NPM: System response time

Hi,

I am havinga question about the percent loss alert in NPM.

I want to create an alert that will be triggered when there is more than 40% loss and the alert reset must be when the packetloss is less than 80 %.

The idea behind this is, we want to receive an alert a soon as the system is offline and we want to receive an alert when the system starts to respond again.

The only problem is when you set the alert that I described you create a loop, a because you have got the alert time between 40% and the 80%.

I was thinking to use something like the “respons time“ option in the alert. But whatis a normal respons time for a system? Because a system in your LAN will give afaster response then a system in china.

Can someonehelp me with my problem.

This iswhat I have got for so far.

Triggercondition
Percent loss is greater than or equal to 40

Resetcondition
Response time is less than or equal to ?????

Already thank !

J-P

Find more posts tagged with

Alert

advanced_alert_setup_question

alert_condition

Alerts

npm_9.5.1

Accepted answers

All comments

mcbridea

It sounds like you are using response time to determine if a system is up. Keep in mind that response time is checked every 2 minutes by default so this won't work for real-time down alerts.

Questionario

if you really want to do it that way you could set it for >1000, that way it won't alarm on just high latency...

but try if it even works this way... I have never tried that and wouldnt know what happens if the response time is null or something like that...

if that doesnt work you could try to negate it to something like "where not all of the following apply" and as only condition in that group have "is smaller than or equal to" 1000

but whats the problem with using the regular "down"?

Comic_nl

The problem is a s follow, we want to receive an alert as soon as the system misses 2 pings en when a system is back online we want to receive immediately an warning. This because the systems that we are monitoring a crucial for a company.

And with the normal alerts you receive within let’s say a minute that a system is offline, but the trap for that the system is back-online can take up to 4 minutes

already thanks for the reply's

mcbridea

When pings are missed, NPM will fast ping the device and look for it to come up. The up notification will be in the order of seconds, where as the down notification will be determined by response time polling settings and the point in the cycle when the device goes down. Add a reset condition action to notify you and you should be good to go.