To: | nv-l@lists.us.ibm.com |
---|---|
Subject: | RE: [nv-l] Teach me to fish |
From: | Leslie Clark <lclark@us.ibm.com> |
Date: | Mon, 23 Aug 2004 10:02:21 -0400 |
Delivery-date: | Mon, 23 Aug 2004 15:16:48 +0100 |
Envelope-to: | nv-l-archive@lists.skills-1st.co.uk |
In-reply-to: | <155FD4A036C9A643B7745049D08EC2F5440E94@pntmail1.foremost.com> |
Reply-to: | nv-l@lists.us.ibm.com |
Sender: | owner-nv-l@lists.us.ibm.com |
I went through this with one customer after an AIX upgrade in which they decided to change the default community from public to something else. From the snmpd.log I could tell that the mib variable being requested was the dpid2 port, which led me to this understanding: The hostmibd daemon is an snmp subagent (a smux agent) on AIX which supplies the answers to SNMP queries of variables defined in the MIB II Hosts mib section - system information about disk, memory, cpu, etc. This daemon requires dpid2 to be up first, to handle communication between hostmibd and snmpd. If dpid2 is not up, hostmibd keeps trying forever. The hostmibd daemon uses an SNMP query to find out what port to talk to dpid2 on. It uses the community string of 'public' to make this query unless a different one is specified in its startup in /etc/rc.tcpip. Cordially, Leslie A. Clark IBM Global Services - Systems Mgmt & Networking (248) 552-4968 Voicemail, Fax, Pager
I'm quite certain that the failures are NOT NetView. I can see TCPDUMPs (snmp traces) coming from NetView containing the correct community string and they process correctly. I do not see any other SNMP requests coming in.
Thanks for the thought. -----Original Message-----
You didn't specify when you did a TCPDUMP of where you were tracing. But the cause of the authentication failures may not be NetView. Some other device on the segment may be probing. Next time you can, trace all incoming port 161 requests and not just from the NetView server. Maybe someone is running another app on your network.
From: owner-nv-l@lists.us.ibm.com [mailto:owner-nv-l@lists.us.ibm.com]
On Behalf Of JAMES Robin
Try doing an snmpwalk command on the box and look for the SNMP settings. Somewhere, if you use the community with write access, you should find the SNMP settings for the read and/or write communities. Netview or some other manager must be polling the device with the wrong community name. Your device should allow you to do a SNMP set operation to the OID which defines the read and/or write community name. What community is Netview using for SNMP polling? public is the usual default. Perhaps your device has been set-up to only allow access via different community. Good luck -----Original Message-----
I've got a rather strange problem and hopefully someone
can give me a clue where to start digging. Any idea how I can start digging into this problem?
=== TRAP.RPT =========================================================
|
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | [nv-l] Teach me to fish - Step 2, Stamper, Steve |
---|---|
Next by Date: | Re: [nv-l] submap oddity?, Paul |
Previous by Thread: | RE: [nv-l] Teach me to fish, Stamper, Steve |
Next by Thread: | [nv-l] submap oddity?, Mike |
Indexes: | [Date] [Thread] [Top] [All Lists] |
Archive operated by Skills 1st Ltd
See also: The NetView Web