the engine working and now am looking into =
writing a Berkeley Sockets wrapper to handle the requests.  I found the =
examples in the Perl Cookbook and perlipc.html didn't work for datagrams =
AND the functions for Socket and IO:Socket:UNIX are not documented. I've =
contacted several groups to no avail.

Does anyone have an example in either C or PERL to handle this OR better =
knows where the Socket.pm/UNIX.pm modules are documented?=20

Thanks,
Daniel L. Needles

Re: simple agent

[Gordon S. <tig...@ho...>]

<html><DIV>
<P><BR><BR></P></DIV>
<DIV></DIV>
<DIV></DIV>&gt;From: "Michael J. Slifcak" <MSL...@IS...>
<DIV></DIV>&gt;To: Gordon Shiao <TIG...@HO...>
<DIV></DIV>&gt;CC: net...@li... 
<DIV></DIV>&gt;Subject: Re: simple agent 
<DIV></DIV>&gt;Date: Sat, 02 Dec 2000 11:07:24 -0500 
<DIV></DIV>&gt; 
<DIV></DIV>&gt;Gordon Shiao wrote: 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; &gt; hi ALL, 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; &gt; I'm porting Net-SNMP to an embedded platform. Currently I only need SNMP 
<DIV></DIV>&gt; &gt; v1/v2c with MIBII to get a smaller size agent. 
<DIV></DIV>&gt; &gt; Is there a way for me to extract the files to fit such functions? I can not 
<DIV></DIV>&gt; &gt; run autoconf and ./configure on my platform and thus I will have to edit the 
<DIV></DIV>&gt; &gt; makefiles manually. 
<DIV></DIV>&gt; &gt; Net-SNMP is well designed for compiling on different UNIX platforms, however 
<DIV></DIV>&gt; &gt; when I look into the code and want to port them to our RTOS it's really 
<DIV></DIV>&gt; &gt; tough just because it's hard to locate some pure codes such as v1/v2c agent 
<DIV></DIV>&gt; &gt; with MIBII methods. 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; &gt; Is there a way to get the code(*.c,*.h) for a v1/v2c agent and its MIB-II 
<DIV></DIV>&gt; &gt; stubs? No automatically compilation, no v3 security, no agentx and so forth. 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; &gt; Thanks for any suggestion and helps. 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; &gt; Gordon 
<DIV></DIV>&gt; &gt; 
<DIV></DIV>&gt; 
<DIV></DIV>&gt;Hi, Gordon. 
<DIV></DIV>
<P>&gt;Which embedded platform are you porting to ? </P>
<P>It's Microware OS9000 for PowerPC. It's Unix-like but its file system hierarchy is different from the usual UNIX. It's POSIX compliant and has ANSI C,BSD socket supportability.</P>
<DIV></DIV>
<P>&gt;Which MIBII modules do you require ? </P>
<P>RFC1213 MIBII. and some other self-defined MIBs.</P>&gt; 
<DIV></DIV>&gt;If you have a UNIX like system, you could mold the Makefiles 
<DIV></DIV>&gt;there, then edit them on your target development system to suit. 
<DIV></DIV>&gt;If not, let me know the modules list, and I'll give you Makefiles 
<DIV></DIV>&gt;generated from my Linux system, to give you a starting point. 
<DIV></DIV>
<P>&gt; </P>
<P>Thanks, I built Net-SNMP on Redhat 7.0. And I got two different snmpd size, one is 900K or so and the other is 30K or so. I think the bigger one is using static library,but it's extreme big&nbsp;comparing to what i need. </P>
<DIV></DIV>&gt;If you don't want to use the net-snmp sources, 
<DIV></DIV>&gt;there are plenty of SNMP agent sources available 
<DIV></DIV>
<P>&gt;through the World Wide Web. </P>
<P>Where are they for embedded portability?</P>
<P>I have a strong interest to use Net-SNMP, but I only need the source files I want, such as v1/v2c source code and mibii method calls. My point is how to get them out of the Net-SNMP package before i can work on them manually(changed them somehow and rewrite the makefiles). Is it a good idea to use CMU v1.14. it's two years old and it only supports v1/v2c,but it's simpler with a straightforward unpacked directory.</P>
<DIV></DIV>&gt; 
<DIV></DIV>&gt; 
<DIV></DIV>
<P>&gt;Good Luck, </P>&gt;-Mike Slifcak, Internet Security Systems, Inc. 
<DIV></DIV>&gt;_________________
<P>Thanks you so much for your kind help, I really appreciate it.</P>
<P>Gordon</P>
<P>&nbsp;</P>
<DIV></DIV>&gt;Net-snmp-coders mailing list 
<DIV></DIV>&gt;Net...@li... 
<DIV></DIV>&gt;http://lists.sourceforge.net/mailman/listinfo/net-snmp-coders 
<DIV></DIV><br clear=all><hr>Get more from the Web.  FREE MSN Explorer download : <a href="http://explorer.msn.com">http://explorer.msn.com</a><br></p></html>

