gmane.network.opennms.general

Re: Email Notifications for Alarms based on SNMPTraps

Mike Diehn — 2013-05-25T15:20:33

I'm still not sure I understand. Do you mail over and over again until the
deaf tunnel comes back? I don't know how to do that. I might try something
in vacuumd or a stored procedure in the db.

Sincerely,
Mike
--
Mike Diehn
DevOps SysAdmin
CD-adapco, Lebanon, NH USA
On May 25, 2013 11:11 AM, "Ian MacDonald" <IMacDonald< at >quartetservice.com>
wrote:

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: Email Notifications for Alarms based on SNMP Traps

Ian MacDonald — 2013-05-25T15:08:58

On Thu, 2013-05-16 at 13:51 -0400, Mike Diehn wrote:

Do you mean that after the "Tunnel B is up again" e-mail arrives, you want another to arrive saying "So glad B is back, but Tunnel A is still down, just in case you forgot about that!!"

Yes, or more specifically,   Not receive any emails relating to tunnel B (because it came back up in under 5 minutes),  but still receive an email that Tunnel A is down.

If I add timestamps to the actual tunnel events, this becomes clearer.

H:MM - Event
0:00  - Tunnel Down (Varbind = A)
0:02  - Tunnel Down (Varbind = B)
0:03 - Tunnel Up (Varbind = B)

At 0:06, I'd like to be notified that Tunnel A is still down.     Even if I could be notified "That any tunnel is still down" would meet my minimum requirement, which is just to know something is wrong, but not have to deal with noise from flapping tunnels (due to NAT-T on a low bandwidth connection, or during re-key).

So, any insight on how to do this?

cheers,
Ian


On Thu, May 16, 2013 at 10:39 AM, Ian MacDonald <IMacDonald< at >quartetservice.com<mailto:IMacDonald< at >quartetservice.com>> wrote:
Kelly,

I have ready that page (just re-read to make sure nothing new there).

The event notifications for VPN trap down and up work.  An email arrives when a tunnel goes down; It also arrives when it goes up. And I can delay that by 5 minutes, and tune the destination path for those events without problem.

I want to do this for a specific alarm.

Here is why.  Take the scenario where 2 VPN tunnels are configured (A and B).  The following events occur.

a) Tunnel Down Trap for Tunnel A arrives
b) Tunnel Down Trap for Tunnel B arrives
c) Tunnel UP Trap for Tunnel B arrives

Currently OpenNMS will generate events and emails for all 3 events.
Based on my event configuration, it will generate 2 alarms, and clear one of them on the third event when the Tunnel B trap arrives.

In the Alarms section, because the Reduction-Key includes the tunnel endpoint, I have two VPN-Down alarms.   By using the endpoint in the reduction-key, I can have one alarm for each VPN tunnel.  I can use the clear-key to correlate the up-tunnel trap and clear the alarm.

I would like to receive an email that is telling me that Tunnel A is still down, based on the OpenNMS alarm state.

cheers,
Ian


On Thu, 2013-05-16 at 15:03 +0100, Michael Kelly wrote:
Hi Ian,

have a look at http://www.opennms.org/wiki/Configuring_notifications

If I understand it correctly, you need to configure the destination path with an initial delay of 5m (minutes).

When a 'tunnel-down' event occurs, OpenNMS will process the notification. It will then wait 5 minutes to see if the event has been acknowledged (either manually or by a 'tunn el-up' event).

If the event has not been acknowledged by the end of the delay period, the notification process will continue.


Hope this helps.


Regards,
Michael.



On 16 May 2013 13:52, Ian MacDonald <IMacDonald< at >quartetservice.com<mailto:IMacDonald< at >quartetservice.com>> wrote:
I am trying to setup OpenNMS to monitor VPN tunnels.

My goaI is to have email notifications from OpenNMS alarm, triggered by a VPN tunnel-down trap, when a VPN tunnel-up trap (that clears alarm state) has not arrived within 5 minutes.

Normally we use ICMP packets NAT'd over the tunnel for monitoring notification purposes, but are constrained to doing this monitoring only with traps in this particular scenario.

The hardware we are receiving traps from is a Fortinet 60B that has multiple IPSEC VPN tunnels configured to connect to similarly configured Fortinet devices at remote locations.

