Forum Discussion

Kelemvor's avatar
Kelemvor
Icon for Expert rankExpert
2 months ago

How do you handle Disk Space alerting?

Hi,

The LM standard is to do disk space alerting based on the percentage of space used.  E.g. 90% warning, 95% error, 98% critical (or whatever).  This is great for machines with an average size drive, but is completely useless for machine with giant drives.  We have some machines that might have drives that are multiple Terabytes in size.  Getting an alert when a 2TB drive is 90% full doesn't help because it still has 200 Gigs free.

For all our Windows servers, we've added additional alerting based on a hard coded 10 Gigs free on the C drive because Windows Updates generally have issues if you have less than that.  This has helped a bunch for smaller drives where 10 Gigs free isn't small enough to hit the Percent-based alerting because maybe it's only a 60 Gig C drive.

I'm just wondering what everyone else does for these types of alerts.  Do you use the percentage based alerts for most machines but then create new ones, or change the percentage for large or small servers?  Do you change everything to a hard size limit?  Some other combination?

Just looking for ideas so we can try to reduce the unnecessary alerts for servers with huge drives and get alerts for ones with tiny drives.

Thanks.

alerting
  • Mike_Moniz's avatar
    Mike_Moniz
    Icon for Professor rankProfessor
    2 months ago

    In the past I've pretty much done it both ways. In general I try to stick with the built-in percent options, especially as MSPs with a wide range of environments. I've sometimes even used custom thresholds like 99.63768%. I've also written custom DataSources in the past. I don't recall the specifics, I think one of them may have even switched between using % and free space depending on the size of the drive (I don't have access to it anymore).