> -----Ursprüngliche Nachricht----- > Von: Hans Jørgen Jakobsen > Gesendet: Mo. 25.07.2016 12:47 > An: venerabl...@googlemail.com > Kopie: net-snmp-users@lists.sourceforge.net >, , > Betreff: RE: snmptranslate doesn't show full MIB-Tree on equal Systems > > Found: > hjj@xxx:~/snmptr % grep 'write(2, ' strace-80.txt > write(2, "No log handling enabled - turnin"..., 52) = 52 > write(2, "Undefined OBJECT-GROUP (entPhysi"..., 94) = 94 > write(2, "Undefined OBJECT-GROUP (entPhysi"..., 96) = 96 > write(2, "Undefined OBJECT-GROUP (entPhysi"..., 94) = 94 > write(2, "Undefined OBJECT-GROUP (entPhysi"..., 98) = 98 > write(2, "Undefined OBJECT-GROUP (entityMa"..., 96) = 96 > write(2, "Undefined OBJECT-GROUP (entityLo"..., 97) = 97 > write(2, "Undefined OBJECT-GROUP (entityGe"..., 96) = 96 > write(2, "Undefined OBJECT-GROUP (entityPh"..., 98) = 98 > write(2, "Undefined OBJECT-GROUP (entityPh"..., 98) = 98 > write(2, "Undefined OBJECT-GROUP (entityPh"..., 97) = 97 > hjj@new:~/snmptr % grep 'write(2, ' strace-81.txt > write(2, "No log handling enabled - turnin"..., 52) = 52 > write(2, "Undefined OBJECT-GROUP (ifGenera"..., 117) = 117 > write(2, "Undefined OBJECT-GROUP (ifGenera"..., 117) = 117 > write(2, "Undefined OBJECT-GROUP (ifGenera"..., 117) = 117 > write(2, "Undefined OBJECT-GROUP (ifGenera"..., 117) = 117 > write(2, "Undefined OBJECT-GROUP (ifGenera"..., 117) = 117 > hjj@xxx:~/snmptr % > So there are some error messages...
Yes I know, however I can't fix it. My MIB files are all checked with MG-SOFT MIB Compiler, and the "undefined" objects are all in the MIB file. I can not find any issue at these files. I suppose that the different loading, of the MIB files could be the reason. How you can see the error messages are different on "equal" machines. Regarding your guesses: Good questions, maybe I should ask any code writers? Regards, Venerable > > Below are some guesses made without code reading! > > For some reason DOCS-IF-M-CMTS-MIB is not read at all on "81". It's only > opened once to read mib module name. > Why is it lost from the discovery of mibs in directory to the reading > phase? > > Could there be a table max on number of mib modules that can be handled? > > Your 2100+ mib modules seems high. > Use and load of mibs could be another talk! > > /hjj > > > On Thu, 21 Jul 2016, venerabl...@googlemail.com wrote: > >> Hi Hans, >> >> I have done three thinks. >> >> 1st I have created a MIB-Tree on both machines and I have compared this >> files. (snmptranslate -m ALL -Pu > -Tp) >> There are many OIDs different. >> >> So I have 2nd created the strace files on both machine with my TEST-OID >> "snmptranslate -m ALL -Pu > -Of .1.3.6.1.4.1.4491.2.1.6.2.1.1" >> And as you have assumed the sequence of the MIB files loading is totally >> different. >> So I think that could be the issue. However I can't solve it with "mibs +" >> at snmp.conf, >> because I have about 2000 MIB files for different manufactures. >> >> As 3rd I have build a new machine and if I can't believe it, snmptranslate >> translated me the full OID as > on the first machine. >> >> If you like you can look at the attached files, maybe you find somewhat. >> >> >> Regards, >> Venerable >> >>> -----Ursprüngliche Nachricht----- >>> Von: Hans Jørgen Jakobsen >>> Gesendet: Mi. 20.07.2016 14:05 >>> An: net-snmp-users@lists.sourceforge.net >>> , , >>> Betreff: Re: snmptranslate doesn't show full MIB-Tree on equal Systems >>> >>> I would try to run the program under truss(strace) looking for: >>> a) which files are opened >>> b) difference in sequence >>> c) size of read.. >>> >>> Even more desperate: >>> A forgotten alias. >>> A forgotten install of net-snmp in alternative path. >>> Move binaries between machines and see if it makes diffrence. >>> /hjj >>> >>> On Wed, 20 Jul 2016, venerabl...@googlemail.com wrote: >>> >>>> On 7/15/16, ler...