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

How to read in a mib file (text format)
by Frances Lung 28 Dec '00

28 Dec '00

Hi All, I'm new to libsmi and just download libsmi 0.2.12. Does libsmi can read in a mib file at run time and build a tree structure internally (I mean in memory)? Does smiLoadModule(...) will do the job? Thanks a bunch. Frances -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

net-snmp agent support status
by Bill Anderson 22 Dec '00

22 Dec '00

Can anyone tell me if the libsmi, smiquery netsnmp output format is compatable with the UCD 4.2 release requirements. I've taken a quick look at the netsnmp sample agent which is mib2c generated versus what libsmi generates for the same mib and they appear to differ . Im new to both of these toolkits so I appreciate any help. Thanks Bill -- Bill Anderson wda(a)ironstream.com IronStream, Inc. Phone: (978) 275-9555 x23 150 Industrial Ave. East http://www.ironstream.com Lowell, MA 01852 fax: (978) 275-9994

2 1

Re: [libsmi] Problem on AIX
by Bill Fenner 07 Dec '00

07 Dec '00

>/bin/sh ../libtool --mode=compile /usr/vac/bin/cc -DHAVE_CONFIG_H -I. >-I. -I.. -c parser-smi.tab.c >/usr/vac/bin/cc -DHAVE_CONFIG_H -I. -I. -I.. -c parser-smi.tab.c -DPIC >1506-507 (W) No licenses available. Contact your program supplier to add >additional users. Compilation will proceed shortly. >mv -f parser-smi.o parser-smi.tab.lo >mv: cannot rename parser-smi.o to parser-smi.tab.lo: >No such file or directory Assuming that the license warning meant what it said about compilation proceeding when a license became available, is there a "parser-smi.tab.o" in that directory? It looks to me like cc should have compiled parser-smi.tab.c into parser-smi.tab.o, but libtool for some reason thinks it will be parser-smi.o . If so, this does look like a libtool problem. You could try configuring libsmi to not try to built a shared library with "./configure --disable-shared", since it at least seems to be the shared library that it's having trouble building. (I can't say that it won't have trouble building a static-only library, but it's at least worth a shot). Bill -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

Problem on AIX
by Harrie Hazewinkel 07 Dec '00

07 Dec '00

HI, I was wondering if someone else have had this problem and what the soultion is. System: AIX 4.3 libtool version: 1.3.5 libsmi version 0.2.5 bison: GNU Bison version 1.28 /bin/sh ../libtool --mode=compile /usr/vac/bin/cc -DHAVE_CONFIG_H -I. -I. -I.. -c parser-smi.tab.c /usr/vac/bin/cc -DHAVE_CONFIG_H -I. -I. -I.. -c parser-smi.tab.c -DPIC 1506-507 (W) No licenses available. Contact your program supplier to add additional users. Compilation will proceed shortly. mv -f parser-smi.o parser-smi.tab.lo mv: cannot rename parser-smi.o to parser-smi.tab.lo: No such file or directory make[2]: *** [parser-smi.tab.lo] Error 1 make[2]: Leaving directory `/raid/harrie/ConductorSNMP/obj/powerpc-ibm-aix-4.3.x/libsmi/lib' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/raid/harrie/ConductorSNMP/obj/powerpc-ibm-aix-4.3.x/libsmi' make: *** [all-recursive-am] Error 2 harrie@purple$ I have the idea 'libtool' screws up. In perticular the line below. output_obj=`$echo "X$srcfile" | $Xsed -e 's%^.*/%%' -e 's%\..*$%%'`.${objext} Unfortenate this is libtool and libsmi makes use of it. Cheers, Harrie -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

2 1

question about trap oids
by Dan Stanger 14 Nov '00

14 Nov '00

i am sure that this question has been asked before, but i could not find any documentation on it. why does the trap oid have a 0 before the trap number? where is this added to the oid? thanks, dan stanger dan.stanger(a)ieee.org -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

