How to alert based on rate of disk consumption, not percentage free.

alert disk-free sensor-detection similar-sensor-analysis snmp trending

I am currently using threshold triggers to alert when disk free space drops below a certain percent. However, I would prefer to alert based on the rate of consumption. For example, a drive has been at 11% free for 6 months and has change to 10%, causing an alert. If the system stays true to form, I can expect the drive to sit at 10% for months. Under these circumstances the alert is effectively useless. Now, let's say the drive hit a warning alert at 25% free and 1 minute later hits an error at 10% free, with a system crash a minute later due to loss of space. Is there a way to alert based on the change of percent of bytes between scans? So if there is a 1% change in free space, nothing happens, but if there is a 15% change in space, for example, I can receive an alert.

Created on Dec 19, 2012 5:05:04 PM by Amy Dunker (1) ●1

Last change on Dec 20, 2012 9:05:21 AM by Daniel Zobel [Product Manager]

9 Replies

Accepted Answer

Votes:

0

Your Vote:

Hello,

I'm afraid this indeed cannot be properly addressed in the moment in PRTG with triggers. We do have this on the mind though! Please bear with us.

best regards.

Created on Dec 19, 2012 5:24:08 PM by Torsten Lindner [Paessler Support]

Votes:

0

Your Vote:

I'm resurrecting this request four years later because there was a suggestion that this capability might someday be implemented. Has there been any progress on this? I too need to monitor the rate of change of a sensor.

In my case, it's not disk space, but a custom sensor that monitors the number of items in a mailbox, which alerts us to spammers that have compromised user accounts from phishing attacks. Time is critical in these events, and we need to be alerted to a sudden spike in the number of items, without knowing what the baseline number is.

Created on Mar 3, 2017 2:55:00 PM by David Guertin (30) ●1

Votes:

0

Your Vote:

David, thank you very much for your KB-Post. I'm very much afraid this has not been implemented, because it is a quite complicated topic, and the requests have been very few for it.

Created on Mar 6, 2017 7:40:17 AM by Torsten Lindner [Paessler Support]

Votes:

0

Your Vote:

Thanks. I ended up accomplishing the same thing using a script being called by a PRTG custom sensor.

Dave

Created on Mar 6, 2017 2:06:21 PM by David Guertin (30) ●1

Votes:

1

Your Vote:

@David, can you please share the Script, we want such infos too :)

Created on Jan 26, 2018 9:58:35 AM by Johannes (10)

Votes:

0

Your Vote:

Any more progress on this? I need to monitor the disk consumption as well.

@david can you post your custom sensor details?

Created on Jun 12, 2020 3:11:21 PM by rrael (0)

Votes:

0

Your Vote:

Hello rrael,

a derivative (meaning rate of change) limit is currently not planned, because other features are in higher demand and this time.

Created on Jun 12, 2020 5:45:39 PM by Arne Seifert [Paessler Support]

Votes:

0

Your Vote:

We have a need for this also - as have had several instances where increased or verbose logging on various servers has gone unchecked, eventually resulting in urgent action being required out of hours. I suspect many users would like it but havent asked or are unaware of the possibility. I would say this request should be a top 10 for PRTG. Its a preventative measure rather than an ambulance at the bottom of a cliff.

Created on Jul 23, 2020 11:23:38 PM by Filet-o-fish (30) ●1

Votes:

0

Your Vote:

Hello,

thank you for your input.

Feature requests can now be made through this process. Please allow me to be frank: For the time being, we don't plan to implement a notification trigger operating of data derivatives. The reason ist that other features on the roadmap currently are in higher demand.

Created on Jul 24, 2020 12:38:22 PM by Arne Seifert [Paessler Support]

Last change on Jul 24, 2020 12:39:03 PM by Arne Seifert [Paessler Support]