- Libsmi - lists.ibr.cs.tu-bs.de

objects with hyphens in a SMIv2 MIB (imported from SMIv1) are not in the output using smidump
by Peter Postma 16 Dec '13

16 Dec '13

Dear list users, I am having problems using smidump (v0.4.8) to export a MIB to XML. The MIB is: INTEL-LAN-ADAPTERS-MIB, INTELLAN.mib, v 1.4.3 01/06/2004 (MIB Version = 1.4.5). This MIB is from Intel but provided by Dell in a package of MIBs, is SMIv2 formatted, but seems to be imported from SMIv1. It contains objects and labels with hyphens producing the [5] hyphen-in-object and [4] hyphen-in-label warnings. These are not parsed by smidump at all and are missing from the output (in my case, XML but also for smiv1 and smiv2 formats). Is there any way for smidump to parse these entries and provide them in the output, disregarding the warnings? With kind regards, Peter Postma

2 1

libsmi-0.4.8:dump-cm.c:469:Same expression on both sides of '>'
by David Binderman 02 Dec '13

02 Dec '13

Hello there, I just ran the static analysis tool "cppcheck" over the source code of libsmi-0.4.8. It said [dump-cm.c:469] -> [dump-cm.c:469]: (style) Same expression on both sides of '>'. Source code is if (x1 == x2 && y1> y1) con = 1; I suspect if (x1 == x2 && y1> y2) con = 1; might be better code. Regards David Binderman

2 1

[PATCH] fix clang / Xcode compilation error
by Gerald Combs 18 Oct '13

18 Oct '13

Compiling Wireshark with the current version of Apple's developer toolchain results in the following error: /usr/local/include/smi.h:319:1: error: empty struct has size 0 in C, size 1 in C++ [-Werror,-Wc++-compat] } SmiElement; Would it make sense to remove the __GNUC__ check in smi.h as per the attached patch?

2 1

'error: unknown object identifier label' even though object is defined
by James Gosnell 13 Aug '13

13 Aug '13

I'm trying to separate concerns and create some modules in individual files. I have a 'registrations' file, the 'notifications' file and a 'conformance' file. I'm having an issue where I'm getting the following warning and error and I can't seem to find the issue. I think the warning might be causing the error, but I'm not sure. I've attached all three source files and would appreciate any input. The command I'm running to produce the following smilint output is: smilint -p ./TEKNO-CONFORMANCE-MIB.txt ./TEKNO-REGISTRATIONS-MIB.txt ./ADMIN-ILM-MIB.txt Here's a breakdown of my files: TEKNO-CONFORMANCE-MIB.txt is my conformance objects I'm learning more about and trying to separate from the actual notification MIB TEKNO-REGISTRATIONS-MIB.txt is our high-level module that will describe our corporate product MIB trunks ADMIN-ILM-MIB.txt is an actual notifications MIB that imports the above files Here is the warning and error msg I get when I run the above `smilint` command: ./TEKNO-REGISTRATIONS-MIB.txt:1: warning: module `TEKNO-REGISTRATIONS-MIB' is already loaded, aborting parser on this file ./ADMIN-ILM-MIB.txt:184: unknown object identifier label `teknoConformanceGroups' Any input and feedback, even to the design of my MIB, would be welcome. Thank you! -- James Gosnell, ACP

3 4

crash on attempt to traverse mib nodes
by Jonathan Walker 06 Jun '13

06 Jun '13

Hello, I'm finding that libsmi crashes when I remove every "RFC" mib from "mibs/ietf" directory, and then execute a program which performs the following: 1. Add each directory under "mibs" to the smi path (smiSetPath). 2. Traverse each directory under "mibs", and for each mib found get the module and traverse each node in the module. It tends to crash at attempt to traverse nodes of a mib which uses IMPORTS of an RFC- prefixed mib. Though I understand how missing dependencies could cause problems loading/traversing a mib, I'm not sure I understand why it's crashing and no error logs are present. Thanks, Jonathan ________________________________ This e-mail and any attachments are confidential. If it is not intended for you, please notify the sender, and please erase and ignore the contents.

