There are other SNMP management applications
for you servers. Usually the hardware comes with one that is SNMP-capable.
For instance, Dell servers have Dell OpenManage, and IBM servers have IBM
Director. Those are not simple, either, but they know a lot about the machines
they run on.
To use Netview as the trap focalpoint
for those management applications, you will also need to load the mibs
for those applications on the Netview server (they usually come with the
applications) and format the traps that are defined in those mibs. There
is a mib2trap command to help with that on Netview.
The TEC event log adapter is for monitoring
the contents of log files and sending events to the Tivoli Enterprise Console,
when you have a Tivoli environment. It is not SNMP.
Cordially,
Leslie A. Clark
IBM Global Services - Systems Mgmt & Networking
(248) 552-4968 Voicemail, Fax, Pager
"Bernhard Zanker"
<bernhard.zanker@serpro.gov.br>
Sent by: owner-nv-l@lists.us.ibm.com
06/30/2005 03:53 PM
|
|
To
| nv-l@lists.us.ibm.com
|
|
cc
|
|
|
Subject
| Re: [nv-l] Service Down and
Service UP TRAPs ... |
|
Hi !
Its my first TRAP configuration with NetView ...
I don't understand why my hosts (win 2000 SP4) don't send TRAP information
by default ... I need to customize the TRAP information using the MS application
evntwin.exe ... I only see TRAPs comming in Port 162 when I use this MS
application (and I think that a hard way to manage Windows TRAPs ???)
When I do that I see so many work to do ! I need to customize every service
to send TRAPs to my NetView Server ... and when I do that, I have many
OIDs ... and I must create the Enterprises to work with this "new"
TRAPs ...
I think that exist a easy way to manage windows TRAPs in NetView .... but
I don't know how ?
I need to install another application/service in my Win2000 Hosts ? I have
read something about TEC - EventLog Adaptor : tecad_win.exe
.or. tecadwins.exe ...
Thanks James !!!
Regards,
Bernhard
|
|
James Shanks <jshanks@us.ibm.com>
Enviado Por: owner-nv-l@lists.us.ibm.com
30/06/05 12:42
Favor responder a nv-l
|
Para: nv-l@lists.us.ibm.com
cc:
Assunto: Re: [nv-l]
Service Down and Service UP TRAPs ... |
-
Sorry, but we still have a failure to communicate.
I don't think you have a clear idea of how this monitoring is going to
occur, or at least you haven't explained it.
What process is monitoring these different services? NetView's servmon
or
something else?
If servmon, then the name of the service being monitored will be in varbind
4 of the Service Up and Service Down traps, so you can change the display
of these traps, using Trap Settings, to say not just "$3"
as it does
now, but "$3 $4" so that the varbind 4 information will also
be displayed.
These other OIDs you mention will not be in the NetView trap.
The actual OIDs you describe belong to some MIB or other, which is not
queried by NetView by default. So unless you have some other mechanism
in
mind to get this information, (snmpcollect?) they aren't relevant.
If
you are querying an agent using some MIB with snmpcollect, the OID might
appear as part of the snmpcollect threshold trap, provide you configure
snmpcollect to send you threshold traps, but you will not get that OID
as
separate enterprise.
If none of this helps, I suggest you implement your monitoring process
for
a few of these services and see how ti works, before you worry about
configuring the traps. That can always be the last step.
Hope this helps,
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
"Bernhard Zanker"
<bernhard.zanker@
serpro.gov.br>
To
Sent by:
nv-l@lists.us.ibm.com
owner-nv-l@lists.
cc
us.ibm.com
Subject
Re: [nv-l]
Service Down and Service
06/30/2005 10:54
UP TRAPs ...
AM
Please respond to
nv-l
I have this problem :
Windows have many services, and each service have one OID !!! And NetView
Enterprise have just one TRAPs for ServiceDown (58916976) ...
For Example : TSM Server have one OID
(1.3.10.65.100.115.109.83.101.114.118.101.114), Telnet Service another
OID,
DNS Service another, etc ...
I need something like a Translator ... to translate many ServiceDown TRAPs
to NetView TRAP Service Down ... and then have just one Enterprise :
NetView.
Its that possible ? Or I need to create many Enterprises in TRAP Settings
?
I want to manage up to 100 windows services, so I need to create up to
100
Enterprises ???
Regards,
Bernhard
James Shanks
Para:
<jshanks@us.ibm.com> nv-l@lists.us.ibm.com
Enviado Por:
cc:
owner-nv-l@lists.us.ibm.com Assunto:
Re: [nv-l]
29/06/05 15:21
Service Down and Service UP TRAPs ...
Favor responder a nv-l
-
I'm not certain just what you are asking.
Service traps would be issued by servmon provided he is configured to
montior for that service. If he sent them, you would get them. You
can
send dummy versions of them with the event command to mimic this behavior,
if you want to check how your Perl script behaves. The command "event
-l"
( small letter L) will list all the traps supported by the event command
and then you can issue those that appear in the list either by name or
number. For example, "event -l" shows that the two traps
you are
interested in are
SUP_EV 0058916975
3 Service Up
SDWN_EV 0058916976 3
Service Down
so you could issue either "event -e SUP_EV" or "event
-E 58916975" to get
a dummy Service Up event to be issued. You can combine that with
other
parameters on the event command to send more data in the event, as for
example, "event -h test1 -a junk -e 58916975" which would
give you a
Service Up trap from a host named "test1" with the work "junk"
in varbind
4, which usually contains, among other things, the service name.
If you are complaining that servmon is not sending the trap, then I would
check his configuration, and his log. I believe that he runs on a
timer
basis and does not react to traps from other devices, but rather performs
his checks based on what in his config file. So for him to generate
his
traps, the Service would have to be up or down at the time that he checked
it, which means that there may be a considerable time lag. servmon
is
designed to notify you about services running on servers which are not
capable of monitoring them and sending traps about them when they go up
or
down. If your server can send his own trap, that will almost always
arrive
at your box before servmon will get a chance to go check the service.
Or are you asking about Service Impact traps sent to TEC? That's
a
completely different issue.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Tivoli Software / IBM Software Group
"Bernhard Zanker"
<bernhard.zanker@
serpro.gov.br>
To
Sent by:
nv-l@lists.us.ibm.com
owner-nv-l@lists.
cc
us.ibm.com
Subject
[nv-l] Service
Down and Service UP
06/29/2005 01:00
TRAPs ...
PM
Please respond to
nv-l
Hi again !!!
I'm trying to run Scripts in PERL whenever I receive TRAPs from my network
...
I receive only TRAPs NodeDown and NodeUP, and TRAPs 'AMS2DefaultTrap' from
Enterprise 'AMS2ALERT' (Anti-virus Server) ... but don't Service Down and
Service UP from Enterprise 'NetView' ...
I have connected to UDP port 162 to see if TRAP packets ServiceDown/UP
are
arriving from my network ... but I don't see it !!!
Its the problem in my Windows 2000 Host ???
Enviroment :
Server : Tivoli 7.1.3 Windows
Hosts : Windows 2000 SP4 with SNMP TRAP enabled to Tivoli Server IP
Regards,
Bernhard
|