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

smilint crash when INTGER values are equal
by Arndt Jonasson 31 Jan '08

31 Jan '08

BUG DEFINITIONS ::= BEGIN IMPORTS private, MODULE-IDENTITY, OBJECT-TYPE, Integer32, Unsigned32, Counter32, Counter64, Gauge32, IpAddress FROM SNMPv2-SMI; top MODULE-IDENTITY LAST-UPDATED "" ORGANIZATION "" CONTACT-INFO "" DESCRIPTION "" ::= { private 2 } obj OBJECT-TYPE SYNTAX INTEGER {blue(1),yellow(1),green(1)} MAX-ACCESS read-create STATUS current DESCRIPTION "" ::= { top 1 } END The above MIB gives a segmentation violation with smilint. With just two symbols defined for 'obj', it gives the expected warning: bug.mib:16: redefinition of number `1' in number enumeration

2 1

an "unspecified" error while validating a MIB
by Pete Wilson 26 Dec '07

26 Dec '07

Had an "unspecified" error while validating a MIB. The file report.txt says: mibs/SPM-BASE-MIB.MIB:13: [1] {internal-other} syntax error, unexpected DISPLAY_HINT Additional descriptions of some error/warning messages: Error: internal-other (level 1) Message: %s Description: An unspecified error occured. Please submit a detailed bug report to the libsmi maintainer or the libsmi mailing list at <libsmi(a)ibr.cs.tu-bs.de>. The first few lines of my MIB: SPM-BASE-MIB DEFINITIONS ::= BEGIN IMPORTS Counter32, Counter64, Gauge32, Integer32, IpAddress, MODULE-IDENTITY, NOTIFICATION-TYPE, Opaque, OBJECT-TYPE, OBJECT-IDENTITY, TimeTicks, Unsigned32, enterprises FROM SNMPv2-SMI TEXTUAL-CONVENTION, RowStatus, DisplayString, TruthValue, DISPLAY-HINT FROM SNMPv2-TC OBJECT-GROUP FROM SNMPv2-CONF; Can anyone tell me how I can fix this error? Thanks! -- Pete ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping

2 1

sizeof(SmiUnsigned32) = 8?
by Arndt Jonasson 03 Dec '07

03 Dec '07

In dump-xsd.c, the variable numSubranges is used in several places for allocating an array of SmiUnsigned32. In one of those places, the number 4 is used literally instead of sizeof(SmiUnsigned32). There exist platforms where sizeof(SmiUnsigned32) == 8 (64-bit AMD 64, for example), and there the program may crash.

2 1

Messages on Dup Imports
by David T. Perkins 16 Nov '07

16 Nov '07

HI, In cleaning up a MIb module, it had the following imports: IMPORTS OBJECT-TYPE FROM SNMPv2-SMI OBJECT-TYPE, IpAddress, MODULE-IDENTITY, enterprises FROM SNMPv2-SMI DisplayString, TruthValue FROM SNMPv2-TC; With a level 5 or greater, smilint generated the warning: warning: identifier `OBJECT-TYPE' imported from module `SNMPv2-SMI' is never used This is misleading. I was expecting the following two warnings: 1) duplicate definition of "OBJECT-TYPE" 2) all of the items from a module must be imported together Regards, /david t. perkins

2 3

Missing warning
by David T. Perkins 07 Nov '07

07 Nov '07

HI, I'm doing some cleanups of sme MIB modules. In them are many examples of tables that have a RowStatus column (with no restriction of values) and have MAX-ACCESS of "read-write" on the columns. I was expecting to see a warning that "read-create" should have been used instead of "read-write". If this warning is presently generated, I suggest that it be added. Regards, /david t. perkins

2 2

Bug in pattern generation for ranges
by Arndt Jonasson 06 Nov '07

06 Nov '07

In the file tools/fortopat.c is this code: for (range = smiRange, lp = 0; range; range = smiGetNextRange(range)) { lengths[lp++] = smiRange->minValue.value.unsigned32; lengths[lp++] = smiRange->maxValue.value.unsigned32; This ought to be for (range = smiRange, lp = 0; range; range = smiGetNextRange(range)) { lengths[lp++] = range->minValue.value.unsigned32; lengths[lp++] = range->maxValue.value.unsigned32; or the ranges after the first one will effectively be ignored.

