Recommended SNMP traps

Hello All,
We are planning to implement SNMP monitoring in our network. Are there any recommended SNMP traps that should be enabled on the devices so we dont get clobbered with thousands of traps.
We have the following devices on our network-
Cisco 7206 with ATM interface, running BGP, EIGRP, PBR.
Cisco 6509, 6513 switches- Running EIGRP, Cisco FWSM
Cisco 1841 routers- have serial links, running BGP, Frame relay.
Cisco 2800 router- acting as DHCP server
Cisco ASA - 1) Perimiter firewall, 2) SSLVPN firewall.
We wish to monitor not only the external parameters like the circuits, routing protocols etc,, but also hardware and environmental parameters if possible.
If someone can recommend a list of traps that should be enabled for these devices, I wold much appreciate it.
Regards,
Imran.

Clearly you should test and refine this list based on your run time experience, but if you enable these trap keywords, that should provide you with a good set of signal to manage.
General:
snmp
entity
envmon
cpu
memory
EIGRP:
eigrp
BGP:
bgp
Frame-relay:
frame-relay
Some of your routing and DHCP management may only be possible with IP SLA and the Embedded Event Manager. In that case, you may want to enable the following two keywords, which will allow you to build IP SLA collectors and EEM policies that can generate SNMP traps:
ipsla
event-manager
For the ASA, enable the following trap keywords:
entity
ipsec
snmp

Similar Messages

3750 Cluster and SNMP traps

I have a 3750 cluster and I want to know what are the recommended snmp traps to be sent. We definitely want to know when one of the switches in the cluster fails.
I've read about snmp-server enable traps stackwise and snmp-server enable traps cluster. What do these traps actually do?

stackwise would be useful, here's a description:
http://tools.cisco.com/Support/SNMP/do/BrowseOID.do?objectInput=ciscoStackWiseMIB&translate=Translate&submitValue=SUBMIT
also have a look at:
http://www.cisco.com/en/US/docs/switches/lan/catalyst4500/12.1/13ew/command/reference/S1.html#wp1126420

Cisco Prime SNMP Traps Best Pratice

The Cisco Prime documentation recommends configuring switches to send SNMP traps. However it does not give any more details.
I was wondering what sorts of SNMP traps people in the community are using with Cisco Prime 2.1. I'm looking for some sort of best practice or for an idea of what traps would be the most useful to configure on the switches, to send to Prime.

Hi ,
Snmp traps need to be configured only on device end , there is no config need to be done on PI.
you can enable all the traps that you want. for e.g
snmp-server enable traps syslog
snmp-server enable traps ipsec start stop
snmp-server enable traps memory-threshold
snmp-server enable traps interface-threshold
snmp-server enable traps connection-limit-reached
snmp-server enable traps cpu threshold rising
etc......
and you can monitor then in PI (Administration > System Settings > Severity Configuration, Link down)
check the below link as well:
https://supportforums.cisco.com/discussion/11919481/prime-infrastructure-20-link-status-alarms
Thanks-
Afroz
***Ratings Encourages Contributors ***

Curious message during activation of snmp traps in cisco router 2800

I activated snmp traps in cisco router 2800 (I didn't specify a set of them, so all of them were activated). The thing is, when i configure "snmp-server enable traps", appears a message in the next line :
" %Cannot enable both sham-link state-change interface traps.
%New sham link interface trap not enabled "
Anyway, traps are activated and are completely functional.
I would like to know, why this message appears... and also what is the difference between informs and traps, because I can activate both of the in a router to be sent to the network admin pc.
Thanks in advance.

Hi Marcelo,
the snmp-server enable traps command enables just all types of traps that the IOS version supports.
The message apperes because of this two, which are mutually exclusive:
R1(config)#snmp-server enable traps ospf cisco-specific state-change shamlink interfaceR1(config)#snmp-server enable traps ospf cisco-specific state-change shamlink interface-old% Cannot enable both sham-link state-change interface traps.% Deprecated sham link interface trap not enabled.
It's recommended to only enable the traps you really need.
Informs were introduced with SNMPv2, and they have the same format and purpose as traps.
The main difference is that traps are send in a hit-or-miss fashion whereas informs expect an acknowledge and will be re-send if unacknowledged.
Hope that helps
Rolf

Configuring SNMP Trap receiver on AIP-SSM sensor

I receive the following error message from my ASA5520 firewall when attempting to forward SNMP traps from my AIP-SSM20 sensor to a server on my Inside interface that is configured to receive SNMP traps:
ASA-4-418001: Through-the-device packet to/from management-only network is denied: udp src management: 10.3.21.2/32768 dst Inside: PPC0ES/162
Can I reconfigure the management IP address of the AIP-SSM sensor to connect to the Inside interface instead of the management vlan or does my SNMP server have to reside on the management vlan with the sensor?