So far, I have updated event definitions to properly identity the VPN traps (Slight modification to existing OpenNMS Fortinet v3 Event xml to support the FortIOS v4 MIBs). I have included these below for reference.

Currently we can receive traps with OpenNMS and generate email notifications for those traps indicating Up/Down. The VPN Up and VPN Down trap notifications contain appropriate varbinds to identify the particular tunnel by its remote tunnel endpoint.

Our goal is to use Alarms so that we can correlate the Down/Up traps for a specific location using the tunnel endpoint pararmeter in the Alarm reduction label.  We also want to suppress notifications for brief tunnel flaps, or delay notification to see if the situation resolves itself.

Reading the following posts, I am able to generate events, and generate notifications for those events, and I now even have Alarms setup that seem to reflect the events.

We can't seem to figure out how to trigger notifications from the alarms, as the emails still arrive based on the event state, not the alarm state.

Specifically, I want an email when <alarm-data reduction-key="%parm[#3]%!%parm[#4]%!%nodeid%"> is triggered, and would like to suppress/delay it when the clear-key is received.

http://www.opennms.org/wiki/Configuring_alarms
http://www.opennms.org/wiki/Creating_Threshold_Alarms

Below is my event.xml for the Traps we are generating alarms for:

<event>
        <mask>
                <maskelement>
                        <mename>id</mename>
                        <mevalue>.1.3.6.1.4.1.12356.101.2</mevalue>
                </maskelement>
                <maskelement>
                        <mename>generic</mename>
                        <mevalue>6</mevalue>
                </maskelement>
                <maskelement>
                        <mename>specific</mename>
                        <mevalue>301</mevalue>
                </maskelement>
        </mask>
        <uei>uei.opennms.org/traps/fortinet/fnTrapVpnTunUp<http://uei.opennms.org/traps/fortinet/fnTrapVpnTunUp></uei>
        <event-label>FORTIOS-400-MIB defined trap event: fnTrapVpnTunUp</event-label>
        <descr>
&lt;p&gt;Indicates that the specified VPN tunnel has been brought up.&lt;/p&gt;&lt;table&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnSysSerial&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#1]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        sysName&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#2]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnVpnTrapLocalGateway&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#3]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnVpnTrapRemoteGateway&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#4]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/table&gt;
        </descr>
                <logmsg dest='logndisplay'>&lt;p&gt;
                        fnTrapVpnTunUp trap received
                        fnSysSerial=%parm[#1]%
                        sysName=%parm[#2]%
                        fnVpnTrapLocalGateway=%parm[#3]%
                        fnVpnTrapRemoteGateway=%parm[#4]%&lt;/p&gt;
                </logmsg>
        <severity>Cleared</severity>
        <alarm-data reduction-key="%parm[#3]%:%parm[#4]%:%nodeid%" clear-key="%parm[#3]%!%parm[#4]%!%nodeid%" alarm-type="2" auto-clean="false"/>

</event>
<event>
        <mask>
                <maskelement>
                        <mename>id</mename>
                        <mevalue>.1.3.6.1.4.1.12356.101.2</mevalue>
                </maskelement>
                <maskelement>
                        <mename>generic</mename>
                        <mevalue>6</mevalue>
                </maskelement>
                <maskelement>
                        <mename>specific</mename>
                        <mevalue>302</mevalue>
                </maskelement>
        </mask>
        <uei>uei.opennms.org/traps/fortinet/fnTrapVpnTunDown<http://uei.opennms.org/traps/fortinet/fnTrapVpnTunDown></uei>
        <event-label>FORTIOS-400-MIB defined trap event: fnTrapVpnTunDown</event-label>
        <descr>
&lt;p&gt;The specified VPN tunnel has been brought down.&lt;/p&gt;&lt;table&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnSysSerial&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#1]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        sysName&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#2]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnVpnTrapLocalGateway&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#3]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;
        &lt;tr&gt;&lt;td&gt;&lt;b&gt;

        fnVpnTrapRemoteGateway&lt;/b&gt;&lt;/td&gt;&lt;td&gt;
        %parm[#4]%;&lt;/td&gt;&lt;td&gt;&lt;p&gt;&lt;/p&gt;&lt;/td&gt;&lt;/tr&gt;&lt;/table&gt;
        </descr>
                <logmsg dest='logndisplay'>&lt;p&gt;
                        fnTrapVpnTunDown trap received
                        fnSysSerial=%parm[#1]%
                        sysName=%parm[#2]%
                        fnVpnTrapLocalGateway=%parm[#3]%
                        fnVpnTrapRemoteGateway=%parm[#4]%&lt;/p&gt;
                </logmsg>
        <severity>Major</severity>
        <alarm-data reduction-key="%parm[#3]%!%parm[#4]%!%nodeid%" alarm-type="1" auto-clean="false"/>