Re: the strange cores related to wide characters are back?

[G. S. M. <ma...@ti...>]

update: make distclean, configure --without-efence, rebuild and it goes
away

-GSM


"G. S. Marzot" wrote:

> Just did another make distclean and rebuild on my RH 6.1 linux box and
> the strange cores that seem related to wide character support have
> resurfaced. Maybe my linux box is just screwed up? Is anyone else using
> RH6.1?
>
> I don't know why it thinks I am using wide chars at all...this line
> seems strange to me as well.
>
> Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.
>
> full output below.
>
> -GSM
>
> [gmarzot@donzi testing]$ gdb -c /tmp/snmp-test-26766/core
> ../apps/snmpget
> GNU gdb 4.18
> Copyright 1998 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you
> are
> welcome to change it and/or distribute copies of it under certain
> conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for
> details.
> This GDB was configured as "i386-redhat-linux"...
> Core was generated by `snmpget -d -p 8765 -v 1 localhost testcommunity
> system.sysUpTime.0'.
> Program terminated with signal 11, Segmentation fault.
> Reading symbols from /lib/libdl.so.2...done.
> Reading symbols from /usr/lib/librpm.so.0...done.
> Reading symbols from /lib/libdb.so.2...done.
> Reading symbols from /usr/lib/libz.so.1...done.
> Reading symbols from /lib/libm.so.6...done.
> Reading symbols from /lib/libc.so.6...done.
> Reading symbols from /lib/ld-linux.so.2...done.
> Reading symbols from /usr/lib/libbz2.so.0...done.
> Reading symbols from /lib/libnss_files.so.2...done.
> Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.
> #0  gconv (step=0x40ec1fc8, data=0xbfffeb18, inbuf=0xbfffeb0c,
>     inbufend=0x40176d24 "", written=0xbfffeb10, do_flush=0)
>     at ../iconv/skeleton.c:204
> 204 ../iconv/skeleton.c: No such file or directory.
> (gdb) where
> #0  gconv (step=0x40ec1fc8, data=0xbfffeb18, inbuf=0xbfffeb0c,
>     inbufend=0x40176d24 "", written=0xbfffeb10, do_flush=0)
>     at ../iconv/skeleton.c:204
> #1  0x400f893b in __mbrtowc (pwc=0xbfffec00, s=0x40176d23 ".", n=1,
>     ps=0xbfffec04) at mbrtowc.c:67
> #2  0x400e30f6 in _IO_vfscanf (s=0x40403f50, format=0x809676a "%s
> %[^\n]\n",
>     argptr=0xbfffec60, errp=0x0) at vfscanf.c:254
> #3  0x400e66c8 in __vfscanf (s=0x40403f50, format=0x809676a "%s
> %[^\n]\n",
>     argptr=0xbfffec60) at vfscanf.c:1294
> #4  0x400e66ef in fscanf (stream=0x40403f50, format=0x809676a "%s
> %[^\n]\n")
>     at fscanf.c:31
> #5  0x8055a8b in add_mibdir (
>     dirname=0x403fffd4 "/usr/local/builds/net-snmp/testing/../mibs")
>     at parse.c:4073
> #6  0x804cada in init_mib () at mib.c:1168
> #7  0x8057002 in init_snmp (type=0x80999e0 "snmpapp") at snmp_api.c:642
> #8  0x80681da in snmp_parse_args (argc=9, argv=0xbffffa64,
> session=0xbffff94c,
>     localOpts=0x8092cb0 "C:", proc=0x8049ed0 <optProc>)
>     at snmp_parse_args.c:411
> #9  0x8049fa0 in main (argc=9, argv=0xbffffa64) at snmpget.c:130

the strange cores related to wide characters are back?