Re: [libsmi] question about trap oids
by Frank Strauss 14 Nov '00

14 Nov '00

Dan> why does the trap oid have a 0 before the trap number? Dan> where is this added to the oid? This is done to ensure SMIv2 NOTIFICATION-TYPE definitions that can unabiguously be mapped to SNMPv1 trap identifications. See RFC 2578, Section 8.5. -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

libsmi release 0.2.8
by Frank Strauss 13 Nov '00

13 Nov '00

Hi! Release 0.2.8... o introduces a new behavior of those smidump drivers that create multiple files (the internal driver interface has changed significantly), o reorganizes the test suite, and adds 5 new tests to the test suite (the tar.gz is now >2MB :-/), o checks for 64 bit types during the configure process and uses the result also for *printf() formats, o adds --with-pathseparator configure option (useful in cygwin environments?), o updates the IANA maintained modules, and o fixes some bugs. See the ChangeLog file for more details. As usual the current release can be found at http://www.ibr.cs.tu-bs.de/projects/libsmi/ ftp://ftp.ibr.cs.tu-bs.de/pub/local/libsmi/ Frank -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

Re: [libsmi] libsmi release 0.2.7
by Frank Strauss 09 Nov '00

09 Nov '00

Yigal> On a slightly different topic: Yigal> I added "-Wall -Wformat" to CFLAGS in tool/Makefile. Yigal> It looks like this: Yigal> CFLAGS = -g -O2 -Wall -Wformat Yigal> The -Wformat checks the validity of printfs syntax. Yigal> I discovered several places with printf syntax problems. If you add Yigal> those flags and recompile you will see those warnings/errors. Yigal> I suggest to add both flags as a standard to CFLAGS in Makefile. This Yigal> will help in creating a better quality source code. I should keep in remind to usually add it to the --with-cflags options when I build using gcc. However, I did not add -Wall to the CFLAGS by default, because it's compiler specific. We could do it in cases where configure finds and uses gcc, but that would be just half of the story. BTW, -Wall includes -Wformat. On which platform did you get the warnings? On Linux everything is fine. No warnings at all with -Wall. If your warnings appear on W32, could you please send me (not to the list) a dump of your make output? -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0

Re: [libsmi] libsmi release 0.2.7
by Frank Strauss 09 Nov '00

09 Nov '00

Yigal> I built the new libsmi described below on NT4 using cygwin. The built Yigal> went ok. I use the following config file where I changed the path Yigal> directive to include : in front of the path. Yigal> ================<< start quote ============================ Yigal> # Last edit: Thu Nov 09 11:39:37 2000 Yigal> # By: Yigal Hochberg Yigal> # Yigal> # File: smi.cfg Yigal> # libsmi initialization file Yigal> # Yigal> # path of mib modules Yigal> # you can use the posix notation: //x/<path> to specify drive x: Yigal> # e.g //c/dmhsnmp-agent/mibdir/ Yigal> # Yigal> path ://f/yigal/dmh-libsmi/libsmi-0.2.7/mibs/ietf/ Yigal> path :../mibs/ietf/ Yigal> path :../mibs/iana/ Yigal> ================== end quote >>============================== Yigal> Unfortunately with or without the : smidump cannot find files in the Yigal> mib directories. Yigal> Any idea what is wrong? Note that up to and including libsmi 0.2.5 Yigal> smidump works very well in this environment. The path separator character on `#ifdef _WIN32' systems is `;' instead of `:'. I cannot confirm the comments of your configuration file on the `posix notation'. I would expect that something like path ;f:\yigal\dmh-libsmi\libsmi-0.2.7\mibs\ietf path ;..\mibs\ietf\ path ;..\mibs\iana\ is ok. -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

2 2

Re: [libsmi] libsmi release 0.2.7
by Frank Strauss 09 Nov '00

09 Nov '00

>> The path separator character on `#ifdef _WIN32' systems is `;' instead >> of `:'. Yigal> I changed the config file to use ; instead of : Yigal> This works fine. Yigal> The problem is that the different path-separator char (; vs. :) now Yigal> requires different configuration file for wintel environment and unix Yigal> environment. Ok. So, what would you think about a configure option for optionally specifying the path separator character? -- !! This message is brought to you via the `libsmi' mailing list. !! Please do not reply to this message to unsubscribe. To subscribe or !! unsubscribe, send a mail message to <libsmi-request(a)ibr.cs.tu-bs.de>. !! See http://www.ibr.cs.tu-bs.de/projects/libsmi/ for more information.

1 0