</event>



------------------------------------------------------------------------------
AlienVault Unified Security Management (USM) platform delivers complete
security visibility with the essential security capabilities. Easily and
efficiently configure, manage, and operate all of your security controls
from a single console and one unified framework. Download a free trial.
http://p.sf.net/sfu/alienvault_d2d
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss




--
Ian B. MacDonald | Director, Network Services
___________________________________
Quartet Service Inc.
1867 Yonge Street
Suite 600
Toronto, ON, M4S 1Y5
o. 416-483-8332 Ext.2138<tel:416-483-8332%20Ext.2138>
c. 416-550-1728<tel:416-550-1728>

Service Desk
GTA: 416-483-4357<tel:416-483-4357>
Toll Free: 1-877-442-4357<tel:1-877-442-4357>

www.quartetservice.com<http://www.quartetservice.com/>

[Quartet Service Inc.]<http://www.quartetservice.com/>

Note: This email and any attachments are confidential and are intended only for the person(s) to whom it is addressed. If you are not the addressee named above, any use, copying, distribution or disclosure is strictly unauthorized. If you have received this information in error, please delete it and any attachments and notify me immediately by return email or by calling 416-483-8332<tel:416-483-8332>.

------------------------------------------------------------------------------
AlienVault Unified Security Management (USM) platform delivers complete
security visibility with the essential security capabilities. Easily and
efficiently configure, manage, and operate all of your security controls
from a single console and one unified framework. Download a free trial.
http://p.sf.net/sfu/alienvault_d2d
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss




--
Mike Diehn
Development Operations
CD-adapco - Lebanon, NH
603 643 9993 x24129<tel:603%20643%209993%20x24129>


--
Ian B. MacDonald | Director, Network Services
___________________________________
Quartet Service Inc.
1867 Yonge Street
Suite 600
Toronto, ON, M4S 1Y5
o. 416-483-8332 Ext.2138
c. 416-550-1728

Service Desk
GTA: 416-483-4357
Toll Free: 1-877-442-4357

www.quartetservice.com<http://www.quartetservice.com/>

[Quartet Service Inc.]<http://www.quartetservice.com/>

Note: This email and any attachments are confidential and are intended only for the person(s) to whom it is addressed. If you are not the addressee named above, any use, copying, distribution or disclosure is strictly unauthorized. If you have received this information in error, please delete it and any attachments and notify me immediately by return email or by calling 416-483-8332<tel:416-483-8332>.
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Removing capsd?

Jos Backus — 2013-05-24T22:12:54

Hi,

What is the plan to remove capsd and move entirely to provisiond? This
would include updating the documentation which today largely uses capsd for
its examples, as if it's not deprecated at all, and lacks matching
provisiond examples in many cases. It's frustrating because it adds
needless complication to trying to understand how things work, and without
proper documentation describing how to use provisiond it's hard or
downright impossible to migrate.

Why keep capsd around, if its functionality is entirely subsumed by
provisiond? And if that's not the case, why not make code
simplification/cleanup a priority by making provisiond reach feature parity
first? I'm just trying to understand the prioritization here, and what the
project's goals are.

Thanks for any light you can shed on this!

Cheers,
Jos

Re: openNMS interesting metrics/graphs

John Blake — 2013-05-24T14:39:15

Seems like it could be done by;

option 1, Use the local snmpd to launch a script to get the data and have
ONMS query it via snmp
option 2. Use ONMS to screen scrape itself. have it log into itself and run
the charts and get the data from that screen.
option 3. Since ONMs can grab data from dbases , have it log into its own
dbase as a service and the queries and put that in an rrd.

John

From:"Seibold, Michael" <Michael.Seibold< at >gkvi.de>
To:General OpenNMS Discussion
<opennms-discuss< at >lists.sourceforge.net>,
Date:05/24/2013 10:00 AM
Subject:Re: [opennms-discuss] openNMS interesting metrics/graphs

Hi Manuel,

great idea!