[G. S. M. <ma...@ti...>]

Just did another make distclean and rebuild on my RH 6.1 linux box and
the strange cores that seem related to wide character support have
resurfaced. Maybe my linux box is just screwed up? Is anyone else using
RH6.1?

I don't know why it thinks I am using wide chars at all...this line
seems strange to me as well.

Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.

full output below.

-GSM

[gmarzot@donzi testing]$ gdb -c /tmp/snmp-test-26766/core
../apps/snmpget
GNU gdb 4.18
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for
details.
This GDB was configured as "i386-redhat-linux"...
Core was generated by `snmpget -d -p 8765 -v 1 localhost testcommunity
system.sysUpTime.0'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /lib/libdl.so.2...done.
Reading symbols from /usr/lib/librpm.so.0...done.
Reading symbols from /lib/libdb.so.2...done.
Reading symbols from /usr/lib/libz.so.1...done.
Reading symbols from /lib/libm.so.6...done.
Reading symbols from /lib/libc.so.6...done.
Reading symbols from /lib/ld-linux.so.2...done.
Reading symbols from /usr/lib/libbz2.so.0...done.
Reading symbols from /lib/libnss_files.so.2...done.
Reading symbols from /usr/lib/gconv/ISO8859-1.so...done.
#0  gconv (step=0x40ec1fc8, data=0xbfffeb18, inbuf=0xbfffeb0c,
    inbufend=0x40176d24 "", written=0xbfffeb10, do_flush=0)
    at ../iconv/skeleton.c:204
204 ../iconv/skeleton.c: No such file or directory.
(gdb) where
#0  gconv (step=0x40ec1fc8, data=0xbfffeb18, inbuf=0xbfffeb0c,
    inbufend=0x40176d24 "", written=0xbfffeb10, do_flush=0)
    at ../iconv/skeleton.c:204
#1  0x400f893b in __mbrtowc (pwc=0xbfffec00, s=0x40176d23 ".", n=1,
    ps=0xbfffec04) at mbrtowc.c:67
#2  0x400e30f6 in _IO_vfscanf (s=0x40403f50, format=0x809676a "%s
%[^\n]\n",
    argptr=0xbfffec60, errp=0x0) at vfscanf.c:254
#3  0x400e66c8 in __vfscanf (s=0x40403f50, format=0x809676a "%s
%[^\n]\n",
    argptr=0xbfffec60) at vfscanf.c:1294
#4  0x400e66ef in fscanf (stream=0x40403f50, format=0x809676a "%s
%[^\n]\n")
    at fscanf.c:31
#5  0x8055a8b in add_mibdir (
    dirname=0x403fffd4 "/usr/local/builds/net-snmp/testing/../mibs")
    at parse.c:4073
#6  0x804cada in init_mib () at mib.c:1168
#7  0x8057002 in init_snmp (type=0x80999e0 "snmpapp") at snmp_api.c:642
#8  0x80681da in snmp_parse_args (argc=9, argv=0xbffffa64,
session=0xbffff94c,
    localOpts=0x8092cb0 "C:", proc=0x8049ed0 <optProc>)
    at snmp_parse_args.c:411
#9  0x8049fa0 in main (argc=9, argv=0xbffffa64) at snmpget.c:130

Re: simple agent

[Michael J. S. <msl...@is...>]

Gordon Shiao wrote:
> 
> hi ALL,
> 
> I'm porting Net-SNMP to an embedded platform. Currently I only need SNMP
> v1/v2c with MIBII to get a smaller size agent.
> Is there a way for me to extract the files to fit such functions? I can not
> run autoconf and ./configure on my platform and thus I will have to edit the
> makefiles manually.
> Net-SNMP is well designed for compiling on different UNIX platforms, however
> when I look into the code and want to port them to our RTOS it's really
> tough just because it's hard to locate some pure codes such as v1/v2c agent
> with MIBII methods.
> 
> Is there a way to get the code(*.c,*.h) for a v1/v2c agent and its MIB-II
> stubs? No automatically compilation, no v3 security, no agentx and so forth.
> 
> Thanks for any suggestion and helps.
> 
> Gordon
> 

Hi, Gordon.
Which embedded platform are you porting to ?
Which MIBII modules do you require ?

If you have a UNIX like system, you could mold the Makefiles
there, then edit them on your target development system to suit.
If not, let me know the modules list, and I'll give you Makefiles
generated from my Linux system, to give you a starting point.

