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

patch to enhance range unions in smidump generated python & perl
by Randy Couey 13 Jul '06

13 Jul '06

The python & perl smidump modules (and possibly others) don't generate useful code for size/range constraint unions. For example, an object defined with syntax: SYNTAX OCTET STRING (SIZE (1|8)) will result in the following dumped python code: "syntax" : { "type" : { "basetype" : "OctetString", "range" : { "min" : "1", "max" : "1" } "range" : { "min" : "8", "max" : "8" }, }, } This isn't very useful, since the second assignment to the 'range' dictionary key blows away the first one. I have written a patch for the python & perl dumpers to fix this problem. It creates a new key called 'ranges' which references a list of range dictionaries. To ensure backward compatibility, the last 'range' key/value pair is retained. With this patch, the above syntax clause will now get dumped as: "syntax" : { "type" : { "basetype" : "OctetString", "ranges" : [ { "min" : "1", "max" : "1" }, { "min" : "8", "max" : "8" }, ], "range" : { "min" : "8", "max" : "8" }, }, }, This allows knowledgeable apps to iterate through the 'ranges' list and properly deal with constraint unions, while older apps can still use the original, but possibly incorrect, 'range' key. thanks, --randy

2 5

Re: Antwort: Re: [libsmi] patch to enhance range unions in smidump generated python & perl
by Randy Couey 12 Jul '06

12 Jul '06

[...] > I am using the python output format to get MIB module information > (mainly for OID->ObjectName resolution and type information) > for usage based on the PySMNP 3.5.x framework > <http://pysnmp.sourceforge.net>. > I am currently not using the range info. > Therefore, I should have no problem with either change of the range > info. > > PySNMP 4.x has also a script (another one) that uses the > smidump python output format to generate SNMP managed object classes > (and MIB modules). > It will probably have a problem with any change in the range info. > It will be nice to inform the project maintainer of this change. That's actually why I submitted the patch :) I am currently using libsmi2pysnmp from PySNMP 4.x. I have already posted a patch to the pysnmp mailing list that will take advantage of the python smidump changes I've provided to you. I specifically designed both patches so that nothing should break for those users who don't simultaneously upgrade both patches. If you integrate my libsmi patch, I'm sure the pysnmp maintainers will accept by libsmi2pysnmp patch as well. thanks, --randy

1 0

[PATCH] Memory leak in smidump netsnmp output
by Adam Kropelin 31 May '06

31 May '06

smidump generates netsnmp wrappers that appear to contain a small memory leak if snmp_open() fails. For example... > int powernet_mib_mgr_get_upsPhaseResetValues(struct snmp_session *s, upsPhaseResetValues_t **upsPhaseResetValues) > { > struct snmp_session *peer; > struct snmp_pdu *request, *response; > struct variable_list *vars; > int status; > > request = snmp_pdu_create(SNMP_MSG_GETNEXT); request is allocated... > snmp_add_null_var(request, upsPhaseResetMaxMinValues, sizeof(upsPhaseResetMaxMinValues)/sizeof(oid)); > > peer = snmp_open(s); > if (!peer) { > return -1; ...and here we potentially return without freeing it. > } > > status = snmp_synch_response(peer, request, &response); snmp_synch_response() frees request in all cases, so there are no leaks after this point. > if (status != STAT_SUCCESS) { > if (response) snmp_free_pdu(response); > snmp_close(peer); > return -2; > } > > ... My knowledge of netsnmp is limited, but I believe the analysis above is correct. Below is a small patch to solve the problem. --Adam --- libsmi-0.4.5/tools/dump-netsnmp.c 2005-11-25 04:13:58.000000000 -0500 +++ libsmi-0.4.5/tools/dump-netsnmp.new.c 2006-05-30 18:30:36.000000000 -0400 @@ -1149,6 +1149,7 @@ "\n" " peer = snmp_open(s);\n" " if (!peer) {\n" + " snmp_free_pdu(request);\n" " return -1;\n" " }\n" "\n"

2 1

libsmi release 0.4.5
by Frank Strauß 29 May '06

29 May '06

Hi! A few days ago, I bundled a snapshot to a 0.4.4 release (without any announcement on this list). Some people found that a file was missing so that it was not possible to compile parts of the package. So, here we go with 0.4.5. I took also the chance to catch up with the current IETF MIB files. As always, thanks to the people who submitted bug reports and patches! http://www.ibr.cs.tu-bs.de/projects/libsmi/ ftp://ftp.ibr.cs.tu-bs.de/pub/local/libsmi/ Enjoy, -frank