Hi Subodh,
Yes, the AIP-SSM can operate in either inline (IPS) or promiscuous (IDS) mode. I would recommend you start by reviewing the following config guide, which shows you how to configure the ASA to pass traffic to the SSM for inspection:
http://www.cisco.com/en/US/products/ps6120/products_configuration_example09186a00807335ca.shtml
If you have any other specific questions, feel free to post back.
Hope that helps.
-Mike

Snmp trap versus syslog message

Hi,
Most network devices will send snmp traps and syslog messages to a central server.
For analyzing purpose this server runs software to display the messages or traps.
My question is, what is the difference between syslog messages and snmp traps?
What is best practise?
Thank you very much.
Hansruedi

From the very basic level, traps and syslog differ in the encoding. Syslog messages are typically text messages sent within a UDP packet. There is a bit of binary encoding to indicate the syslog facility and severity. SNMP traps have encoded ASN.1 fields (called variable bindings). These varbinds are not ASCII text like syslog messages. Instead they are encoded object identifiers that can be translated into object names using MIB definitions.
More syslog messages exist than SNMP traps because syslog messages do not have as much governance associated with them. However, we typically recommend that customers enable both as there are some details available in traps that you may not get in syslog messages. Traps can also be processed in a more programmatic fashion because of the documentation that goes into the MIBs that define them.

SNMP traps send by Java Classes

Hi,
I am now trying to create a java service to generate SNMP traps.
I found AdventNet classes but I think they are more for building a management tool.
What I need is just a java package that I can use in my application.
Thank you for the comments,
F.

1.) I can recommend the OpenSource SNMP library joeSNMP�supporting SNMPv1 and SNMPv2.
http://www.opennms.org/files/releases/joeSNMP/latest
It has some little bugs regarding the sending of traps (because its original purpose is to allow development of NMS), but they can be easily overcome. Because it's completely free it should be your first choice for developing a simple agent.
2. AdventNet offers an agent toolkit which is very easy to use (unfortunately rather expensive), also from iReasoning you can obtain a JMX agent toolkit with an SNMP adapter, both supporting SNMPv1-v3. JDMK would also be an option (still I have no experience with it).
3. Take a look at http://www.snmplink.org/
Regards
Hounce

ACE and ANM, Syslog and SNMP Traps

Hi guys.. another ACE/ANM question.
I configured the ACE devices to send Syslog and SNMP messages to the ANM server. But i got a couple of questions:
Whats the difference between using the:
logging history 4 (this would send logging messages as SNMP traps according to doc)
And:
snmp-server host x.x.x.x traps version 2c public
snmp-server trap-source vlan 1000
This of course I think should do the same..
The funny and weird thing, in the ANM Event viewer, I can only see syslog messages, not one snmp event.
Thanks!
Omar
PS: ACE ver A2.4
ANM Ver 4.2

Hi Omar,
Let's see if I can clarify your questions.
As you mentioned, the "logging history 4" command specifies that, syslog messages of severity 4 and higher will be sent as SNMP traps. After you configure it, you need the "snmp-server host x.x.x.x traps version 2c public" command to specify what will be the destination IP and SNMP community for these traps.
It would only make sense to use the "logging history 4" command if your monitoring application doesn't support receiving syslog messages. However, since ANM is able to get syslog messages from the ACE without issues, I would just configure a destination for syslog message instead (with "logging host x.x.x.x")
I hope this makes this point more clear.
Now, moving on to why you are not seeing any SNMP traps in your ANM, the first things you would need to check are:
-- Did you enable traps? You would use the "ACE(config)# snmp-server enable traps" command for this
-- Are traps being sent? You can use the "show snmp" command and check if the "Trap PDUs" counter increases
-- Is ANM getting these traps? This is the most complicated step. For this, I would recommend getting a traffic capture on the ANM server (if it's installed on linux) or as close as possible to it if it's a ANM appliance
I hope this helps
Daniel