If you don't want to use the net-snmp sources,
there are plenty of SNMP agent sources available
through the World Wide Web.


Good Luck,
-Mike Slifcak, Internet Security Systems, Inc.

Re: smux-peer and net-snmp

[Jochen F. <jo...@sc...>]

Hi Dave,

> > Attached patch ( against 4.2.pre2 ) suppresses sending previous
> > getRequest before setRequest to smux-peer. This works.
> 
> I have to say that your proposed change bring an uncomfortably
> large amount of protocol-specific code into the guts of the agent.
> 
> Can I suggest you might like to consider an alternative approach
> that I used for the AgentX support.   In this model, the 'get'
> routine is essentially a dummy handler, that simply adds the
> requested object to an outgoing request, and trivially returns success.
> Nothing is actually sent to the subagent until all the variables
> have been processed.

I guess this would also fix the second case of duplicate requests:

If a subagent has more than one registered tree and a GETNEXT request to
the subagent rolls over to the next registered subtree, the master agent
will ignore the response and repeat it later.

A cached response would avoid the second subagent call.

Cheers,
Jochen

RE: help required : *SNMP managers* and *indexing*

[Shshank_Sharma <Shs...@sa...>]

Hi Wes,
I guess I am not pestering you too much already.
I'll just take this opportunity to explain what I am doing, ofcourse not
wanting to eat into your useful time :-)

What I have is this :
I am writing an SNMP sub-agent (implementing three MIBs) for an H.323
Gatekeeper( the NE).
I am using a freeware sub-agent development toolkit for the same. It
provides a master-agent (MA) - sub-agent (SA) paradigm and provides APIs to
register the sub-agent and its tables with the MA.
I can specify my own access method prototypes for the MIB variables when I
compile the smi file (that maps the MIB variables to corrsponding
C-variables)
Inside those prototype implementations, I have to retrieve *real-life*
parameters from the NE ( I am using UDP messages for doing that).
Guess that is the little background that will help clarify the aspect we
have been discussing.

Would you really say that I retrieve actual values from the NE before I
register the row with the master agent ?

Thanks for your time.
--Shshank

> -----Original Message-----
> From:	Wes Hardaker [SMTP:wjh...@uc...]
> Sent:	Friday, December 01, 2000 11:38 PM
> To:	Shshank_Sharma
> Cc:	Wes Hardaker; ucdavissnmpcoders; Varaprasad_Ajjampudi;
> PradeepKumar_Keloth
> Subject:	Re: help required : *SNMP managers* and *indexing*
> 
> >>>>> On Fri, 1 Dec 2000 21:30:16 +0530 , Shshank_Sharma
> <Shs...@sa...> said:
> 
> Shshank> If I am initializing the indexing variables for a table
> Shshank> instance with some dummy values when I am registering the row
> Shshank> with the master agent, does that make a difference ?
> 
> Yes.  You must use the real index when registering an index level
> registration with a master agent.
> 
> Unfortunately, its hard for me to picture exactly what you're doing so
> my answer is partially based on a guess of what you're doing.
> 
> -- 
> Wes Hardaker
> Please mail all replies to net...@li...

A minor patch.

[Harrie H. <ha...@co...>]

HI all,

I have attached a small patch that avoids two warnings
of type conflicts. These warnings are on

	AIX 4.3 using the native  compiler.

Cheers,
Harrie 
0- Harrie Hazewinkel ---------------------------------------0
 mailto:ha...@co...            phone:+1-415-536-5221
0-----------------------------------------------------------0

Re: snmpd dies on snmpset -- urgent !

[<ni...@ba...>]

On Wed, 15 Nov 2000 10:33:45 +0530 Kapil Verma <Ka...@ne...> wrote:
>    I am using zebra 0.89/ and ucd 4.1.2/ and using smux for BGP MIB
>support. I have bgpd and snmpd running on a machine and bgp peer running on
>another machine. when I do an snmpwalk, i get 19 variables only. But when i
>try to do an snmpset on any table-entry variable in the BGP MIB(which has a
>read-write permission), the snmpd is killed, smux connection is closed, core
>file is generated and there is no reponse.  

I have tried duplicating your problem with our 4.2.pre2 beta version, but it
just seems to work. Could you please try that version if see if it works for
you?