1 0

libsmi 0.4.4: building problems
by Jose Pedro Oliveira 28 May '06

28 May '06

Hi, This is a report of two building problems that make the building process fail, a report of test failures and a small patch to the man pages: 1) build failure due to missing file The latest version of libsmi (0.4.4) fails to build due to a missing file - dump-svg-script-js. This file hasn't been included in the libsmi-0.4.4 tarball but exists in svn. 2) build failure due to FORTIFY_SOURCE This version of libsmi (and at the least 0.4.3) also fail to build in Fedora Core 4 and Fedora Core 5 systems if we use the same CFLAGS as in the base distro. This failure is due to FORTIFY_SOURCE, a buffer overflow prevention/detection mechanism built in gcc and glibc. This mechanism replaces almost all C library functions that deal with buffer copies, namely snprintf. The attached patch - libsmi-0.4.3-snprintf.h.patch - fixes this problem. 3) test failures The "make check" step reports two test failures. 4) man pages The second patch - libsmi-0.4.4-lib-manpages.patch - just replaces the pattern "@prefix@/include" by "@includedir@" in several man pages. Regards, jpo -- José Pedro Oliveira * mailto: jpo(a)di.uminho.pt * http://gsd.di.uminho.pt/jpo * * gpg fingerprint = F9B6 8D87 859D 1C94 48F0 84C0 9749 9EB5 91BD 851B *

2 1

FW: [libsmi] Regarding Unsigned64 interpretation by SMILINT
by Rohit Palthad Venkataramana (rovenkat) 17 May '06

17 May '06

Hi , I am using the windows version of libsmi , and instead of setting the SMIPATH env variable , I want to pass configuration file(containing the mibs directory path) while compiling a mib file . I tried the below steps without any success : I defined the path variable in a Configuration file (libsmiConfig) and excecuted the smilint still it did not recognize the mibModules The configuration file contains - Path :D:\Documents and Settings\rovenkat\smi\mibs\ietf ( I tried both types I mean Path and path ) Executed statement : smilint -m -s -l 6 -c "D:/Documents and Settings/rovenkat/smi/libsmiConfig" "D:/Documents and Settings/rovenkat/mibs/CISCO-IP-URPF-MIB.my" Errors : D:\Documents and Settings\rovenkat\mibs\CISCO-IP-URPF-MIB.my:25: [1] {module-not-found} failed to locate MIB module `SNMPv2-SMI' D:\Documents and Settings\rovenkat\mibs\CISCO-IP-URPF-MIB.my:29: [1] {module-not-found} failed to locate MIB module `SNMPv2-CONF' D:\Documents and Settings\rovenkat\mibs\CISCO-IP-URPF-MIB.my:32: [1] {module-not-found} failed to locate MIB module `SNMPv2-TC' D:\Documents and Settings\rovenkat\mibs\CISCO-IP-URPF-MIB.my:34: [1] {module-not-found} failed to locate MIB module `SNMP-FRAMEWORK-MIB' ------ many more Can you plz let me know what is wrong in the configuration file . Thanks Rohit

1 0

RE: [libsmi] Regarding Unsigned64 interpretation by SMILINT
by Rohit Palthad Venkataramana (rovenkat) 16 May '06

16 May '06

Hi Frank , Thanks you for ur modification . Is this fixed on windows version also . How can I set the SMIPATH ( mib directory path ) at runtime for smilint ? . How do I write configuration file for smilint ?, so that I can specify the mibs path in the configuration file . Can u plz help me out . I am using the windows binary of libsmi . Thanks Rohit -----Original Message----- From: Frank Strauß [mailto:strauss@ibr.cs.tu-bs.de] Sent: Monday, May 15, 2006 7:24 PM To: Rohit Palthad Venkataramana (rovenkat); libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] Regarding Unsigned64 interpretation by SMILINT Juergen Schoenwaelder wrote: > On Thu, May 04, 2006 at 01:39:04PM +0530, Rohit Palthad Venkataramana (rovenkat) wrote: > >> We import it using CISCO-TC MIB . > > OK. So you essentially have a definition for Unsigned64 which should > be valid as long as it is used by SMIv2 modules but it should be > invalid if used by SPPI modules and libsmi always complains. I am not > sure how easy it will be to fix this because libsmi treats Unsigned64 > internally as a base type and Counter64 actually as a type derived > from Unsigned64 (basically following the SMIng model). Frank, do you > have an idea how to address this name clash? > > For now, I have written a test case for libsmi and checked it into the > SVN repository which demonstrates the problem. I have just modified the parser so that it now accepts (and warns) about such type definitions that match existing SPPI or SMI base type names. $ SMIPATH=:test/mibs ./tools/smilint test/mibs/LIBSMI-TEST-018-MIB test/mibs/LIBSMI-TEST-018-MIB:41: warning: definition of identifier `Unsigned64' which is already a SMI or SPPI basetype