I added following graphs to our “Charts” – page:
- Number of threshold events (…exceeded, …rearmed) per minute
for today
- Number of threshold events (…exceeded, …rearmed) per minute
for yesterday
- Number of threshold events (…exceeded, …rearmed) per day for
the last two months
- Number of events per hour for the last two months

After some time using those graphs you will get used to the typical
“picture” they have and you will most probably see when there is something
strange going on.

It might even be more performant and handy to find some way to write some
sort of data collection for those values and store them in rrd files. This
way it should be possible to define thresholds and generate
alarms/notification. I once did this using automations in vacuumd to count
the number of ospf neighbor changes coming via syslog-events. Too many
changes -> generate notification. Using datacollections would be much nicer
as everything else like thresholding, graphing etc. is working as always.

-Michael

Von: Manuel Villarejo [mailto:mjvillarejo< at >gmail.com]
Gesendet: Freitag, 24. Mai 2013 13:09
An: General OpenNMS Discussion
Betreff: [opennms-discuss] openNMS interesting metrics/graphs

Hello,

I just tried to implement some graphs (and future thresholds) in openNMS to
measure the work that openNMS does.
I found a couple of them which are interesting like:

- Number of events
eventtime BETWEEN (CURRENT_TIMESTAMP - INTERVAL '1 day') AND
CURRENT_TIMESTAMP;", eventCount],

- number of nodes
'D';", nodeCount],

- ongoing alarms vs cleared
# select the outgoing alarms
alarms where alarmtype=1;", AlarmCountCur],
# select the outgoing cleared alarms
alarms where alarmtype=2;", AlarmCountClear],

- active services monitors vs unmanaged
# number of active services
where status='A';", ServActive],
# number of disabled services
ifservices where status='N';", ServDisabled]

if you consider they are not 'quite' right or if you have more suggestion,
please feel free to share here so we can try to compose a set of them.

Regards

--
Manuel Villarejo
e: mjvillarejo< at >gmail.com

AVISO LEGAL
Este mensaje va dirigido exclusivamente a su destinatario y es
confidencial. Si por error lo recibe, por favor, comuníquelo por esta vía y
elimínelo. Cualquier uso de este mensaje o sus anexos sin autorización está
prohibida por ley.

LEGAL NOTICE
This message is intended to be read exclusively by the addressee and is
confidential. Should you receive it by error, please contact the sender by
email and delete it. Any use of this message or its attachments without due
authorisation is prohibited by law.
------------------------------------------------------------------------------

Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service

that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of
this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: weeding out unwanted link-up-downtrapsfromswitches

Seibold, Michael — 2013-05-24T14:20:57

Yes, that's what I would try.

 

-Michael

 

Von: Mike Diehn [mailto:mike.diehn< at >cd-adapco.com] 
Gesendet: Freitag, 24. Mai 2013 16:09
An: General OpenNMS Discussion
Betreff: Re: [opennms-discuss] weeding out unwanted link-up-down
trapsfromswitches

 


are you suggesting creating event definitions that match these traps and
donotpersist them just before the more general defn that sends link
up/down traps to the translator? 

Sincerely,
Mike
--
Mike Diehn
DevOps SysAdmin
CD-adapco, Lebanon, NH USA

On May 24, 2013 10:03 AM, "Seibold, Michael" <Michael.Seibold< at >gkvi.de>
wrote:

Hi Mike,

 

can you distinguish the enduser ports from uplink ports in some way
within the traps - portname configured as "uplink" or something like
this? Than you could change the event definition for Link up/down traps
filtering out the unwanted traps (there is the possibility to not
display/not store them).

 

-Michael

 

 

Von: Mike Diehn [mailto:mike.diehn< at >cd-adapco.com] 
Gesendet: Freitag, 24. Mai 2013 15:15
An: General OpenNMS Discussion
Betreff: [opennms-discuss] weeding out unwanted link-up-down traps
fromswitches

 

 

Hi folks,

 

We have a lot of laptops that dock and undock throughout the day and
their docks are connected to Dell PE 5548 switches.  Those switches send
a nearly continuous stream of linkUp and linkDown traps to OpenNMS,
along with a few more interesting traps.

 

My problem is that all those linkUp and linkDown traps distort our view
in OpenNMS of the health of those switches.

 

I don't see a method of preventing the switches sending just those
traps. (if you know the secret, please share!)