/Niels


-- 
Get your firstname@lastname email for FREE at http://Nameplanet.com/?su

simple agent

[Gordon S. <tig...@ho...>]

hi ALL,

I'm porting Net-SNMP to an embedded platform. Currently I only need SNMP 
v1/v2c with MIBII to get a smaller size agent.
Is there a way for me to extract the files to fit such functions? I can not 
run autoconf and ./configure on my platform and thus I will have to edit the 
makefiles manually.
Net-SNMP is well designed for compiling on different UNIX platforms, however 
when I look into the code and want to port them to our RTOS it's really 
tough just because it's hard to locate some pure codes such as v1/v2c agent 
with MIBII methods.

Is there a way to get the code(*.c,*.h) for a v1/v2c agent and its MIB-II 
stubs? No automatically compilation, no v3 security, no agentx and so forth.

Thanks for any suggestion and helps.

Gordon
_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com

Re: help required : *SNMP managers* and *indexing*

[Wes H. <wjh...@uc...>]

>>>>> On Fri, 1 Dec 2000 21:30:16 +0530 , Shshank_Sharma <Shs...@sa...> said:

Shshank> If I am initializing the indexing variables for a table
Shshank> instance with some dummy values when I am registering the row
Shshank> with the master agent, does that make a difference ?

Yes.  You must use the real index when registering an index level
registration with a master agent.

Unfortunately, its hard for me to picture exactly what you're doing so
my answer is partially based on a guess of what you're doing.

-- 
Wes Hardaker
Please mail all replies to net...@li...

Re: snmpconf needs testing

[Wes H. <wjh...@uc...>]

>>>>> On Fri, 01 Dec 2000 10:03:03 -0500, "Michael J. Slifcak" <msl...@is...> said:

Michael> One very small problem arises when responding to prompts for
Michael> SNMPv3 user.  I can enter "mike noauth" for user name, then
Michael> "auth" for min. security level, which will generate the
Michael> illegal line :

Michael> rwuser mike noauth noauth

fixed.

Michael> Thanks for building a wonderful admin tool!

Well, its a hack, but it'll hopefully help...

-- 
Wes Hardaker
Please mail all replies to net...@li...

Still Perl problems with latest 4.2

[<gm...@no...>]

I have fixed a number of the issues with the perl/SNMP module but I am
still getting oddities (cores during make test on linux) and test
failures.

I have added more support for targeted perl module builds so people can
use a specific net-snmp install (yahoo - no more accidentally building
against the wrong libsnmp and subsequent email to net-snmp-coders)

perl Makefile.PL -NET-SNMP-PATH=/usr/local/net-snmp-stage

Another issue that came up was the impact of the
"create_user_from_session" behaviour for environements like perl/SNMP
that create multiple sessions. There was an undesired side effect that
if a session was created for user A with some given parameters then
another session created for user A with more detailed parameters (more
security etc.) would not augment those parameters. I made a change which
allows the existing user to be augmented when the subsequent session is
created but it is still problematic if the second session contradicts
user parameters in the first session.

Typically in perl you just want to "do what I mean" for the users but I
am not sure how to provide that in this case. One way would be to go
back to the "if user exists, then don't change it" model and perhaps
provide a user creation/modification API in the module. thoughts?

-GSM

BTW, I will not be at the IETF so hoist a few for me and regards to all

-- 
G.S. Marzot                        email: gm...@no...
Nortel Networks                    voice: (978)288-3990
600 Tech Park  M/S E65-60-405      
Billerica, MA  01821                 fax: (978)670-8145

Re: snmpconf needs testing

[Wes H. <wjh...@uc...>]

>>>>> On Fri, 1 Dec 2000 11:39:35 +0100 , RADMAN Stefan <Ste...@ct...> said:

RADMAN> VERY useful stuff. This really could take the crypt out of snmp*.conf :)

That was my goal.

RADMAN> local/snmpconf.dir/snmd.conf/monitor:
RADMAN> change
RADMAN> "multiple file Do you want to configure the agents ability to file sizes?"
RADMAN> to
RADMAN> "multiple file Do you want to configure the agents ability to monitor file
RADMAN> sizes?"
RADMAN> and add info about the units of "maxsize".

Good points, thanks.  I've made these changes.

-- 
Wes Hardaker
Please mail all replies to net...@li...