2 3

Some cleanups on the package
by David T. Perkins 01 Nov '07

01 Nov '07

HI, I just did a checkout through subversion and found a few items that would be nice if they were cleaned up. These include: 1) the README file leaves out a few pieces of info that would have saved me from having to figure them out. These include: a) it doesn't mention that you have to run script autogen.sh to generate the configure script b) the manifest is out of date, since it doesn't include autogen.sh nor smi.conf-example c) it would be helpful to list the files that were to be copied for "make install", and the source and dest directories d) More info needs to be given about interesting options to set for the configure script. (I didn't use the prefix option, and then make install failed (since I don't have admin capabilities on the system). I then when back and used --prefix=$HOME on configure, then rebuilt, then installed.) I'm not sure yet if there are "features" that I need to turn on or off. e) There is a passing reference to the smi_config(3) man page. A little more info is needed, with some suggestions for setting up a local (.smirc) and system wide (smi.conf) files. 2) The "make check" failed. It appears that the expected results were not updated after libsmi was changed. And there might be a bug or two. Someone should check. 3) I looks like the list of unversioned files to ignore should be updated so that svn doesn't give bogus warnings. 4) Note that on the WEB pages, it would be nice if there was a style sheet for printers so that when, say, a man page was printed, the left side nav <div> was not shown. Regards, /david t. perkins

1 0

smiFreeData and BITS
by Arndt Jonasson 02 Oct '07

02 Oct '07

I think there is an error in the function smiFreeData in lib/data.c. For a MIB of mine, smidump had a segmentation fault there, while the memory was being freed on exit. The code is: if (objectPtr->typePtr) { if ((objectPtr->typePtr->export.basetype == SMI_BASETYPE_OCTETSTRING || objectPtr->typePtr->export.basetype == SMI_BASETYPE_BITS)) { smiFree(objectPtr->export.value.value.ptr); It seems to be the case that for SMI_BASETYPE_OCTETSTRING, the contents is not a pointer, but an integer, and shouldn't be freed with smiFree. Then the correct code would be: if (objectPtr->typePtr) { if ((objectPtr->typePtr->export.basetype == SMI_BASETYPE_OCTETSTRING)) { smiFree(objectPtr->export.value.value.ptr);

2 2

errormacros.h
by Pete Wilson 25 Sep '07

25 Sep '07

Hi everyone -- In building libsmi for Win32, my very first hurdle is the absence of errormacros.h. It seems not to be included in the libsmi 0.4.5 zip file I downloaded the other day. Can some kind sould tell me where I can find this file? Thanks so much! - Pete Wilson ____________________________________________________________________________________ Catch up on fall's hot new shows on Yahoo! TV. Watch previews, get listings, and more! http://tv.yahoo.com/collections/3658

2 1

Win32 llibsmi intro doc
by Pete Wilson 25 Sep '07

25 Sep '07

Hi folks -- I'm interested in Win32 libsmi. Because so few Win32 guys outside this group know anything about it; and those who do know are (like me) afraid of it, I decided to start writing a user guide for Win32 libsmi development. It's at http://www.pwilson.net/Win32HowTo.html I'd appreciate it if you libsmi experts could take a look and tell me if this is or is not the right approach to introducing libsmi to the wider Win32 audience that it deserves. On another note, I'm stuck in my own development of this package. I have a libsmi.dll that I got from Luis, but I need a libsmi.obj. And my obsessive personality inists that I need a real, formal Visual Studio project/solution. Yigal or someone else, can yu help me with this? Thanks, everyone. -- Pete Wilson http://www/pwilson.net/ ____________________________________________________________________________________ Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games. http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow

2 1