Questions on Receiving SNMP Traps

Hi:
- I have more questions on receiving SNMP traps:
1) the OEM plug-in can receive traps now, but when I click the metric, I see:
Error getting data for target test20. Exception: ORA-20216: ORA-06512: at "SYSMAN.EMD_MNTR", line 817 ORA-01403: no data found ORA-06512: at line 1
- the push descriptor looks like:
<PushDescriptor RECVLET_ID="SNMPTrap">
<Property NAME="MatchEnterprise" SCOPE="GLOBAL">...</Property>
<Property NAME="MatchGenericTrap" SCOPE="GLOBAL">6</Property> <Property NAME="MatchSpecificTrap" SCOPE="GLOBAL">31</Property> <Property NAME="MatchAgentAddr" SCOPE="INSTANCE">target_ip</Property>
<Property NAME="EventsnChasFanIndexOID" SCOPE="GLOBAL">...</Property>
<Property NAME="ContextsnChasFanDescriptionOID" SCOPE="GLOBAL">...</Property>
<Property NAME="SeverityCode" SCOPE="GLOBAL">WARNING</Property>
</PushDescriptor>
- is the Key Property needed ?
2) The alerts for some reason do not filter back to the all targets home page.
- When I click the Home tab and goto to the 'All Targets' pane, I do not see the alert generated by the OEM plug-in.
- What I am doing wrong ?
3) Is it okay to receive traps with the metric usage set to either: HIDDEN or HIDDEN_COLLECT ?
- Does this cause the errors I see in Q 1) ?
Thanks
John
Edited by: user8826739 on Feb 23, 2010 7:05 AM

Hi John,
Can you post the full definition of the metric? You would need to use the Key property for each key column in the metric description.
With the SNMP receivelet you can set up definitons for data points or alerts. I would assume (as I've never tried this ;) that if you set up the definition to be a data point, you would see data from the All Metrics page. To me, it wouldn't make sense for a metric that used the PushDescriptor SNMPTrap to have data to be viewed as the result of the SNMP trap coming is would be an alert. I will have to look into that. My gut reaction is that a metric with PushDescriptor SNMPTrap shouldn't even appear on the All Metrics page ...
To be clear are you saying that you don't see the Warning number under "All Targets Alerts" increase by 1 when you SNMP trap is caught and alert is generated? When this occurs do you see the alert on the target instance homepage?
In regards to HIDDEN and HIDDEN_COLLECT, I don't know what effect they would have on a metric defined for an SNMP trap to raise an alert. You definitely wouldn't want to use HIDDEN as this setting is for temporary metrics that are used in the caclulation of other metrics. HIDDEN metrics are not collected (or hence uploaded to the OMS) and don't appear in the UI. HIDDEN_COLLECT are collected, but do not show up in the UI and are not uploaded. I've never used this settings with SNMP trap metrics that are for alerts. If your metrics for the SNMP trap alerts do show up on the All Metrics page (I'd have to get something set up to look at this), then it could make sense to use the HIDDEN_COLLECT as the alert would still be generated, but the metric itself wouldn't be shown in the UI.
Let me find out the expected behavior from someone on the agent team.
Dave

What is the minimum server layer OEM version supports SNMP trap reception ?

Hi:
- I have been trying to enable SNMP trap reception on an OEM plug-in.
- I turned on debug channel for recvlets.snmp and saw:
2009-10-16 16:07:42,808 Thread-3028552624 ERROR recvlets.snmp: Duplicate threshold : test900, oracle_guide, interfaces, status
and
2009-10-16 16:09:08,382 Thread-3021634480 INFO recvlets.snmp: Trap received is to convert Data Point
2009-10-16 16:09:08,379 Thread-3021634480 INFO recvlets.snmp: Sending Data Point ...
2009-10-16 16:09:08,379 Thread-3021634480 INFO recvlets.snmp: Listening for TRAP
So, it looks like the OEM agent can receive traps but no data point or alert appears.
And, the agent always issues an error about duplicate thresholds.
- Does the agent have to be patched ?
My agent is:
Oracle Enterprise Manager 10g Release 5 Grid Control 10.2.0.5.0.
Copyright (c) 1996, 2009 Oracle Corporation. All rights reserved.
Agent Version : 10.2.0.5.0
OMS Version : 10.2.0.1.0
Protocol Version : 10.2.0.0.0
Agent is Running and Ready
- on the server layer, the oms is:
Oracle Enterprise Manager 10g Release 10.2.0.1.0
Copyright (c) 1996, 2005 Oracle Corporation. All rights reserved.
Oracle Management Server is Up.
Is a patch needed for OMS ?
Should OMS be version 10.2.0.5.0 ?
Thanks
John
Edited by: user8826739 on Feb 23, 2010 7:17 AM

