Georg,
Perhaps you can get 50% of the way. The trap in question (according to
a MibDepot search) has only one varbind - SlbRealServerState. This can
have 14 possible values, but you seem most interested in a couple of
groups of them: failed, probeFailed, inbandProbeFailed (3,9,13), and
testing, probeTesting (5,10). You're heading down a path of using
rulesets and correlation have the ability to do comparisions on the
varbind and act accordingly.
Jon Austin
>>> Georg.Gangl@brz.gv.at 03/16/04 07:17AM >>>
Hi Group members,
In my Events Display Window I found an unknown trap with Enterprise ID
1.3.6.1.4.1.9.9.161.2, Specific 2 and Generic 6. Using the SNMP
Object
Navigator I found out that this Enterprise ID specifies
"CiscoSlbMIBnotifications". So I loaded the "CISCO-SLB-MIB.my" MIB via
Web Console Loader and the Traps via "mib2trap" and we are well
informed
now about state changes on virtual and real servers.
The trouble is that the traps coming in with one Enterprise ID and one
Specific Trap number are indeed 4 different traps (TCP health probe
failed for server X, TCP health probe re-activated for server X, HTTP
health probe failed for server X and HTTP health probe re-activated
for
server X) and I do not know how to separate the "failed" from the
"re-activated" and the "TCP" from the "HTTP" traps or in other words
how
to produce 4 different traps - one for each different type of state
change.
Does anybody know how this problem could be tackled ? Any suggestions
are appreciated very much.
Cordially,
Georg Gangl
_____________________________
Ing. Georg Gangl
Bundesrechenzentrum Ges.m.b.
Hintere Zollamtsstrasse 4
A-1033 Wien
Gruppe I-PS-NW Netzwerke
Tel.: 00 43 (1) 711 23 - 3891 DW
Fax.: 00 43 (1) 711 23 - 3897 DW
mailto:Georg.Gangl@brz.gv.at
--------------------------------------------------
|