Libsmi September 2002

libsmi@ibr.cs.tu-bs.de

9 participants
11 discussions

How to validate libsmi functionality?
by Mark W. Alexander 01 Oct '02

01 Oct '02

Greetings all, Thanks for your work on libsmi. I'm working on a python application to connect snmp traps to a TCP stream oriented management application and I've found libsmi to be exactly what I need to decode trap PDU's to tokens and arguments that the management application can parse. To do so, I've implemented a hand coded Python wrapper around (so far) smi_config (except SetErrorHandler), most of smi_module (no Revision functions) and some of smi_node (no Element or Refinement functions). The advantage to hand-coding over SWIG is that the smi data structures are properly reflected in real python objects. With SWIG, a nodes smiValue reference becomes a string representation of a C pointer, providing no real value to the python programmer. By properly producing the return objects, the python Node object contains a "real" Value object that properly represents the equivalent C structure. First question: Is there a test suite anywhere that I can use to help generate a python test suite to verify that I'm not mucking up the results of the underlying smi calls? Second question: Is anyone else interested in a full Python wrapper? I'm no SMI expert (beyond SNMP MIB's using it), and I can't really pursue much more functionality since I'm basically clueless about Elements, Refinements, Types, and Macros. If there _is_ a test suite, I could implement based on that, if not I could use advice and suggestions on where to learn more so I can extend as time permits. If nobody else cares, I'll just keep it as is until I need more from it. TIA -- Mark W. Alexander slash(a)dotnetslash.net

2 1

Problem in Indexing Table..
by Vivekanandan V 30 Sep '02

30 Sep '02

Hi, In Rfc:2674 (P-BRIDGE MIB), an Table Object, which is having read-write access is later used as an Index for another Table. Is this Correct??? To be more precise, dot1dRegenUserPriority is an object of Dot1dUserPriorityRegenEntry, with read-write access. In later part of the mib, the same dot1dRegenUserPriority is used as an Index for the Table dot1dPortOutboundAccessPriorityTable. Libsmi is not giving any Warnings/Errors for this?? So can I assume that this is a correct usage?? But the tool we are using gives an error saying Index element should be not-accessible. So which is correct?? Thanks & Regards, Vivekanandan.V *************************************************************************** This message is proprietary to Future Software Limited (FSL) and is intended solely for the use of the individual to whom it is addressed. It may contain privileged or confidential information and should not be circulated or used for any purpose other than for what it is intended. If you have received this message in error, please notify the originator immediately. If you are not the intended recipient, you are notified that you are strictly prohibited from using, copying, altering, or disclosing the contents of this message. FSL accepts no responsibility for loss or damage arising from the use of the information transmitted by this email including damage from virus. ***************************************************************************

2 1

smi.xsd
by Chris Ross 27 Sep '02

27 Sep '02

Does smi.xsd actually exist somewhere? Thanks.

2 1

Limited number of loaded modules?
by Heiko.Eichberger＠RSBICK.rohde-schwarz.com 27 Sep '02

27 Sep '02

Hi, does anybody know if there's a limited number of modules to load (with smiLoadModule)? Regards, Heiko

2 1

libsmi on Windows
by Henning Eggers 23 Sep '02

23 Sep '02

Hello, this is to anybody who works with smilib on Windows. I am trying to use libsmi-0.4.0 on WindowsNT. I am using VC6.0. The lib and my testing code compile with no problem but at run-time I see this assertion: ----------------------------------------------------- Debug Assertion Failed! Pogram: D:\projects\libsmitest\Debug\libsmitest.exe File: dbgheap.c Line: 1084 Expression: pHead->nBlockUse == nBlockUse ----------------------------------------------------- The code in question is: smiInit(NULL); smiSetPath("d:\\projects\\smi\\mibs\\ietf"); smiLoadModule("SNMPv2-SMI"); smiLoadModule("SNMPv2-TC"); smiLoadModule("SNMPv2-CONF"); smiLoadModule("SNMPv2-MIB"); The assertion gets fired on the last call to LoadModule. If I leave that call out, the program runs cleanly but doesn't do anything useful, of course. I saw a problem like this on this list in June but the answer was simply to call smiInit first which is the case with my code. Running the debugger shows that the error occurs when smiparse() gets called but this is what the call stack contains: _free_dbg(void * 0x00309630, int 2) line 1084 + 43 bytes __crtsetenv(const char * 0x0030aa90, const int 1) line 174 + 17 bytes LIBSMITEST! putenv + 54 bytes timegm(tm * 0x00000002) line 165 + 10 bytes mainCRTStartup() line 206 + 25 bytes KERNEL32! 77f1b9ea() Unfortunately I cannot trace into smiparse() because the debugger does not seem to be able to handle the flex/bison-code or it is simply not there. The whole thing does not happen on Linux (gcc)! Any help is appreciated! Henning