10.2.0.5 should be fine ...
Dave

Creating SNMP Trap Monitors in 2012 R2

Hello,
For the past week or so I have been attempting to create a MP for my company's UPSs to be monitored in SCOM 2012 R2. I have created a custom MP with custom datasources and objects. All seems to go according to the plan until I attempt to create a SNMP
Trap monitor for those objects. I am able to verify that the device is discovered and is sending the traps successfully with a view. The problem seems to lie within the setup of the monitor. I have followed the steps as much as I can from this page http://technet.microsoft.com/en-us/library/hh563870.aspx
to create a monitor that will be triggered by an incoming trap saying the UPS is on Battery Power.
This is the initial or trigger event information received by SCOM:
<EventData>
<DataItem type="System.SnmpData" time="2014-07-03T08:48:07.6990732-07:00" sourceHealthServiceId="26DF02E5-DEA0-BED0-443B-3605A05D4790">
<Source>10.51.85.40</Source>
<Destination>127.0.0.1</Destination>
<ErrorCode>1</ErrorCode>
<Version>1</Version>
<SnmpVarBinds>
<SnmpVarBind>
<OID>.1.3.6.1.2.1.1.3.0</OID>
<Syntax>67</Syntax>
<Value VariantType="19">17567220</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.6.3.1.1.4.1.0</OID>
<Syntax>6</Syntax>
<Value VariantType="8">.1.3.6.1.4.1.318.0.5</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.4.1.318.2.3.3.0</OID>
<Syntax>4</Syntax>
<Value VariantType="8">UPS: On battery power in response to an input power problem.</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.6.3.1.1.4.3.0</OID>
<Syntax>6</Syntax>
<Value VariantType="8">.1.3.6.1.4.1.318</Value>
</SnmpVarBind>
</SnmpVarBinds>
</DataItem>
</EventData>
This is the event should signal that everything is operating normally once again.
<EventData>
<DataItem type="System.SnmpData" time="2014-07-03T08:48:12.9133769-07:00" sourceHealthServiceId="26DF02E5-DEA0-BED0-443B-3605A05D4790"><Source>10.51.85.40</Source>
<Destination>127.0.0.1</Destination>
<ErrorCode>1</ErrorCode>
<Version>1</Version>
<SnmpVarBinds>
<SnmpVarBind>
<OID>.1.3.6.1.2.1.1.3.0</OID>
<Syntax>67</Syntax>
<Value VariantType="19">17567730</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.6.3.1.1.4.1.0</OID>
<Syntax>6</Syntax>
<Value VariantType="8">.1.3.6.1.4.1.318.0.9</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.4.1.318.2.3.3.0</OID>
<Syntax>4</Syntax>
<Value VariantType="8">UPS: No longer on battery power.</Value>
</SnmpVarBind>
<SnmpVarBind>
<OID>.1.3.6.1.6.3.1.1.4.3.0</OID>
<Syntax>6</Syntax>
<Value VariantType="8">.1.3.6.1.4.1.318</Value>
</SnmpVarBind>
</SnmpVarBinds>
</DataItem>
</EventData>
You may notice that these traps are being sent as SNMP v1. I am unable to change that on the UPS so i have modified the monitors to account for that. Here is the code for the monitor:
<UnitMonitor ID="UIGeneratedMonitorc2c65f3c85fe42f48ce8d2580c57bbae" Accessibility="Public" Enabled="true" Target="APC.MP.UPS" ParentMonitorID="Health!System.Health.AvailabilityState" Remotable="true" Priority="Normal" TypeID="SNL!System.NetworkManagement.SnmpTrapProvider.2SingleEvent2StateMonitorType" ConfirmDelivery="false">
<Category>Custom</Category>
<AlertSettings AlertMessage="UIGeneratedMonitorc2c65f3c85fe42f48ce8d2580c57bbae_AlertMessageResourceID">
<AlertOnState>Warning</AlertOnState>
<AutoResolve>true</AutoResolve>
<AlertPriority>Normal</AlertPriority>
<AlertSeverity>Error</AlertSeverity>
</AlertSettings>
<OperationalStates>
<OperationalState ID="UIGeneratedOpStateId2750c1ab2cd14b4a8a8e9f4633eba0b0" MonitorTypeStateID="SecondEventRaised" HealthState="Success" />
<OperationalState ID="UIGeneratedOpStateId5753748399984d08953c1a46895df99c" MonitorTypeStateID="FirstEventRaised" HealthState="Warning" />
</OperationalStates>
<Configuration>
<FirstIP>$Target/Property[Type="SNL!System.NetworkManagement.Node"]/SNMPAddress$</FirstIP>
<FirstVersion>1</FirstVersion>
<FirstOIDProps>
<OIDProp>.1.3.6.1.4.1.318</OIDProp>
</FirstOIDProps>
<FirstExpression>
<SimpleExpression>
<ValueExpression>
<XPathQuery Type="String">SnmpVarBinds/SnmpVarBind[OID=".1.3.6.1.6.3.1.1.4.1.0"]/Value</XPathQuery>
</ValueExpression>
<Operator>Equal</Operator>
<ValueExpression>
<Value Type="String">.1.3.6.1.4.1.318.0.5</Value>
</ValueExpression>
</SimpleExpression>
</FirstExpression>
<SecondIP>$Target/Property[Type="SNL!System.NetworkManagement.Node"]/SNMPAddress$</SecondIP>
<SecondVersion>1</SecondVersion>
<SecondOIDProps>
<OIDProp>.1.3.6.1.4.1.318</OIDProp>
</SecondOIDProps>
<SecondExpression>
<SimpleExpression>
<ValueExpression>
<XPathQuery Type="String">SnmpVarBinds/SnmpVarBind[OID=".1.3.6.1.6.3.1.1.4.1.0"]/Value</XPathQuery>
</ValueExpression>
<Operator>Equal</Operator>
<ValueExpression>
<Value Type="String">.1.3.6.1.4.1.318.0.9</Value>
</ValueExpression>
</SimpleExpression>
</SecondExpression>
</Configuration>
</UnitMonitor>
I tried the following things while attempting to troubleshoot this issue:
Leaving First
SnmpTrapProvider and Second
SnmpTrapProvider blank while creating the monitor.
Adding "" and also '' around the value as it is a string and perhaps that was the cause.
Changing the First and Second SNMPTrapProvider to .1.3.6.1.6.3.1.1.4.1.0 since that is the specific OID I am looking for the value of.
Reading every page of the internet and bashing my head on the keyboard.
I am guessing that the issue will reside in the the monitor creation and definition, but I have not found anything that clearly describes how to fill out the SNMP Trap Monitor wizard given the trap info. Especially since the trap I am receiving does
not seem to contain the usual OID with a common Int or string value. Hopefully someone can provide assistance as I am nearing my wits end with this problem.