 

I'd like to just discard them as they come into OpenNMS, either at trapd
or eventd. Maybe the event translator? I'm pretty sure I could use
vacuumd to clean them out of the DB.

 

Do you have any suggestions for me?

 

Sincerely,

Mike

Re: openNMS interesting metrics/graphs

Manuel Villarejo — 2013-05-24T14:17:31

Hello!

< at >Markus, yeah, I got them enabled as well but I'm afraid I do not have
enough information of how to check i.e. which operations are pending to be
processed, is there any page where I can get that info?

< at >Michael, they seem to be interesting metrics as well, I'm creating graphs
out of the values I'm collecting so if you want to share your queries /
graphs I can summarize them together and share them right after.

thanks


On Fri, May 24, 2013 at 3:58 PM, Seibold, Michael
<Michael.Seibold< at >gkvi.de>wrote:

Re: weeding out unwanted link-up-down trapsfromswitches

Mike Diehn — 2013-05-24T14:09:06

are you suggesting creating event definitions that match these traps and
donotpersist them just before the more general defn that sends link up/down
traps to the translator?

Sincerely,
Mike
--
Mike Diehn
DevOps SysAdmin
CD-adapco, Lebanon, NH USA
On May 24, 2013 10:03 AM, "Seibold, Michael" <Michael.Seibold< at >gkvi.de>
wrote:

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: weeding out unwanted link-up-down trapsfromswitches

Seibold, Michael — 2013-05-24T14:02:15

Hi Mike,

 

can you distinguish the enduser ports from uplink ports in some way
within the traps - portname configured as "uplink" or something like
this? Than you could change the event definition for Link up/down traps
filtering out the unwanted traps (there is the possibility to not
display/not store them).

 

-Michael

 

 

Von: Mike Diehn [mailto:mike.diehn< at >cd-adapco.com] 
Gesendet: Freitag, 24. Mai 2013 15:15
An: General OpenNMS Discussion
Betreff: [opennms-discuss] weeding out unwanted link-up-down traps
fromswitches

 

 

Hi folks,

 

We have a lot of laptops that dock and undock throughout the day and
their docks are connected to Dell PE 5548 switches.  Those switches send
a nearly continuous stream of linkUp and linkDown traps to OpenNMS,
along with a few more interesting traps.

 

My problem is that all those linkUp and linkDown traps distort our view
in OpenNMS of the health of those switches.

 

I don't see a method of preventing the switches sending just those
traps. (if you know the secret, please share!)

 

I'd like to just discard them as they come into OpenNMS, either at trapd
or eventd. Maybe the event translator? I'm pretty sure I could use
vacuumd to clean them out of the DB.

 

Do you have any suggestions for me?

 

Sincerely,

Mike

Re: openNMS interesting metrics/graphs

Seibold, Michael — 2013-05-24T13:58:44

Hi Manuel,

great idea!

I added following graphs to our "Charts" - page:

- Number of threshold events (...exceeded, ...rearmed) per minute for today

- Number of threshold events (...exceeded, ...rearmed) per minute for yesterday

- Number of threshold events (...exceeded, ...rearmed) per day for the last two months

- Number of events per hour for the last two months

After some time using those graphs you will get used to the typical "picture" they have and you will most probably see when there is something strange going on.

It might even be more performant and handy to find some way to write some sort of data collection for those values and store them in rrd files. This way it should be possible to define thresholds and generate alarms/notification. I once did this using automations in vacuumd to count the number of ospf neighbor changes coming via syslog-events. Too many changes -> generate notification. Using datacollections would be much nicer as everything else like thresholding, graphing etc. is working as always.

-Michael

Von: Manuel Villarejo [mailto:mjvillarejo< at >gmail.com]
Gesendet: Freitag, 24. Mai 2013 13:09
An: General OpenNMS Discussion
Betreff: [opennms-discuss] openNMS interesting metrics/graphs

Hello,

I just tried to implement some graphs (and future thresholds) in openNMS to measure the work that openNMS does.

I found a couple of them which are interesting like:

- Number of events

- number of nodes

- ongoing alarms vs cleared

# select the outgoing alarms

# select the outgoing cleared alarms

- active services monitors vs unmanaged

# number of active services

# number of disabled services

if you consider they are not 'quite' right or if you have more suggestion, please feel free to share here so we can try to compose a set of them.

