Custom alert text help

Question

Hello,

I have a custom alert that triggers when fixed disk space changes by 10% over a 2 hour period. The alert works great! The only problem is that we cannot identify if the change was because 10% disk space was consumed or made available. + or -

Trigger:

LEFT JOIN VolumeUsage_Detail V

ON Volumes.VolumeID = V.VolumeID

AND Volumes.VolumeType = 'fixed disk'

AND DATEDIFF(HH,DateTime, GETDATE()) < 2

GROUP BY Volumes.NodeID, Volumes.VolumeID, Volumes.FullName

HAVING MAX(PercentDiskUsed) - MIN(PercentDiskUsed) > 10

Alert message body:

An issue on an object you are monitoring occurred at ${N=Alerting;M=AlertTriggerTime;F=DateTime}.

Server: ${NodeName}

IP: ${Node.IP_Address}

Volume: ${SQL:Select Substring ('${Caption}',1,3) AS Caption}

Volume Size: ${SQL:Select CAST( ROUND (${VolumeSize}/1048576, 0) AS Decimal(18,0)) AS Gigabytes} MB

Space Available: ${SQL:Select CAST( ROUND (${VolumeSpaceAvailable}/1048576, 0) AS Decimal(18,0)) AS Gigabytes} MB

Percent Available: ${N=SwisEntity;M=VolumePercentAvailable}

Volume Graph: ${VolumeDetailsURL}

This message was brought to you by the alert named: ${N=Alerting;M=AlertName}

Actual alert message:

An issue on an object you are monitoring occurred at Wednesday, February 08, 2017 8:30 PM.

Server: Server1
 IP: 192.168.1.10
 Volume: F:\
 Volume Size: 307197 MB
 Space Available: 112760 MB
 Percent Available: 37 %
 Volume Graph: SWServer:443/.../View.aspx

This message was brought to you by the alert named: Disk change 10% in 2 Hrs

I need either two triggers or something in the alert message body that indicates if it was an + or -.

Thanks in advance!

jvb · Answer

I did some more looking and I wasn't able to find a way to do that directly in the alert logic but

I was able to put together a very basic script that I then added to a template:

Volume Space Diff / Compare

It is really just a proof of concept and could probably be improved but it hits the basic points.

Poll every 15 minutes and then compare space used now against the last poll. (Could easily be modified for your slow creep use case too)

Have a look and see if that helps.

mnostrom · Answer

The exact request from our vendor was alerts for:

* Rapid increase detection* Sampling Frequency: 15 minutes
* Threshold: 100 MB increase since last sample

* Slow creep detection* Sampling Frequency: 4 hours
* Threshold: 1 GB increase since last sample

Not sure if this is even possible.

jvb · Answer

My apologies... You are correct, that variable is not going to be useful.

I had misunderstood what you were trying to do. I see what you mean now.

I will try to tinker with this in my lab a little later and see if I can find something.

The trick is going to be determining what to compare against... We know within a 2

hour time frame what the minimum and maximum amount of percent available was.

Now we have to find something to compare against to say if that is net loss or net gain....