I tried the queries you provided and they worked
great! I’ll probably create a script to update the rest of the routers we
are monitoring. Thanks again for the help.
From:
opennms-discuss-bounces <at> lists.sourceforge.net
[mailto:opennms-discuss-bounces <at> lists.sourceforge.net] On Behalf Of John Blake
Sent: Wednesday, November 01, 2006
8:53 AM
To: General
OpenNMS Discussion
Subject: Re: [opennms-discuss]
SNMP partially discovered/No SNMP Performance
You’ll have to get all the
nodeid’s thafor th eprimary interfaces
Select nodeid, ipaddr from
ipinterface where issnmpprimary = ‘P’.
Once you figure out which ones you need to
change;
To change an ip from Primary;
update ipinterface set issnmpprimary
= 'N' where nodeid = 435 and ipaddr = 10.10.10.10’;
to set it to pirmary;
update ipinterface set issnmpprimary
= 'P' where nodeid = 435 and ipaddr = '10.10.10.10';
John
From:
opennms-discuss-bounces <at> lists.sourceforge.net [mailto:opennms-discuss-bounces <at> lists.sourceforge.net]
On Behalf Of Michael Koprowski
Sent: Tuesday, October 31, 2006
3:57 PM
To: General
OpenNMS Discussion
Subject: Re: [opennms-discuss]
SNMP partially discovered/No SNMP Performance
I didn’t think of making the
modification directly in the database and would agree that trying to create
800+ packages would be time consuming.
Do you have example queries to list and
change the Primary interface?
Thanks,
Mike
From:
opennms-discuss-bounces <at> lists.sourceforge.net
[mailto:opennms-discuss-bounces <at> lists.sourceforge.net] On Behalf Of John Blake
Sent: Tuesday, October 31, 2006
3:26 PM
To: General
OpenNMS Discussion
Subject: Re: [opennms-discuss]
SNMP partially discovered/No SNMP Performance
The package would have to
“include” the primary ip and “exclude” all the others.
If it didn’t
“exclude” the others, ONMS will still test them and go thru its
normal procedure for marking the Primary IP.
How often are you going to rescan
the devices?
You might just do the normal discovery
process and go into the dbase to set the primary interface for each node.
John
From:
opennms-discuss-bounces <at> lists.sourceforge.net
[mailto:opennms-discuss-bounces <at> lists.sourceforge.net] On Behalf Of Michael Koprowski
Sent: Tuesday, October 31, 2006
3:19 PM
To: General
OpenNMS Discussion
Subject: Re: [opennms-discuss]
SNMP partially discovered/No SNMP Performance
Thanks for the quick response John. Since
OpenNMS is multihomed on the client’s network and ours, I don’t
have the option of creating an acl to restrict the “other IPs”.
Going on my other thought, if I create a
collectd package specific for each device, would this effectively mark the
specific defined address in the package the primary interface?
Mike
From:
opennms-discuss-bounces <at> lists.sourceforge.net
[mailto:opennms-discuss-bounces <at> lists.sourceforge.net] On Behalf Of John Blake
Sent: Tuesday, October 31, 2006
3:12 PM
To: General
OpenNMS Discussion
Subject: Re: [opennms-discuss]
SNMP partially discovered/No SNMP Performance
- No.
- No.
The only way to do it is to not allow ONMS direct access to those other
IP’s.