Regards

Re: openNMS interesting metrics/graphs

Markus Neumann — 2013-05-24T13:30:47

Am Fr 24 Mai 2013 13:09:03 CEST schrieb Manuel Villarejo:

Hey Manuel,

to monitor OpenNMS and the jvm it's using you can checkout the jmx 
etc/examples /jvm or jmx folder.
There are datacollection and graph definitions for OpenNMS.

Cu

Tak

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

weeding out unwanted link-up-down traps fromswitches

Mike Diehn — 2013-05-24T13:15:27

Hi folks,

We have a lot of laptops that dock and undock throughout the day and their
docks are connected to Dell PE 5548 switches.  Those switches send a nearly
continuous stream of linkUp and linkDown traps to OpenNMS, along with a few
more interesting traps.

My problem is that all those linkUp and linkDown traps distort our view in
OpenNMS of the health of those switches.

I don't see a method of preventing the switches sending just those traps.
(if you know the secret, please share!)

I'd like to just discard them as they come into OpenNMS, either at trapd or
eventd. Maybe the event translator? I'm pretty sure I could use vacuumd to
clean them out of the DB.

Do you have any suggestions for me?

Sincerely,
Mike

openNMS interesting metrics/graphs

Manuel Villarejo — 2013-05-24T11:09:03

Hello,

I just tried to implement some graphs (and future thresholds) in openNMS to
measure the work that openNMS does.
I found  a couple of them which are interesting like:

- Number of events
eventtime BETWEEN (CURRENT_TIMESTAMP - INTERVAL '1 day') AND
CURRENT_TIMESTAMP;", eventCount],

- number of nodes
'D';", nodeCount],

- ongoing alarms vs cleared
# select the outgoing alarms
alarms where alarmtype=1;", AlarmCountCur],
# select the outgoing cleared alarms
alarms where alarmtype=2;", AlarmCountClear],
 - active services monitors vs unmanaged
# number of active services
where status='A';", ServActive],
# number of disabled services
ifservices where status='N';", ServDisabled]

if you consider they are not 'quite' right or if you have more suggestion,
please feel free to share here so we can try to compose a set of them.

Regards

Re: Delaying OpenNMS Sending Notification

Ron Roskens — 2013-05-24T01:35:44

When you create the destination path, you can specify an initial delay 
to pause for before it sends the notification out. You can also edit the 
destination path later on. (Admin > Configure Notifications > Configure 
Destination Paths)

Ron

On 5/23/2013 7:33 PM, Gary Jarrel wrote:


------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: Delaying OpenNMS Sending Notification

TAN Lists — 2013-05-24T01:03:27

Yep there is the facility for both of those.

You should probably consider raising the timeout rather than the
notification time

Depending on your setup it will be in capsd-configuration.xml or
poller-configuration.xml


On 24 May 2013 10:33, Gary Jarrel <garyjarrel< at >gmail.com> wrote:

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Delaying OpenNMS Sending Notification

Gary Jarrel — 2013-05-24T00:33:13

Hi All,

I am reasonably new to OpenNMS and we are trialing it out to monitor a few
remote servers that are at our site offices and are connected to head
office via an ipsec VPN where the OpenNMS Server is installed.

What we are getting is some false notifications of services going down at
the remote sites. Probably due to heavy traffic across the VPNs. Eg, we
would get a HTTP service down at a remote site notification and then 60
seconds later (approximately) we get the RESOLVED notification.

Is there a way to delay the notification, such that at least two
consecutive failures have to occur before the notification is raised, or
even to increase the time out before notification is sent!

Thank you for your assistance. It seems something simple but I've been
reading the tutorials, Googling and can't seem to find the solution.

Thank you

GJ
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: Getting min/max/avg values

Les Mikesell — 2013-05-23T17:53:28

Something like this? (Oddly, a cut/paste from old browser history).

http://172.22.181.29:8980/opennms/summary/results.htm?filterRule=ipaddr+iplike+172.22.181.162&startTime=1366999000&endTime=1366999029&attributeSieve=CpuRawKernel|CpuRawSystem|CpuRawWait

Change the server IP, the node IP, and the times (and maybe the
specific snmp items).

If your browser isn't logged in, you'll get a redirect for
authentication but it should go on and work. To automate it you'll
need something that knows how to follow the redirection. The
parseurl() from perl's XML::Twig works for me.