1 1

Re: [libsmi] SMI API
by barora＠hss.hns.com 19 Sep '02

19 Sep '02

Hi, This thread seems to be getting a bit to long !!! But, I am sorry to say that the problem persists with the pre0.4.1 code that you sent me. It works fine with 0.3.1 though!!! Am attaching the binary I compiled on my machine for smiquery. This still displays the language as SMIv1. (See attached file: smiquery.zip) - Regards, Brijesh Arora Frank Strauss <strauss(a)ibr.cs.tu-bs.de> on 09/18/2002 03:07:11 PM To: Brijesh Arora/HSS@HSS cc: libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] SMI API Hi! barora> I tried with libsmi 0.3.1and the language is correctly barora> identified as SMIv2 for this MIB so there seems to be a barora> problem with that in 0.4.0. barora> As I am behind a firewall, I cannot access the CVS code barora> through my CVS client. Is there some other way that I can get barora> the current snapshot of the CVS code ? (I downloaded barora> libsmi-current.tar.gz but that seems to be same as 0.4.0) I will send you a snapshot in a separate email. -frank -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/libsmi.

2 1

Re: [libsmi] SMI API
by barora＠hss.hns.com 18 Sep '02

18 Sep '02

Hi, I tried with libsmi 0.3.1and the language is correctly identified as SMIv2 for this MIB so there seems to be a problem with that in 0.4.0. As I am behind a firewall, I cannot access the CVS code through my CVS client. Is there some other way that I can get the current snapshot of the CVS code ? (I downloaded libsmi-current.tar.gz but that seems to be same as 0.4.0) - Regards, Brijesh Frank Strauss <strauss(a)ibr.cs.tu-bs.de> on 09/12/2002 03:06:46 PM To: Brijesh Arora/HSS@HSS cc: libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] SMI API Hi! Frank> Are you sure? I get ``Language: SMIv2'' for this MIB. barora> Absolutely! [...] BTW, what are the parameters by which the barora> language is guessed ? Could you give a try to the current snapshot from CVS? Perhaps, there is a bug in 0.4.0 due to the new support for SPPI PIBs which introduced new language detection code. The code evolved a bit since then, and from the current code I cannot reproduce the problem of wrongly identified MIB languages. The detection is based on imports from certain SMI language specific modules and from constructs like MODULE-IDENTITY, etc. -frank -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/libsmi.

2 1

Re: [libsmi] smidump question
by barora＠hss.hns.com 13 Sep '02

13 Sep '02

Hi, Try the -p option of smidump. It is used to preload some mibs that are are imported by the target MIB. Hope that helps, Brijesh Arora Senior Software Engineer Network Management Group Hughes Software Systems India. Vinoth Palaniappan <vinoth(a)mihy.mot.com> on 09/12/2002 11:37:05 PM To: libsmi(a)ibr.cs.tu-bs.de cc: (bcc: Brijesh Arora/HSS) Subject: [libsmi] smidump question Hi, I downloaded and built libsmi-0.4.0 in solaris-2.6 machine. I ran configure script with --prefix and --exec-prefix options and all got properly installed. When i ran smidump as: "smidump -f tree ./IF-MIB" in <prefix-dir>/share/mibs/ietf directory, i got the errors as follows. I understand that the location of mib modules to be imported needs to be given to the smidump utility. ./IF-MIB:6: failed to locate MIB module `SNMPv2-SMI' ./IF-MIB:9: failed to locate MIB module `SNMPv2-TC' ./IF-MIB:11: failed to locate MIB module `SNMPv2-CONF' . . Then i re-ran configure with ---with-smipath=<prefix-dir>/share/mibs/ietf:<prefix-dir>/share/mibs/irtf... and then re-built everything. But this time, the mib files were not copied to the <prefix-dir>/share/mibs/ directory. Again running the "smidump -f tree ./IF-MIB" in <install-dir>/mibs/ietf/ i got the same error. Can you help me with what i'm missing? ____________________________________________________________________________ ___________ Vinoth Motorola, India ____________________________________________________________________________ ___________ -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/libsmi.

