General OpenNMS Discussion.

David Poole | 14 Aug 2012 20:10

Thresholding servicedown or nodedown

Hello,

I have a situation in where I do not want notifications sent out unless the service (ie: http, snmp, ftp, etc.) on a node or the node itself (icmp) is down and this is triggered x number of times.

I know I can set triggers on thresholds, but since this is a servicedown and a nodedown event instead of an snmp threshold, is there a way to accomplish this?

Reason: sometimes there is a momentary disconnect in a remote network on the WAN for a couple of seconds. I don't want alert notifications being sent via email for a temporary "outage". I'd rather have openNMS poll a couple more times to verify the nodes are actually down before generating notifications.

Thank you

David

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

_______________________________________________
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

headers

Fuhrmann, Marcel | 14 Aug 2012 21:10

Re: Thresholding servicedown or nodedown

Hi David,

you can configure your destination paths to send notifications delayed. 
I set it to 5 minutes. 

Von meinem iPhone gesendet

Am 14.08.2012 um 20:15 schrieb "David Poole" <david.sydney.poole <at> gmail.com>:

> Hello,
> 
> I have a situation in where I do not want notifications sent out unless the service (ie:  http, snmp, ftp,
etc.) on a node or the node itself (icmp) is down and this is triggered x number of times.
> 
> I know I can set triggers on thresholds, but since this is a servicedown and a nodedown event instead of an
snmp threshold, is there a way to accomplish this?
> 
> Reason:  sometimes there is a momentary disconnect in a remote network on the WAN for a couple of seconds.  I
don't want alert notifications being sent via email for a temporary "outage".  I'd rather have openNMS
poll a couple more times to verify the nodes are actually down before generating notifications.
> 
> 
> Thank you
> 
> David
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and 
> threat landscape has changed and how IT managers can respond. Discussions 
> will include endpoint security, mobile security and the latest in malware 
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> 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

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
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

headers

David Poole | 15 Aug 2012 17:53

Re: Thresholding servicedown or nodedown

Marcel,

Thank you. I have thought of doing that but my primary concern is that when I have threshold alarms that need to be acted upon quickly (ie: disk space running very low or a CPU spike), I don't want notifications like that to be delayed.

It would be nice if there was a "trigger" counter, much like in Thresholds, for the Polling service that could be configured so that the service does not set an event or alarm until the trigger has been fired x number of times. This would be especially useful if it was configurable at the service level in the polling configuration.

David

On Tue, Aug 14, 2012 at 12:10 PM, Fuhrmann, Marcel <Marcel.Fuhrmann <at> lux.ag> wrote:

Hi David,

you can configure your destination paths to send notifications delayed.
I set it to 5 minutes.

Von meinem iPhone gesendet

Am 14.08.2012 um 20:15 schrieb "David Poole" <david.sydney.poole <at> gmail.com>:

> Hello,
>
> I have a situation in where I do not want notifications sent out unless the service (ie: http, snmp, ftp, etc.) on a node or the node itself (icmp) is down and this is triggered x number of times.
>
> I know I can set triggers on thresholds, but since this is a servicedown and a nodedown event instead of an snmp threshold, is there a way to accomplish this?
>
> Reason: sometimes there is a momentary disconnect in a remote network on the WAN for a couple of seconds. I don't want alert notifications being sent via email for a temporary "outage". I'd rather have openNMS poll a couple more times to verify the nodes are actually down before generating notifications.
>
>
> Thank you
>
> David

> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> 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

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
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

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

_______________________________________________
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

headers

David S Hustace | 19 Aug 2012 16:22

Re: Thresholding servicedown or nodedown

On Aug 15, 2012, at 11:53 AM, David Poole <david.sydney.poole <at> gmail.com> wrote:

It would be nice if there was a "trigger" counter, much like in Thresholds, for the Polling service that could be configured so that the service does not set an event or alarm until the trigger has been fired x number of times. This would be especially useful if it was configurable at the service level in the polling configuration.

This behavior is implied when setting the delay on the notification when you consider that OpenNMS not only doesn't stop polling but ale increasing polling, by default, for a short period of time after the outage is detected. This is known as the downtime polling model and is defined in each polling package. In the default configuration, if you set an initial delay in the notification for 2 minutes, then you will have 4 polling attempts after the original failing poll. Also, in you default configuration, this problem means that each of those 4 polling attempts will have 2 tests (initial + 1 retry). So, that means 8 attempts at testing the service before the notification is sent. If you want to "trigger" your notification on more or less than that number of failed attempts, you can simply alter your initial delay in the notification or change the polling package polling interval and number retries on each service in the package as well as downtime polling model to meet your requirements.

HTH,

David

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/

_______________________________________________
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