To repeat:
1) Make sure netmon is configured to
use whatever seedfile you are editing.
2) You can edit your seedfile manually
if you like, followed by 'netmon -y'
3) Leave out the @limit_discovery statement
4 Add some SNMP-capable routers to the
seedfile. Put all of your core routers or switches in there by loopback
address.
5) To do your exclusions, add !10.16-19.12.10-40
Do these things, and then check your
results with
ovtopdump -l
to see how many things you are discovering.
Cordially,
Leslie A. Clark
IT Services Specialist, Network Mgmt
Information Technology Services Americas
IBM Global Services
(248) 552-4968 Voicemail, Fax, Pager
"sudha g" <sudha.unix@gmail.com>
Sent by: nv-l-bounces@lists.ca.ibm.com
05/31/2007 08:52 AM
|
Please respond to
Tivoli NetView Discussions <nv-l@lists.ca.ibm.com> |
|
|
To
| "Tivoli NetView Discussions"
<nv-l@lists.ca.ibm.com>
|
|
cc
|
|
|
Subject
| Re: [NV-L] Netview Event Browser Hangs |
|
The seed file is not working for me.
I cleared the database and used the default netmon.seed
file to rediscover, but most of the nodes are missing.
Now the requirement is, i need to rediscover all my nodes
and have to block devices from the following IP's
10.16.12.10-40
10.17.12.10-40
10.18.12.10-40
10.19.12.10-40
Please help me in doing this.
On 5/30/07, Leslie Clark <lclark@us.ibm.com>
wrote:
You are close, but you need to turn on discovery in the nmpolling dialog.
And you need to put the seedfile into effect BEFORE you restart automatic
map generation.
The ranges do not force discovery, they control discovery. In order to
consider a node for discovery, Netview has to hear about it first. This
can happen in one of several ways.
1) Seeds. Put a couple of SNMP-enabled routers, like your core devices,
in the seedfile. Netmon will work through their ARP cache and routing tables
and find out about devices, and work outwards from there, checking everything
it finds out about against those ranges. If any address on a device matches
any range, it is allowed. Discovery must be enabled.
2) Or you can ping a few things from the Netview commandline. Netmon hears
about them when the responses come back, and considers them for addition.
Then it checks their ARP cache, etc, and works out from there. Discovery
must be enabled.
3) Put all devices in the seedfile (one address per device is preferable).
You can leave Discovery disabled in this case, but it is nuisance to administer,
and can get quite large.
How you go about this depends on your mission. But a seedfile of
just ranges, on a new discovery, could take a long time. I like to see
a discovery just zoom along until it is done.
Cordially,
Leslie A. Clark
IT Services Specialist, Network Mgmt
Information Technology Services Americas
IBM Global Services
(248) 552-4968 Voicemail, Fax, Pager
(1)No i restarted all of them using ovstop & ovstart
commands.
(2) Event History--Query--->Display events is not working since it hangs
when i clicked Event History itself (showing busy icon)
(3) event -h hostname command send nodeup events, but didn't received in
trapd.log file. It was fine when i checked the same command in another
test machine and there i cans ee the event in trapd.log file.
(4) Finally, we are in the process of stopping traps from Openwave Messaging
Server and need to check the other events like Node up and Node down. Also
i want netview to stop discovering the network range, but it's not working
through the seed file i created. Please check the steps i followed and
the seed file and advice me.
Steps:
1.run this command "smit nv6000"
2. Maintain-->Clear Object/topology/map databases, remove customizations
3. Control-->Restart Automatic Map Generation
4. run this command "nmpolling"
5. uncheck "Discover new nodes" & "Discover
new services"
6. run this command "launch_seededit"
Limit by IP Address
In Include, i added the following IP range (can be
seen below from the output of seedfile)
In Exclude list i added the IP range which is not
to be discovered.
@limit_discovery # Limit discovery to the contents of this seed
file.
10.16.51.*
10.16.14.*
10.16.15.*
10.16.45.*
10.16.16.*
10.16.52.*
10.16.53.*
10.16.54.*
10.17.14.*
10.17.16.*
10.17.51.*
10.17.17.*
10.18.17.*
10.18.14.*
10.18.16.*
10.18.51.*
10.19.17.*
10.19.16.*
10.19.51.*
10.19.52.*
10.19.53.*
10.19.54.*
192.168.*.*
172.*.*.*
!10.16.12.10-40
!10.17.12.10-40
!10.18.12.10-40
!10.19.12.10-40
Is this right?
Will netview identify all the devices in 10.16.*.* ,10.17.*.*,10.18.*.*,10.19.*.*
and
If discovered i need this to arrange in 4 locations.
Like if i click location A, it should list all devices discovered under
10.16.*.* and if location B. it should show 10.17.*.*
Kindly advice me on the same.
On 5/29/07, James Shanks <jshanks@us.ibm.com>
wrote:
I would undo your last change, turning off the traps from Openwave, and
see whether things return to normal.
As I said in my last post, it sounds to me like you have flooded the box.