Note that the default setup for the jrbs is to start aggregating
timeslices after 2 weeks so you lose resolution, and the values from
the data export interface are at the worst resolution of the timespan
you request. That is if you had a short spike at 1 week ago and ask
for a 3 week interval, you'll get the downsampled 'max' instead of the
actual sample value you would get if you ask for a week. For things
like this I do nightly runs and update my own database whenever the
max is higher - or for some things just save the daily value.

--
Les Mikesell
lesmikesell< at >gmail.com

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: Getting min/max/avg values

Graeme Robinson — 2013-05-23T17:12:33

From: Les Mikesell [mailto:lesmikesell< at >gmail.com] 
Sent: 23 May 2013 17:32

I don't quite follow what you are trying to do.  Can't you pass an ip address and the snmp values that you want to retrieve to the data export interface?  The ip address will just identify the node - you'll actually get back the data from all interfaces in the xml for interface-related values.


----------------

Hi,

Thank you for the reply, I'm specifically trying to get the peak CPU and average CPU usage over the last month for a set of nodes.  I'll use this information to then populate another system.

The opennms system we run is gathering this information, so I need a way of retrieving it out of the opennms rrd (java) files.  That's when I found http://www.opennms.org/wiki/Data_Export .  Which would be ideal if I could get it to respond.

Thank you




------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Re: Getting min/max/avg values

Les Mikesell — 2013-05-23T16:31:54

I don't quite follow what you are trying to do.  Can't you pass an ip
address and the snmp values that you want to retrieve to the data
export interface?  The ip address will just identify the node - you'll
actually get back the data from all interfaces in the xml for
interface-related values.

--
  Les Mikesell
    lesmikesell< at >gmail.com

------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Availability metric calculation

Narayani — 2013-05-22T21:42:32

Hi,

I was wondering how the availability metrics is calculated in opennms. I
see the overall availability metric for a node is an aggregate of values of
all
probes configured  for a particular node. But I am not sure how the
individual availability values are calculated for each probe .
Is there  a specific formula used for calculation. Also, how is the outage
being considered in the calculation.


Thanks,
Narayani
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Detector for page sequence monitor

Narayani — 2013-05-22T22:49:09

Hi,

I am trying to use page sequence monitor to monitor URLs behind a proxy
server. From the
wiki<http://www.opennms.org/wiki/Page_Sequence_Monitor_(PSM)_Setup> ,
I see capability scanning can be configured using HTTP / HTTPS
plugin. I was wondering if there is a suitable detector for page sequence
monitor to configure provisiond or I could configure provision for
HTTP/HTTPS
in the same way as capsd configuration.


Thanks,
Narayani
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss

Getting min/max/avg values

Graeme Robinson — 2013-05-23T15:31:38

Hello,

I'm fairly new to opennms and am trying to gather information from the graphing system system.

I've followed the first example in http://opennms.org/wiki/ResourceID_Replacement combined with the wget format in http://www.opennms.org/wiki/Data_Export without getting any result. (using my own login credentials and a node id that does exist)

I'm specifically trying to hit... https://XXXXX/opennms/summary/results.htm?resourceId=node[XXXX].nodeSnmp[]&reports=all<https://XXXXX/opennms/summary/results.htm?resourceId=node%5bXXXX%5d.nodeSnmp%5b%5d&reports=all>  , I receive a 200 header with 0 bytes.

I can see the following errors in the web-map.log
2013-05-23 08:16:54,077 DEBUG [qtp1668324806-13773001] SecurityAuthenticationEventOnmsEventBuilder: ServletRequestHandledEvent received - url=[/opennms/summary/results.htm]; client=[XX.XX.XX.XX];
method=[GET]; servlet=[dispatcher]; session=[XXXXXXXXXX]; user=[grobinson]; time=[4541ms]; status=[failed: java.lang.NullPointerException]

We are running opennms Version 1.11.90.

Does anyone have a suggestion on either another way of gathering this information, or a suggestion on correcting the error above?

Thank you,
Graeme
------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may_______________________________________________
Please read the OpenNMS Mailing List FAQ:
http://www.opennms.org/index.php/Mailing_List_FAQ

opennms-discuss mailing list

To *unsubscribe* or change your subscription options, see the bottom of this page:
https://lists.sourceforge.net/lists/listinfo/opennms-discuss