2 5

smidump --format=smiv2 output contains some errors in compliance SYNTAX/WRITE-SYNTAX clauses.
by Jens Engel 06 Jun '13

06 Jun '13

Hello, I just stumbled over the following problem in libsmi 0.4.8 and development trunk-tip rev34557 (as of today). The problem resides in the SYNTAX/WRITE-SYNTAX clauses in COMPLIANCE statements. Instead of a "type" the invalid term "(null)" is present. AFFECTED-MIBS (at least): - BRIDGE-MIB - SNMP-VIEW-BASED-ACM-MIB - ... => grep "(null)" *-MIB EXAMPLE: smidump --format=smiv2 BRIDGE-MIB ... OBJECT dot1dStpPriority SYNTAX (null) (0 | 4096 | 8192 | 12288 | 16384 | 20480 | 24576 | 28672 | 32768 | 36864 | 40960 | 45056 | 49152 | 53248 | 57344 | 61440) ... OBJECT dot1dStpPortPriority SYNTAX (null) (0 | 16 | 32 | 48 | 64 | 80 | 96 | 112 | 128 | 144 | 160 | 176 | 192 | 208 | 224 | 240) BUT: Original BRIDGE-MIB that is distributed w/ libsmi contains: OBJECT dot1dStpPriority SYNTAX Integer32 (0 | 4096 | 8192 | 12288 | 16384 | 20480 | 24576 | 28672 | 32768 | 36864 | 40960 | 45056 | 49152 | 53248 | 57344 | 61440) ... OBJECT dot1dStpPortPriority SYNTAX Integer32 (0 | 16 | 32 | 48 | 64 | 80 | 96 | 112 | 128 | 144 | 160 | 176 | 192 | 208 | 224 | 240) smidump --format=smiv2 SNMP-VIEW-BASED-ACM-MIB ... OBJECT vacmViewTreeFamilyMask WRITE-SYNTAX (null) (SIZE(0)) ... SHOULD-BE (as in original MIB distributed with libsmi): ... OBJECT vacmViewTreeFamilyMask WRITE-SYNTAX OCTET STRING (SIZE(0)) I assume that smidump does not have the info (only a NULL pointer) and just dumps it as textual info. NOTE: "smidiff <ORIGINAL-MIB> <SMIDUMPED-MIB>" does not seem to show a difference (except warnings, ...). Ciao, Jens

3 2

Segfault on quering type SNMPv2-TC::RowStatus
by Luiz Angelo Daros de Luca 28 May '13

28 May '13

Hello, This simple command: smiquery type SNMPv2-TC::RowStatus is crashing smiquery. $ smiquery --version smiquery 0.4.8 $ smiquery type SNMPv2-TC::RowStatus Type: SNMPv2-TC::RowStatus Basetype: Enumeration Parent Type: Enumeration Numbers: active(1) notInService(2) notReady(3) createAndGo(4) createAndWait(5) destroy(6) Declaration: TEXTUAL-CONVENTION Status: current Segmentation Fault Is this a known problem? Regards, --- Luiz Angelo Daros de Luca, Me. luizluca(a)gmail.com

2 1

smiGetFirstRevision() does not return most recent revision
by Jonathan Walker 21 Mar '13

21 Mar '13

Hello, I have a MIB that currently has 70 revisions (IANAifType-MIB), and yet smiGetFirstRevision() is returning the 31st revision, and not the first. I'm using a DLL that I compiled myself, although I did not make any changes to the code base. LibSMI version 0.4.5. Please let me know if you require more information. Thank you, Jonathan ________________________________ This e-mail and any attachments are confidential. If it is not intended for you, please notify the sender, and please erase and ignore the contents.

2 1

Smidump python output issues
by Tuure Laurinolli 18 Mar '13

18 Mar '13

Hi, It appears that smidump turns the variable list defined in a NOTIFICATION-TYPE into a dict. This seems fine on the surface, but does not preserve ordering of the variables. As per section 8.1 of RFC 2578, the sequence of variables should be ordered. Is there a known workaround for this?

1 0

Libsmi 64bit windows
by Doron Tsur 28 Nov '12

28 Nov '12

Hi, We've been using libsmi for a long time and it's been serving us well. I'm in the process of porting one of our infrastructures project to 64bit systems. I've been having difficulties understanding what to compile in order to create a 64 bit windows lib. The version I'm currently using (don't laugh) is 0.2.13 which is well before my time. Could you assist me with that? Good Day, Doron Tsur

2 1