Uri,
from your description of the problem I don't have a clear idea (so
forgive me if I tell you things you already know), but I can give you some
clues:
1. To get sure that a node will be snmp managed you must try snmpwalk
<nodename> without the -c option. If that works then Netview will succeed
when making a snmp poll to the node. If it doesn't work then you must check
that Netview is polling the node with the right community name (try
getcommunity <nodename> and check if that's the right community name).
2. Once you are sure that Netview knows the good community name for that
node, check with ovtopodump if your node is discovered with all IP
addresses. If you can see only one, it may mean that the object was
discovered before snmp was active (or Netview didn't know that). If that's
the case I would suggest deleting and rediscovering the node.
Regards and good luck.
Juan.
Juan Echevarría López
AT&T Global Network Services
Internet: jechevarria@att.com
"mtfuri" <mtfuri@fibimail.co.il> on 29/03/2001 07:18:16
Please respond to IBM NetView Discussion <nv-l@tkg.com>
To: nv-l@tkg.com
cc:
Subject: [NV-L] snmp objects discovery problem.
I am using netview 6.01 on aix 4.3.3 and try to snmp- manage some Cisco
routers.
doing snmp -c <community> from the Aix shell produce the right results, but
doing a demand poll from netview
bring only the "ping" information and I am getting an snmp time out
message. The object is not discovered as router to the database. those
router has 4 legs and I have a static route to two of the relevant nets.
I am able to recognize the same type of Cisco routers on another, more
simple net so I don't think that a mib is missing on my machine.
can any one help ??
TIA
Uri Cohen
|