John
From:
opennms-discuss-bounces <at> lists.sourceforge.net
[mailto:opennms-discuss-bounces <at> lists.sourceforge.net] On Behalf Of Michael Koprowski
Sent: Tuesday, October 31, 2006
3:01 PM
To: General
OpenNMS Discussion
Subject: [opennms-discuss] SNMP
partially discovered/No SNMP Performance
I seem to keep running into the same situation when discovering our
client’s Cisco equipment. I created an entry in the snmp-config.xml file
containing the correct community string, and I validated I can walk SNMP for
the device. I limited the capsd scan in the capsd-configuration.xml to only
scan ICMP, SNMP and Router. During the discovery of the nodes, I see in the
capsd.log OpenNMS correctly identifies SNMP is accessible on the device and
obtains mib information. The information in the capsd.log seems to run as
expected. Below is the output for the discovered host ip 10.255.252.211:
[root <at> NOCLI-ONMS007 etc]# cat ../logs/capsd.log |grep 10.255.252.211
2006-10-31 13:37:56,588 DEBUG [Capsd:BroadcastEventProcessor]
BroadcastEventProcessor: onMessage: Adding interface to suspectInterface Q:
10.255.252.211
2006-10-31 13:59:33,997 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor: running collection for
10.255.252.211
2006-10-31 13:59:33,997 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
IfCollector.run: run method invoked to collect information for address 10.255.252.211
2006-10-31 13:59:33,998 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 testing plugin ICMP
2006-10-31 13:59:34,107 DEBUG [ICMPReceiver39966] ReplyReceiver:
process: received matching echo reply from host 10.255.252.211
2006-10-31 13:59:34,107 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 protocol ICMP supported? true
2006-10-31 13:59:34,107 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 plugin ICMP completed!
2006-10-31 13:59:34,107 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 testing plugin SNMP
2006-10-31 13:59:34,524 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 protocol SNMP supported? true
2006-10-31 13:59:34,524 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 plugin SNMP completed!
2006-10-31 13:59:34,524 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 testing plugin Router
2006-10-31 13:59:34,940 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 protocol Router supported? false
2006-10-31 13:59:34,940 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
10.255.252.211 plugin Router completed!
2006-10-31 13:59:34,941 DEBUG [Capsd Suspect Pool-fiber0]
IfSnmpCollector: run: collecting for: /10.255.252.211 with agentConfig:
AgentConfig[Address: /10.255.252.211, Port: 161, Community: nocreadstring,
Timeout: 1000, Retries: 3, MaxVarsPerPdu: 50, Max request size: 65535, Version:
2]
2006-10-31 13:59:34,941 DEBUG [Capsd Suspect Pool-fiber0]
JoeSnmpWalker: Walking system/ifTable/ifXTable/ipAddrTable for /10.255.252.211
using version SNMPv2
2006-10-31 13:59:35,384 DEBUG [SnmpPortal--1] JoeSnmpWalker: Received a
tracker pdu from /10.255.252.211 of size 56 errorStatus = 0, errorIndex = 0
2006-10-31 13:59:35,846 DEBUG [SnmpPortal--1] JoeSnmpWalker: Received a
tracker pdu from /10.255.252.211 of size 68 errorStatus = 0, errorIndex = 0
2006-10-31 13:59:35,850 DEBUG [SnmpPortal--1] IpAddrTableEntry: Storing
Result: alias: ipAdEntAddr [.1.3.6.1.2.1.4.20.1.1].[10.255.252.211] =
10.255.252.211
2006-10-31 13:59:35,850 DEBUG [SnmpPortal--1] IpAddrTableEntry: Storing
Result: alias: ipAdEntIfIndex [.1.3.6.1.2.1.4.20.1.2].[10.255.252.211] = 4
2006-10-31 13:59:35,850 DEBUG [SnmpPortal--1] IpAddrTableEntry: Storing
Result: alias: ipAdEntNetMask [.1.3.6.1.2.1.4.20.1.3].[10.255.252.211] =
255.255.255.255
2006-10-31 13:59:35,850 DEBUG [SnmpPortal--1] IpAddrTableEntry: Storing
Result: alias: ipAdEntBcastAddr [.1.3.6.1.2.1.4.20.1.4].[10.255.252.211] = 1
2006-10-31 13:59:36,304 DEBUG [SnmpPortal--1] JoeSnmpWalker: Received a
tracker pdu from /10.255.252.211 of size 66 errorStatus = 0, errorIndex = 0
2006-10-31 13:59:36,716 DEBUG [SnmpPortal--1] JoeSnmpWalker: Received a
tracker pdu from /10.255.252.211 of size 71 errorStatus = 0, errorIndex = 0
2006-10-31 14:00:02,438 DEBUG [Capsd Suspect Pool-fiber0] IfCollector:
IfCollector.run: run method exiting after collecting information from address
10.255.252.211
2006-10-31 14:00:52,429 DEBUG [Capsd Suspect Pool-fiber0]
CapsdConfigFactory: isInterfaceInDB: attempting to lookup interface
10.255.252.211 in the database.
