Start a Conversation

Unsolved

This post is more than 5 years old

35321

March 7th, 2014 13:00

Change Alert Threshold

Hey Everyone,

I have Dell OpenManage Essentials 1.3 (just upgraded).

How do you mdify the alert threshold for an alert? Example:

You have an alert action defined that monitors for critical events such as "System is Down". I get this alert way to often and its a false positive. Perhaps the server is very briefly not reachable (maybe 10 seconds), but I only want this alert if the server is unreachable for more than 30 or 45 seconds.

Thanks,

Nick.


Enterprise OID: .1.3.6.1.4.1.674.11000.1000.100.1 Specific Trap OID: 1001 Generic Trap OID: 6 Alert Category Name: System Events Alert Source Name: omeAlertSystemDown Package Name: MIB-Dell-OME

March 7th, 2014 13:00

Hey Rob,

My settings are as follows:

Enable OnDemand Poll - Checked
Enable Status Polling - Checked
Schedule - 0 Days, 1 Hours, 30 Minutes
Status Polling Speed - Set to the middle, inbetween slow and fast.

The health state change alert was enabled, but I have now disabled this.

The ignore alert isn't exactly a threshold setting. As far as I understand it, this will prevent the same alert being sent out multiple times, but this isn't going to change the threshold.

I think it is basic functionality to be able to define that I only want to receive an alert if the condition persists for x amount of time. This logic must at least be built in and predefined. I am very surprised I can't find a setting as almost all alerting software (SolarWinds, WhatsUp Gold) allows this to be defined.

Thanks,

Nick.

2.8K Posts

March 7th, 2014 13:00

Thanks Nick,

Just to check...how often were you getting that alert in the console?  If your status poll was set to 90 minutes, was it showing up more frequent than that?  As you noticed, the status poll frequency is configurable.

But I think what you want is to be able to specify how long the condition exists before the decision is made to trigger an alert right?

Just making sure I understand so I can fwd the feedback.

Thx

Rob

2.8K Posts

March 7th, 2014 13:00

Hi there Nick...

Couple of quick questions:

1. What is your status polling time set to?  (default is 60min) setting is on the Discovery Portal page.

2. Is the health state change alert checkbox checked?  (seems like yes, can't recall if it is default) in Prefs.

You actually don't have to have this checked if your managed devices are set up to send SNMP traps to the OME console.  This is kind of an extra indicator for folks that don't use SNMP traps.

There isn't really a threshhold settings, but some folks will create a new ignore filter that let's you specify a time.

thx

Rob

March 7th, 2014 14:00

Hey Rob,

Your wording is correct. I want to be able to specify how long the condition exists before the decision is made to trigger an alert. Some alerts are subject to false positive, because they have variable values that change over time, and others are a one time event.

Here are some examples:

If a server looses network connectivity for 2 seconds, I don't want to know about that.

If a server looses network connectivity for 30 seconds, I want to know about that.

If a server gets too hot for 5 seconds, I don't want to know about that.

If a server gets too hot for 3 minutes, I want to know about that.


Other alerts would not be subject to the requirement of a threshold. For example, if a server lost power on a power supply, or someone opened a chassis, I would just want those ones to send their alert out right away every time.

Thanks for your help Rob,

Nick.

2.8K Posts

March 7th, 2014 14:00

got it...will send it across to the right folks for consideration.

thx

Rob

No Events found!

Top