@gmail.com wrote >>>>> What happens if you try it with mibReplaceWithLatest off? Found this: >>>>> " > target="_blank">https://docs.oracle.com/cd/E18752_01/html/816-5166/snmpcmd-1m.html >>>>>> -PR > Replace MIB objects using the last read MIB file. The parser >>>>> replaces MIB objects in its hierarchy whenever it sees a subidentifier >>>>> and name match. > Caution - Setting this option might result in an >>>>> incorrect >>>>> hierarchy. Can be also set with the configuration token > >>>>> mibReplaceWithLatest. >>>>>> Regards, >>>>> Lee >>>> >>>> I have set it to "no" on both machines, no changes. >>>> On Server1 no problems, on server2 the same problem as before. >>>> >>>> It seems to be really tricky. >>>> >>>> Regards, >>>> Venerable >>>> 2016-07-15 0:38 GMT+02:00 Lee : >>>> On 7/14/16, venerabl...@googlemail.com wrote: >>>> > Hi Hans, >>>> > >>>> > Thanks for reply. >>>> > >>>> > The MIB files on both machines are the same. >>>> > >>>> > I have compiled hundreds of MIB files and copied to both machines. >>>> > >>>> > The MIB files from the std packages, have I moved to my home >>>> directory >>>> > before. >>>> > >>>> > The snmp config files are all the same. >>>> > >>>> > I have also checked env, export and net-snmp-config, I can't find >>>> any >>>> > differences. >>>> > >>>> > My snmp.conf looks like: >>>> > >>>> > mibdirs /usr/share/mibs >>>> > mibs ALL >>>> > showMibErrors yes >>>> > strictCommentTerm yes >>>> > mibAllowUnderline yes >>>> > mibReplaceWithLatest yes >>>> > extendedIndex yes >>>> >>>> What happens if you try it with mibReplaceWithLatest off? Found >>>> this: >>>> " > target="_blank">https://docs.oracle.com/cd/E18752_01/html/816-5166/snmpcmd-1m.html >>>> >>>> -PR >>>> Replace MIB objects using the last read MIB file. The parser >>>> replaces MIB objects in its hierarchy whenever it sees a >>>> subidentifier >>>> and name match. >>>> Caution - Setting this option might result in an incorrect >>>> hierarchy. Can be also set with the configuration token >>>> mibReplaceWithLatest. >>>> >>>> Regards, >>>> Lee >>>> >>>> >>>> >>>> >>>> >>>> > Thanks! >>>> > >>>> > Venerable >>>> > >>>> > >>>> > >>>> >> -----Ursprüngliche Nachricht----- >>>> >> Von: Hans Jørgen Jakobsen >>>> >> Gesendet: Di. 12.07.2016 18:50 >>>> >> An: net-snmp-users@lists.sourceforge.net >>>> >> Betreff: Re: snmptranslate doesn't show full MIB-Tree on equal >>>> Systems >>>> >> >>>> >> >>>> >> >>>> >> On Mon, 11 Jul 2016, venerabl...@googlemail.com wrote: >>>> >> >>>> >>> For example: >>>> >>> If I use snmptranslate -m ALL -Pu -Of >>>> .1.3.6.1.4.1.4491.2.1.6.2.1.1 >>>> >>> on Server1 then it translated me >>>> >> >>> .iso.org.dod.internet.private.enterprises.cableLabs.clabProject.clabProjDocsis.docsIfMCmtsMib./ >>>> > >>>> > >>>> docsIfMCmtsConformance.docsIfMCmtsCompliances.docsIfMCmtsCoreDeviceCompliance >>>> >>> >>>> >>> If I use the same command on Server2 then it translated me only >>>> >>> >>>> >>> >>> .iso.org.dod.internet.private.enterprises.cableLabs.clabProject.clabProjDocsis.docsIfMCmtsMib.2.1.1 >>>> >>> >>>> >>> >>>> >>> How can that be? >>>> >> >>>> >> Have you looked at what MIB files are present on both machines. >>>> >> How has the Docsis MIB files entered the machines. They are at >>>> least on >>>> >> my machines not part of std packages. >>>> >> Differences in enviroment variables or (snmp) config files? >>>> >> >>>> >> /hjj> >>>> >> >>>> >> -----Ursprüngliche Nachricht Ende----- >>>> >>>> >>>>> >>> >>> -----Ursprüngliche Nachricht Ende----- >> >>> > > -----Ursprüngliche Nachricht Ende----- ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports.http://sdm.link/zohodev2dev _______________________________________________ Net-snmp-users mailing list Net-snmp-users@lists.sourceforge.net Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users