Re: URGENT: [Fwd: Maximum size of a trap packet]

[Wes H. <wjh...@uc...>]

>>>>> On Fri, 1 Dec 2000 10:44:04 -0500, "Ning Zhu" <nz...@op...> said:

Ning> acl.c          context.c      mib.c          party.c        snmp_api.c
Ning> snmp_client.h  sw_snmp.h      acl.h          context.h      mib.h
Ning> party.h        snmp_api.h     snmp_impl.h    vssver.scc     asn1.c
Ning> md5.c          parse.c        snmp.c         snmp_auth.c    snmp_nt.c
Ning> asn1.h         md5.h          parse.h        snmp.h         snmp_client.c
Ning> snmp_protos.h

Ning> If you are familiar with this version of snmplib, could you
Ning> point it out how I can increase the maximum size of a trap
Ning> packet?  So far we are limited to 80 characters per packet.

Thats not our library, so I doubt we can help you...

-- 
Wes Hardaker
Please mail all replies to net...@li...

RE: help required : *SNMP managers* and *indexing*

[Shshank_Sharma <Shs...@sa...>]

Wes, thanks a lot for your reply.

I guess I did not make myself too clear !...especially in Question #2. 
This is about Question#2 below and the answer thereof.
What I intend to know is this :
If I am initializing the indexing variables for a table instance with some
dummy values when I am registering the row with the master agent, does that
make a difference ?
(By dummy values I mean values which are not the *actual* values as present
on the NE for the corresponding *real-life* parameters.)
Do the values to be given to these variables have to be the same as the
actual values at the NE ?
I am writing functions for getting/setting individual variables at the NE. I
have a store of the same (the MIB !) at the agent also.
The master agent and sub-agent talk SMUX ( which is hidden from me). Thus
when a user-supplied function is called for a particular variable, do the
indexing variables' values get filled by the master agent or do I have to
fill them in, in the function's implementation ?
Do you think I need to contact the NE at the time of the row registration
with Master Agent and get the valid indexing variable values from it, and
then register the row ?

( That was quite a mouthful :-))
Guess I have conveyed now what I wish to ask.
Thanks for your time.
br,
Shshank



> -----Original Message-----
> From:	Wes Hardaker [SMTP:wjh...@uc...]
> Sent:	Friday, December 01, 2000 1:45 AM
> To:	Shshank_Sharma
> Cc:	ucdavissnmpcoders; Varaprasad_Ajjampudi; PradeepKumar_Keloth
> Subject:	Re: help required : *SNMP managers* and *indexing*
> 
> >>>>> On Mon, 27 Nov 2000 18:53:39 +0530, Shshank_Sharma
> <Shs...@ht...> said:
> 
> Shshank> How does an SNMP manager form valid object identifiers(OIDs)
> Shshank> for tabular variables that have multiple variables indexing
> Shshank> them ?  Does it just do an snmpwalk from a higher level and
> Shshank> see what are the valid indices by inpecting the returned OIDs
> Shshank> ? or does it have some way of forming them itself ?
> 
> oid indexes are concatinated together to the end of the OID.  You
> probably need to read a good book on MIBs to truly appreciate how it
> works though...  In short, managers can figure out ahead of time what
> the oid to use should be if they know the index values.
> 
> Shshank> Question # 2
> 
> Shshank> Do I need to give *valid* values for the indexing variables
> Shshank> in a table instance when I am creating that instance in the
> Shshank> sub-agent implementation and registering it with the master
> Shshank> agent ?  (By *valid*- I mean the actual values of those
> Shshank> parameters as present at the NE) Or can I just initialize
> Shshank> those variables to some default values ?
> 
> Your subagent can do whatever it wants if it registers the entire
> table instance with the master agent.  It itself is then responsible
> for whatever lies beneath the registered OID and you're responsible
> for returning the right results.  The master agent doesn't need to
> know about yoru indexes at all.
> 
> -- 
> Wes Hardaker
> Please mail all replies to net...@li...
> _______________________________________________
> Net-snmp-coders mailing list
> Net...@li...
> http://lists.sourceforge.net/mailman/listinfo/net-snmp-coders

RE: URGENT: [Fwd: Maximum size of a trap packet]

[Ning Z. <nz...@op...>]

Hello everyone,

Firstly, thanks to Sudheer and Wes for your kind reply.  This may not be the
right place to ask but...

Instead of using snmp v3, we are still using snmpv1-2.  I am not sure where
we originally obtained the snmp libraries, which were written in C and
modified from cmu's distribution.