As for the rest, I am very PUZZLED by this description:
>>I can see lot of events in my trapd.log file,
but nvevents show blank. If i do ovstop and ovstart
i can see the events through nvevents, but itz
not getting refreshed when i see through display events of Event History.
(1) nvevents starts working if you restart which daemons? All of them?
Just trapd? That would be proof to me that the daemons get overloaded.
By restarting trapd, you cause him to throw away all the events in his
queue and start over.
(2) Events History is a completely different animal from Events. It does
not refresh automatically. You must pull down the Query button periodically
and manually cause it to refresh. It is not meant to be dynamic but rather
a short-term analysis tool for the most recent events.
If you want to test nvevents, I would try using the event command from
the command line,
event -h test1
This will generate three events,
(1) netmon-related application connected to trapd (APUP_EV)
(2) a dummy Node Up event for a nonexistent host called "test1"
(3) netmon-related application has disconnected from trapd (APDN_EV)
You will want to issue the command twice because trapd buffers his output
and does not write the last event to his log until the next one arrives.
That can be changed in 7.1.5 using an environment variable but not in 7.1.4.
The basic NetView events you are looking for are generated by netmon. If
you want to see what he is doing you should turn on the netmon trace (netmon
-M -1) and examine the netmon.trace in /usr/OV/log. Or issue one of the
dozen or more netmon action commands (netmon -a <number>) which get
written to the same place. See the man pages for suggestions.
The Layer 2 events are generated by ITSA and I know nothing about troubleshooting
that. I do know that it keeps topology in memory only and must rediscover
your Layer 2 network afresh every time you restart it, and in some networks
that can take a very long time. Poor performance by ITSA in large networks
is one reason why IBM went to Precision IP for the future direction.
HTH,
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Network Availability Management
Network Management - Development
Tivoli Software, IBM Corp
 "sudha
g" <
sudha.unix@gmail.com>
James,
I have installed Netview Server 7.1.4 and no patches installed in Aix 5.2.
I'm using the forwardall.rs as the default ruleset.
I can see lot of events in my trapd.log file, but nvevents show blank.
If i do ovstop and ovstart i can see the events through nvevents,
but itz not getting refreshed when i see through display events of Event
History.
bash-2.05b# cat /usr/OV/log/trapd.log | grep node
bash-2.05b# cat /usr/OV/log/trapd.log | grep network
bash-2.05b# cat /usr/OV/log/trapd.log | grep interface
I didn't get no results for the above commands from trapd.log, but lot
of events i can see from my Openwave Messaging Application
1180346657 3 Mon May 28 15:34:17 2007 10.18.12.12
A popEvent 6 3 2 args: [1] private.enterprises.2415.1.1.1.2.2.1
(Integer): 1
1180346657 3 Mon May 28 15:34:17 2007 10.18.12.13
A popEvent 6 3 2 args: [1] private.enterprises.2415.1.1.1.2.2.1
(Integer): 1
1180346657 3 Mon May 28 15:34:17 2007 10.17.12.28
A popEvent 6 3 2 args: [1] private.enterprises.2415.1.1.1.2.2.1 (Integer):
1
1180346657 3 Mon May 28 15:34:17 2007 10.17.12.29
A popEvent 6 3 2 args: [1] private.enterprises.2415.1.1.1.2.2.1 (Integer):
1
I don't have any problem in my name resolution. The events i saw above
can be seen in the TEC console only after restarting ovstop and ovstart
daemons. There will be no events in TEC fo longtime unless i do a next
restart.
This problem happens only after configuring our other Openwave Messaging
Servers to send traps to Netview and we also applied a patch for Tivoli
Business Systems Manager Event Enablement.3.1.0-FP0002
Thanks & Regards,
Sudhakar Gandhi
On 5/26/07, James Shanks <
jshanks@us.ibm.com
> wrote:
OK, so this is UNIX. What platform, and what version of NetView? If not
7.1.5, what fixes are installed?
Are you using the forwardall.rs ruleset as the default? You can try turning
on tracing in nvcorrd to see what is happening to your events, with the
command
nvcdebug -d all
Then you can tell in the nvcorrd logs whether events are being processed.
But really these kinds of things are very difficult to diagnose remotely.
Is this a new system? When did you start having the problem? Could your
name resolution be a problem? All events get resolved three times, once
by trapd for the trapd.log, once by nvcorrd if the ruleset requires it
(though forwardall usually does not), and again by nvserverd for display
by nvevents.
HTH,
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Network Availability Management
Network Management - Development
Tivoli Software, IBM Corp
 "sudha
g" <sudha.unix@gmail.com
>
Hi,
My Netview Event Browser suddenly hangs and i have restarted nvcorrd,nvserverd
and actionsvr .
I did ovstop and ovstart too, but no luck. I cannot see any events
in the Event browser and it is blank, when i click the Event History icon,
it got hanged!
Please help me out in fixing this issue.
Thanks,
Sudhakar Gandhi _______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal
IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal
IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to
internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to
internal IBM'ers only)
|