Firstly, thank you Michael Skov, knowing that the First and Second SnmpTrapProvider or <FirstOIDProps> in the xml, needed to be blank got me pointed in the right direction. I figured I would come back and share my final working code snippet in case
others run across a similar issue.
<UnitMonitor ID="UIGeneratedMonitorf9f671e5456c44fd807d6ec2bd09621d" Accessibility="Public" Enabled="true" Target="Disc!APC.Disc.UPS" ParentMonitorID="Health!System.Health.PerformanceState" Remotable="true" Priority="Normal" TypeID="NetworkManagement!System.NetworkManagement.SnmpTrapProvider.2SingleEvent2StateMonitorType" ConfirmDelivery="false">
<Category>Custom</Category>
<AlertSettings AlertMessage="UIGeneratedMonitorf9f671e5456c44fd807d6ec2bd09621d_AlertMessageResourceID">
<AlertOnState>Warning</AlertOnState>
<AutoResolve>true</AutoResolve>
<AlertPriority>Normal</AlertPriority>
<AlertSeverity>Warning</AlertSeverity>
</AlertSettings>
<OperationalStates>
<OperationalState ID="UIGeneratedOpStateId8700c95344254f87a60c92d30d4a3953" MonitorTypeStateID="SecondEventRaised" HealthState="Success" />
<OperationalState ID="UIGeneratedOpStateId7bf68c6cfe3d4faabe83e630781c7327" MonitorTypeStateID="FirstEventRaised" HealthState="Warning" />
</OperationalStates>
<Configuration>
<FirstIP>$Target/Property[Type="NetworkManagement!System.NetworkManagement.Node"]/SNMPAddress$</FirstIP>
<FirstVersion>1</FirstVersion>
<FirstOIDProps />
<FirstExpression>
<SimpleExpression>
<ValueExpression>
<XPathQuery Type="String">SnmpVarBinds/SnmpVarBind[OID=".1.3.6.1.6.3.1.1.4.1.0"]/Value</XPathQuery>
</ValueExpression>
<Operator>Equal</Operator>
<ValueExpression>
<Value Type="String">.1.3.6.1.4.1.318.0.5</Value>
</ValueExpression>
</SimpleExpression>
</FirstExpression>
<SecondIP>$Target/Property[Type="NetworkManagement!System.NetworkManagement.Node"]/SNMPAddress$</SecondIP>
<SecondVersion>1</SecondVersion>
<SecondOIDProps />
<SecondExpression>
<SimpleExpression>
<ValueExpression>
<XPathQuery Type="String">SnmpVarBinds/SnmpVarBind[OID=".1.3.6.1.6.3.1.1.4.1.0"]/Value</XPathQuery>
</ValueExpression>
<Operator>Equal</Operator>
<ValueExpression>
<Value Type="String">.1.3.6.1.4.1.318.0.9</Value>
</ValueExpression>
</SimpleExpression>
</SecondExpression>
</Configuration>
</UnitMonitor>