Here are files in the libraries.

acl.c          context.c      mib.c          party.c        snmp_api.c
snmp_client.h  sw_snmp.h      acl.h          context.h      mib.h
party.h        snmp_api.h     snmp_impl.h    vssver.scc     asn1.c
md5.c          parse.c        snmp.c         snmp_auth.c    snmp_nt.c
asn1.h         md5.h          parse.h        snmp.h         snmp_client.c
snmp_protos.h

If you are familiar with this version of snmplib, could you point it out how
I can increase the maximum size of a trap packet?  So far we are limited to
80 characters per packet.

Many many thanks,
Ning



-----Original Message-----
From: Wes Hardaker [mailto:wjh...@uc...]
Sent: Thursday, November 30, 2000 5:28 PM
To: Sudheer Tumuluru
Cc: Net SNMP Coders; Net SNMP; Ning Zhu; Dave Shield
Subject: Re: URGENT: [Fwd: Maximum size of a trap packet]


>>>>> On Thu, 30 Nov 2000 13:28:55 -0800, Sudheer Tumuluru
<stu...@re...> said:

Sudheer> I originally sent this message about 10 days but it got
Sudheer> drowned in the flood of emails. Could somebody please help on
Sudheer> this issue? I got an email from one Ning Zhu asking me about
Sudheer> the exact same problem he's having. His messages follows mine
Sudheer> below...

AgentX packets are significantly larger than SNMP packets, so you
might have to increase it beyond what you have the snmp packet limit
set to.

However, 64000 would be enough I think...

--
Wes Hardaker
Please mail all replies to net...@li...

RE: snmpconf needs testing

[RADMAN S. <Ste...@CT...>]

continued:
did a make and pure "snmpconf" and I love it.

Btw: found a hardcoded reference to /usr/local in rsnmp:

IDCsoh@idc009:/build/cvs/net-snmp/local$ grep usr/local *.made
rsnmp.made:    open(OUT,"/usr/local/etc/snmpwalk -v 1 $i public
$rtopmib.$rtopid.$rtopnode | grep -v End|");

Guess this should be fixed in local/Makefile.

:)Stefan

-----Original Message-----
From: Michael J. Slifcak [mailto:msl...@is...]
Sent: Friday, 01 December, 2000 16:03
Cc: net...@li...
Subject: Re: snmpconf needs testing


Wes Hardaker wrote:
> 
> If anyone has a chance, build of the cvs snapshot and run "snmpconf -g
> basic_setup" and see if it looks ok.  (or just snmpconf to test other
> aspects of it).
> 
> If you don't want to do a make install, then:
> 
>   cd local
>   ./snmpconf -c snmpconf.dir -g basic_setup
> 
> --
> Wes Hardaker

I like it.

One very small problem arises when responding to prompts for
SNMPv3 user.  I can enter "mike noauth" for user name, then
"auth" for min. security level, which will generate the illegal line :

  rwuser mike noauth noauth

and the agent will complain.

So, please check for single token (or quoted token if username
has embedded whitespace).

Thanks for building a wonderful admin tool!
-Mike Slifcak, Internet Security Systems, Inc.
_______________________________________________
Net-snmp-coders mailing list
Net...@li...
http://lists.sourceforge.net/mailman/listinfo/net-snmp-coders

Re: Some observations with SilverCreek agent tester

[Michael J. S. <msl...@is...>]

Better test results summary for testing CVS branch agent on Linux.
Still can't create a SNMPv3 "context" on the SilverCreek tester.

Re: snmpconf needs testing

[Michael J. S. <msl...@is...>]

Wes Hardaker wrote:
> 
> If anyone has a chance, build of the cvs snapshot and run "snmpconf -g
> basic_setup" and see if it looks ok.  (or just snmpconf to test other
> aspects of it).
> 
> If you don't want to do a make install, then:
> 
>   cd local
>   ./snmpconf -c snmpconf.dir -g basic_setup
> 
> --
> Wes Hardaker

I like it.

One very small problem arises when responding to prompts for
SNMPv3 user.  I can enter "mike noauth" for user name, then
"auth" for min. security level, which will generate the illegal line :

  rwuser mike noauth noauth

and the agent will complain.

So, please check for single token (or quoted token if username
has embedded whitespace).

Thanks for building a wonderful admin tool!
-Mike Slifcak, Internet Security Systems, Inc.