1 0

smidump question
by Vinoth Palaniappan 12 Sep '02

12 Sep '02

Hi, I downloaded and built libsmi-0.4.0 in solaris-2.6 machine. I ran configure script with --prefix and --exec-prefix options and all got properly installed. When i ran smidump as: "smidump -f tree ./IF-MIB" in <prefix-dir>/share/mibs/ietf directory, i got the errors as follows. I understand that the location of mib modules to be imported needs to be given to the smidump utility. ./IF-MIB:6: failed to locate MIB module `SNMPv2-SMI' ./IF-MIB:9: failed to locate MIB module `SNMPv2-TC' ./IF-MIB:11: failed to locate MIB module `SNMPv2-CONF' . . Then i re-ran configure with ---with-smipath=<prefix-dir>/share/mibs/ietf:<prefix-dir>/share/mibs/irtf... and then re-built everything. But this time, the mib files were not copied to the <prefix-dir>/share/mibs/ directory. Again running the "smidump -f tree ./IF-MIB" in <install-dir>/mibs/ietf/ i got the same error. Can you help me with what i'm missing? ____________________________________________________________________________ ___________ Vinoth Motorola, India ____________________________________________________________________________ ___________

1 0

Re: [libsmi] SMI API
by barora＠hss.hns.com 12 Sep '02

12 Sep '02

Hi, barora> 1. I tried 'make install' but it failed as I do not have root barora> permission on the machine that I work on. I want to barora> install the man pages in my home directory. What do I need barora> to change in the makefile? Frank>> Then, I suggest to install everything in your $HOME. The source package Frank>> is flexibly configurable via configure. Please read the ``configure --help'' Frank>> output. You probably want to ``configure --prefix=$HOME''. Then run ``make'' Frank>> and ``make install'' as usual. barora>>> Thanks !!That worked. Actually I had been able to access the man pages barora>>> earlier too through some juggelry but i was trying something like barora>>> "man smiGetNode" which was not working. Found what I was looking barora>>> for in "man smi_node". barora> 2. About the SMI version, I also assumed that the SMI version is barora> guessed from the MIB but the result of the smiquery command barora> on the mib "libsmi/libsmi-0.4.0/mibs/ietf/IP-MIB" given as part barora> of the distribution gives the language as language as "SMIv1" barora> even though it should be SMIv2. Why is this happening ? Frank>>Are you sure? I get ``Language: SMIv2'' for this MIB. barora>>> Absolutely! I ran the command "smiquery module ./IP-MIB" in the barora>>> directory /libsmi/libsmi-0.4.0/mibs/ietf and here are the barora>>> results verbatim from the screen. barora>>> barora>>> Object: IP-MIB::ipMIB barora>>> Organization: IETF SNMPv2 Working Group barora>>> ContactInfo: Keith McCloghrie barora>>> barora>>> Postal: Cisco Systems, Inc. barora>>> 170 West Tasman Drive barora>>> San Jose, CA 95134-1706 barora>>> US barora>>> barora>>> Phone: +1 408 526 5260 barora>>> Email: kzm(a)cisco.com barora>>> Description: The MIB module for managing IP and ICMP implementations, barora>>> but excluding their management of IP routes. barora>>> Language: SMIv1 barora>>> Conformance: 1 barora>>> Loaded: no barora>>> Revision: Tue Nov 1 05:30:00 1994 barora>>> Description: [Revision added by libsmi due to a LAST-UPDATED clause.] barora>>> Revision: Sun Mar 31 05:30:00 1991 barora>>> Description: The initial revision of this MIB module was part of MIB- barora>>> barora>>> barora>>> The language shown is SMIv1 !!! barora>>> By the way, I got the same results for WWW-MIB, TCP-MIB and ISDN-MIB. barora>>> barora>>> Whereas I got the following result for BRIDGE-MIB which is actually SMIv1 barora>>> barora>>> Module: BRIDGE-MIB barora>>> Pathname: ./BRIDGE-MIB barora>>> Language: SMIv1 barora>>> Conformance: 1 barora>>> Loaded: no barora>>> barora>>> Seems to be some other problem? As as result the tools do not work on barora>>> these mibs as is gives SMI conformance errors!!! barora>>> BTW, what are the parameters by which the language is guessed ? - Brijesh Brijesh Arora Senior Software Engineer Hughes Software Systems Network Management Group India

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

Libsmi September 2002