2006-10-31 14:00:52,430 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: getExistingNodeEntry: checking for current target:
/10.255.252.211
2006-10-31 14:00:52,430 DEBUG [Capsd Suspect Pool-fiber0] SuspectEventProcessor:
getExistingNodeEntry: issuing SQL command: SELECT nodeId FROM ipinterface WHERE
ipaddr='10.0.3.1' OR ipaddr='10.255.252.211' OR ipaddr='10.255.12.170' OR
ipaddr='10.254.12.170'
2006-10-31 14:00:52,548 DEBUG [Capsd Suspect Pool-fiber0] SuspectEventProcessor:
SuspectEventProcessor: 10.255.252.211 has sysName: r0003r
2006-10-31 14:00:52,549 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor: 10.255.252.211 has
sysDescription: Cisco Internetwork Operating System Software
2006-10-31 14:00:52,549 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor: 10.255.252.211 has sysLocation:
2006-10-31 14:00:52,549 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor: 10.255.252.211 has sysContact:
2006-10-31 14:00:52,792 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: addInterfaces: 10.255.252.211/10.255.252.211 is managed
2006-10-31 14:00:52,806 DEBUG [Capsd Suspect Pool-fiber0]
PollerConfigFactory: interfaceInPackage: Interface 10.255.252.211 passed filter
for package example1?: true
2006-10-31 14:00:52,806 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: addInterfaces: 10.255.252.211/10.255.252.211 is to be
polled = true
2006-10-31 14:00:52,806 DEBUG [Capsd Suspect Pool-fiber0]
IfSnmpCollector: getIfIndex: retrieving ifIndex for 10.255.252.211
2006-10-31 14:00:52,806 DEBUG [Capsd Suspect Pool-fiber0] IpAddrTable:
getIfIndex: got a match for address 10.255.252.211 index: 4
2006-10-31 14:00:52,806 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor: setting ifindex for
10.255.252.211/10.255.252.211 to 4
2006-10-31 14:00:52,815 DEBUG [Capsd Suspect Pool-fiber0]
PollerConfigFactory: interfaceInPackage: Interface 10.255.252.211 passed filter
for package example1?: true
2006-10-31 14:01:03,204 DEBUG [Capsd Suspect Pool-fiber0]
SuspectEventProcessor: SuspectEventProcessor for 10.255.252.211 completed.
[root <at> NOCLI-ONMS007 etc]#
From the Web UI I click on the node but don’t see the option to
view SNMP Performance. I click on Admin and click on “Configure SNMP Data
Collection per Interface” but don’t see this IP as an option to
collect data. In fact I only see 2. The first is a serial interface without an
IP address and the second is a Fast Ethernet interface with an IP address, but
neither is set to Primary.
I’ve been running into this more and more often. I would think if
I’m discovering a device by its IP address, shouldn’t this signal
to OpenNMS mark this interface as the Primary?
I’ve read the “Data Collection Configuration How-To”
but don’t feel I’m encountering the issue the article speaks about
where there are several interfaces that meet all 3 criteria to be eligible for
the Primary.
The 2 questions I have are:
1) Is
there an easy way to limit interfaces eligible to be Primary for 800 devices
other then creating a collectd-configuration.xml package for each device?
2) Can
OpenNMS be configured to mark the discovered IP as the Primary?
System Info:
Version: 1.3.1-1 – Downloaded from the SourceForge site
Server Time: Tue Oct 31 14:52:47 EST 2006
Client Time: document.write(
new Date().toString()) Tue Oct 31 2006 14:52:52
GMT-0500 (Eastern Standard Time)
Java Version: 1.4.2_04 Sun Microsystems Inc.
Java Virtual Machine: 1.4.2_04-b05 Sun Microsystems Inc.
Operating System: Linux 2.6.9-5.ELsmp (i386)
Servlet Container: Apache Tomcat/4.1 (Servlet Spec 2.3)
User Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1)
Gecko/20061010 Firefox/2.0
Postgres 8.1.4
Regards,
Mike
RSS Feed