3 4

RE: [libsmi] Regarding Unsigned64 interpretation by SMILINT
by Rohit Palthad Venkataramana (rovenkat) 15 May '06

15 May '06

We import it using CISCO-TC MIB . Thanks Rohit -----Original Message----- From: Juergen Schoenwaelder [mailto:j.schoenwaelder@iu-bremen.de] Sent: Thursday, May 04, 2006 1:08 PM To: Rohit Palthad Venkataramana (rovenkat) Cc: libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] Regarding Unsigned64 interpretation by SMILINT On Thu, May 04, 2006 at 01:02:31PM +0530, Rohit Palthad Venkataramana (rovenkat) wrote: > Sorry for the later response , I was out of the office . > > The below is the definition of Unsigned64 object > > ciscoTestRate64 OBJECT-TYPE > SYNTAX Unsigned64 > UNITS "bits/second" > MAX-ACCESS read-only > STATUS current > DESCRIPTION > "The committed policing rate. This is the sustained > rate permitted by policing." > > ::= { ciscoTestRate 11 } > > > The mib is in my controll .I even checked for other mibs also ,smilint > gives same error for Unsigned64 object . So how is Unsigned64 defined / imported? The SMIv2 itself does not have an Unsigned64 data type, only SPPI has it. /js -- Juergen Schoenwaelder International University Bremen <http://www.eecs.iu-bremen.de/> P.O. Box 750 561, 28725 Bremen, Germany

3 2

RE: [libsmi] Regarding Unsigned64 interpretation by SMILINT
by Rohit Palthad Venkataramana (rovenkat) 07 May '06

07 May '06

Hi , SMILINT is not able to detect the mib modules . Is their set the mibs directory without setting the smipath environment variable .. Thanks Rohit -----Original Message----- From: Juergen Schoenwaelder [mailto:j.schoenwaelder@iu-bremen.de] Sent: Thursday, May 04, 2006 2:03 PM To: Rohit Palthad Venkataramana (rovenkat) Cc: libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] Regarding Unsigned64 interpretation by SMILINT On Thu, May 04, 2006 at 01:39:04PM +0530, Rohit Palthad Venkataramana (rovenkat) wrote: > We import it using CISCO-TC MIB . OK. So you essentially have a definition for Unsigned64 which should be valid as long as it is used by SMIv2 modules but it should be invalid if used by SPPI modules and libsmi always complains. I am not sure how easy it will be to fix this because libsmi treats Unsigned64 internally as a base type and Counter64 actually as a type derived from Unsigned64 (basically following the SMIng model). Frank, do you have an idea how to address this name clash? For now, I have written a test case for libsmi and checked it into the SVN repository which demonstrates the problem. /js -- Juergen Schoenwaelder International University Bremen <http://www.eecs.iu-bremen.de/> P.O. Box 750 561, 28725 Bremen, Germany

2 1

RE: [libsmi] Regarding Unsigned64 interpretation by SMILINT
by Rohit Palthad Venkataramana (rovenkat) 04 May '06

04 May '06

Hi , Sorry for the later response , I was out of the office . The below is the definition of Unsigned64 object ciscoTestRate64 OBJECT-TYPE SYNTAX Unsigned64 UNITS "bits/second" MAX-ACCESS read-only STATUS current DESCRIPTION "The committed policing rate. This is the sustained rate permitted by policing." ::= { ciscoTestRate 11 } The mib is in my controll .I even checked for other mibs also ,smilint gives same error for Unsigned64 object . Thanks Rohit -----Original Message----- From: Juergen Schoenwaelder [mailto:j.schoenwaelder@iu-bremen.de] Sent: Sunday, April 30, 2006 3:15 AM To: Rohit Palthad Venkataramana (rovenkat) Cc: libsmi(a)ibr.cs.tu-bs.de Subject: Re: [libsmi] Regarding Unsigned64 interpretation by SMILINT On Sun, Apr 30, 2006 at 01:04:57AM +0530, Rohit Palthad Venkataramana (rovenkat) wrote: > Hi LIBSMI Folks , > > When a MIB containing Unsigned64 object is compiled using > SMILINT . it gives the below error : > > xxxx.my:22: [0] {keyword-illegal-in-mib} the SPPI construct/keyword > `Unsigned64' may not be used in a MIB > > > Although the Unsigned64 Object is defined in the MIB . Please show us the definition of Unsigned64 you are using and the definition of the MIB object which uses Unsigned64. If the MIBs exist under your control, simply prefix the definitions properly. /js -- Juergen Schoenwaelder International University Bremen <http://www.eecs.iu-bremen.de/> P.O. Box 750 561, 28725 Bremen, Germany

2 1