SNMP traps from Solaris 10

Hi,
I'm trying very hard to set up a Sun Fire V245 to send SNMP traps when certain hardware or software related events occur. I've been looking at sma_snmp (net-snmp) and the Fault Management Daemon (SUNWfmd) but they seem to be very limited in their capabilities. I have manged to get some traps sent for filesystem fill-ups and high load averages but that is about it.
Most of all I would like the system to send traps when there is a HW failure such as a faulty FRU or if there are disk failures.
If anyone can point me to some documentation about this, I would be most grateful.
/Mikael

Mikael,
I struggled through the same thing with a Netra 240 recently. The Sun docs are garbage when it comes to this. I opened a ticket with Sun and after 3 days and 6 hours on the phone I finally got hold of someone who knew how to spell SNMP. Yes, it was that bad!
Here's the scoop. In Solaris 10 you run Net-SNMP, a.k.a. SMA, snmpd. The old snmpdx is obsoleted and you shouldn't configure it at all.
Now to get the hardware related traps for the Sunfire and Netra series servers... (what you are really looking for).
You have to load and configure an additional SNMP daemon for the hardware specific traps.
(The first doc is rather old, the last one 819-7978-12 is pretty new and is somewhat more relevant.)
Sun� SNMP Management Agent for Sun Fire� and Netra� Systems: Sun Doc number 817-2559-13
Sun� SNMP Management Agent Addendum for the Netra� 240 Server: Sun Doc Number 817-6238-10
Sun� SNMP Management Agent Administration Guide for Sun Blade� /Sun Fire�/Sun SPARC� Enterprise/Netra� Servers: Sun Doc Number 819-7978-12
And finally the SMA/net-snmp/snmpd guide for the standard Solaris related traps:
Solaris System Management Agent Administration Guide: Sun Doc Number 817�3000�11
There are problems with all of the above documents. None of the Netra/Sunfire docs specifically talk about Solaris 10 so read them with caution. They also talk about configuring and running snmpdx and never reference SMA/net-snmp. This is odd because the instructions I got from Sun (finally) were not to run snmpdx, only to run sma/snmpd and additionally run the sunfire/netra snmpd agent.
The SMA document (817-3000-11) has an undocumented bug, which Sun knows about and is working on but will not reveal to the public. In the section titled "Migration From the Sun Fire Management Agent" it references using a script called masfcnv to convert the sunfire/netra specific snmp config and daemon to work with and through SMA. Since they all use the same ports (161/162) there is some conflict and the masfcnv is script is meant to resolve this by making sma/snmpd a proxy agent to requests toward the sunfire/netra specific hardware daemon.
The problem is the masfcnv script doesn't work properly. In fact, if you run the script you will destroy your other snmp configurations and may have to uninstall and reinstall the packages to clean everything up. This script hasn't ever worked and Sun is working on a fix but they neglect to mention this in the document which is IMO gross negligence and is a reflection of Sun's overall state of affairs (but that's another ranting thread).
So what you must do is configure SMA/net-snmp (or whatever you want to call it), and also configure the sunfire/netra specific snmp (after downloading and installing that package).
Since traps are sent to the remote trapsink using destination port 162, both net-snmp and the netra specific snmp daemons can co-exist here (port 162 is not an open listening port on the machine).
Port 161 is used for receiving SNMP Get requests and can only be bound to one daemon at a time. So either it is used by net-snmp or the netra snmp daemon, but not both. Since my boxes have not been fully integrated still I can't figure out which daemon 161 is bound to. At any rate, in my application the customer is only interested in receiving traps so the outcome here isn't that important.
I realize this isn't complete but I'm no expert here and haven't worked through all the test scenarios on a fully configured system. Hopefully though this will help clear some of the confusion propogated through Sun's stupid documents. Good luck!
/Frank

