Glen,
mib2trap works on files, not on the MIB database. So all the MIB
definitions needed to resolve the traps must be in one file. If they are
spread over multiple files, then you have to gather them all into one big
one.
But before you do that, or any more of this, let me give you a tip.
If you plan running a whole bunch of vendor traps through mib2trap, then my
advice is to build a little mini MIB database for just them.
You would use xnmloadmib2 for that, since it has the same parser as
mib2trap. Two ways to do this.
(1) rename /usr/OV/conf/ snmpv2mib and snmpv2mib.bin to something else for
safe keeping.
(2) start loading all the MIBs you want to run mib2trap on and keep adding
MIBs until you don't get any more warnings or errors. Use the IMPORTS
sections to tell you what other MIBs you need for resolution when you get
error. Usually these are vendor MIBs and not RFC's.
(3) When you finish step (2), run mib2trap on /usr/OV/conf/snmpv2mib. It's
a big text file of all the MIBs you loaded to get one consistent whole.
(4) When mib2trap is finished, you can rename the new /usr/OV/conf/
snmpv2mib and snmpv2mib.bin files to something else (for safe keeping case
you want to do this again) and restore the old ones.
It's also possible to skip the rename game altogether and run xnmloadmib2
from the command line with the -mibDB option to specify a new MIB database
file of your choosing and run mib2trap on that instead. For each vendor
you want to add a lot of traps for, you van build a little mini database of
just his stuff for mib2trap to work with.
HTH
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
James
Shanks/Raleigh/IB
M@IBMUS To
Sent by: nv-l@lists.us.ibm.com
owner-nv-l@lists. cc
us.ibm.com
Subject
RE: [nv-l] error w/ mib2trap
04/06/2006 08:38
AM
Please respond to
nv-l@lists.us.ibm
.com
-T is valid though it may not be documented at you level of code. It tells
addtrap not to send the FMTCHG event to trapd. This flag is added by
mib2trap to prevent trapd from getting into a loop re-reading the
trapd.conf file when many new traps are being added at once.
You problem is the -i, not -T.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
"Glen Warn"
<Glen.Warn@pemcoc
orp.com> To
Sent by: <nv-l@lists.us.ibm.com>
owner-nv-l@lists. cc
us.ibm.com
Subject
RE: [nv-l] error w/ mib2trap
04/05/2006 08:32
PM
Please respond to
nv-l@lists.us.ibm
.com
I think I fixed this. After going thru each parameter I found there is no
T option for addtrap. I also saw that the –i parameter wasn’t being
issued. I did a search and replace on the –T with a –i
1.3.6.1.4.1.9.9.171.2 and reran the script. It seems to have worked fine.
Does anyone have any input on the –T (specifically why would mib2trap put
it there?) Not sure which of the 2 actually was the fix, but it did run
w/o error now.
From: owner-nv-l@lists.us.ibm.com [mailto:owner-nv-l@lists.us.ibm.com] On
Behalf Of Glen Warn
Sent: Wednesday, April 05, 2006 4:57 PM
To: nv-l@lists.us.ibm.com
Subject: [nv-l] error w/ mib2trap
Running NV 7.1.4 FP3 on RH AS.21
Have had this problem before, cannot recall work around.
Trying to get traps defined for CISCO-IPSEC-FLOW-MONITOR-MIB.my
Run mib2trap against it and works well (except for this warning at the
beginning)
Warning: The mib description doesn't seem to be consistent.
Some nodes couldn't be linked under the "iso" tree.
These nodes are left:
And this at the bottom
Enterprise object id not found for trap cikeTunnelStart, (6,1).Enterprise
object id not found for trap cikeTunnelStop, (6,2).Enterprise object id not
found
for trap cikeSysFailure, (6,3).Enterprise object id not found for trap
cikeCertCrlFailure, (6,4).Enterprise object id not found for trap
cikeProtocolFailu
re, (6,5).Enterprise object id not found for trap cikeNoSa,
(6,6).Enterprise object id not found for trap cipSecTunnelStart,
(6,7).Enterprise object id not
found for trap cipSecTunnelStop, (6,8).Enterprise object id not found for
trap cipSecSysFailure, (6,9).Enterprise object id not found for trap
cipSecSetUp
Failure, (6,10).Enterprise object id not found for trap cipSecEarlyTunTerm,
(6,11).Enterprise object id not found for trap cipSecProtocolFailure,
(6,12).En
Now that I have a script to run, I do but it returns this over and over
again.
Usage: /usr/OV/bin/addtrap -n enterprise-name
-l trap-label
-i ent-object-id
-g gen-trap
-s spec-trap
-o source-ID
-S severity-flag
-t status-type
-c category
-F format-specification
[-C command]
[-d display] ]
[-D description] ]
[-N node list] ]
[-e TEC Class Name] ]
[-E TEC Slot Name] ]
[-V TEC Slot Value] ]
[-T do not send FMTCHG event to trapd ]
If arguments contain space, use the double quotation mark.
Here is a sample from the beginning of the script.
/usr/OV/bin/addtrap -l cikeTunnelStart -g 6 -s 1 \
-n cipSecMIBNotifications \
-o A -c "LOGONLY" -t 0 \
-S 1 \
-T \
-D "This notification is generated when an IPsec Phase-1
IKE Tunnel becomes active." \
-e cikeTunnelStart \
-F '$E $G $S $# args: $*'
I’ve also tried running the first step manually and it fails too. Am
tinkering to see where but here is the command I’ve used.
/usr/OV/bin/addtrap -l cikeTunnelStart -g 6 -s 1 -n cipSecMIBNotifications
-o A -c "LOGONLY" -t 0 -S 1 -T -D "This notification is generated when an
IPsec Phase-1 IKE Tunnel becomes active." -e cikeTunnelStart -F '$E $G $S
$# args: $*'
|