Snmp trap on solaris 10

Hi,
how can i make my Solaris 10 generate SNMP trap on specific event?
like high CPU,Load or lack of disk space ?
thanks
Ram

it should be possible to define the "index" in each /etc/hostname. file
I have never done this, but when the system boots it reads the content of these files and passes them to "ifconfig" to configure the NICs
search this page for "index":
http://docs.oracle.com/cd/E19253-01/816-5166/ifconfig-1m/index.html

SCOM 2012 SNMP Trap - Does anyone have it working?

I am wondering if anyone out there has had any luck with getting SNMP Traps to be captured by SCOM 2012? Probes work fine.
We have been unable to get SCOM 2012 to receive a SNMP Trap from network devices. We currently have a case open with Microsoft, but they dont have any answers as of yet.
Here are the cliff notes version of what we have done so far.
Discovery - OK
The nodes discover without issue. I made sure they were set to SNMP only. I did this to verify we were talking SNMP and not just ICMP.
Open Ports/Firewall Issues
No firewalls are in place on the SCOM server and none exist between devices we are trying to get traps from. Using a port tool, I can see that UDP 162 is listening as expected and its the System Center Management Host Process that is running the
process. We have tried having SNMP installed, but disabled, installed and running, and uninstalled. Currently we have the SNMP service installed and running. The SNMP Trap Service is NOT installed.
Data Validation
Using Wireshark running on the SCOM box I have been able to validate the OIDs we are searching for are hitting the SCOM server. I installed an SNMP Trap generator and Receiver on my local machine as well as Jalasoft SNMP Simulator. I was able
to discover my local machine as a network device and if I send a TRAP from the TRAP generator on my local machine to the receiver on my local machine I can validate the data is coming through ok.
What has been done on SCOM side
The simplest place to look for data is with an Event Collection Rule and then an Event View Monitor. We have set up an Event Based Monitor to search for an OID that I know to be accurate and then I set up an Event monitor to look for
anything hitting that Collection Rule. The Event monitor showed nothing so we changed the EVent Monitor to look for anything that hit Node which should show any SNMP traps that have hit from network devices. No results.
We tried to run WFAnalyzer from the 2007R2 Authoring Console, but it wont run against 2012 and from what I have found the 2012 edition doesnt exist anywhere yet.
So has anyone succesfully received a SNMP TRAP into SCOM 2012?

Hi,
As this thread has been quiet for a while, we assume that the issue has been resolved. At this time, we will mark it as "Answered" as the previous steps should be helpful for many similar scenarios.
In addition, we’d love to hear your feedback about the solution. By sharing your experience you can help other community members facing similar problems.
Thanks,
Yog Li
TechNet Community Support
How could you possibly mark this as answered? No answer was given, it was quiet for a couple of days including one of which was a holiday in the US. The previous steps are NOT helpful. The question remains, Does anyone have it working?
People have given troubleshooting advice, but nobody has actually answered the simple question. Do they have it actually WORKING?

SNMP Trap Translation Wrong For Hebrew Language

SR 3-4956842281
sev 2
Cus ISRAEL CREDIT CARDS LTD
=====================
Customer is using OMS 10.2.0.5 on linux server.
They have created an user defined metric on DB cluster instance with alert message written in hebrew.
The OMS console showed the alert in hebrew as expected.
However when they send this alert with snmp trap notification to a second system (HP openview for windows) then the message appeared not cleared (with question marks). This second system get messages in hebrew from other systems without any problems.
+Does setting LANG variable at OMS level will impact this
+Any tracing which can help to find the cause of the issue

Closing as issue found at HP openview parameter settings

Recommended SNMP traps

Similar Messages

Maybe you are looking for