We are wanting to setup some automated procedures for low disk monitoring on server but I haven't been able to figure out just how to accomplish the following.
on the first low disk alert kick off a procedure to run a disk cleanup.
if a second low disk alert is generated then create a ticket.
I currently have a policy setup to do both at the same time but can't figure out how to get a "counter" going so a ticket is only generated on a second alert
We do this today via a smart monitor that we developed. It has:
1. Dynamic Thresholds - It generates a threshold dynamically based on the disk size and its relationship to several "container" sizes. The bigger the disk, the smaller the threshold. We found that about 1 in 800 systems will need a custom override because the free space on a drive doesn't change much but hovers right around the calculated threshold.
2. Transient Suppression - the threshold must be exceeded by 85% on the first check or must exceed the threshold for 3 days before an alert is generated. Prevents alerts from situations where data is placed on a volume temporarily.
3. Self Remediation - our daily disk cleanup utility can be invoked directly by the Smart Monitor with an "aggressive" parameter that reduces the file "old age" value from 7 days to 2.
The Disk Capacity Smart Monitor also performs a 30-day trend capture and projects that usage out another 30 days each day, which can generate a Warning event a few weeks before the threshold is actually crossed.
We have Smart Monitors for Disk Capacity, Antivirus, SafeMode, Internet Fail-Over, and W32Time Config. These are executable code, deployed to agents and then invoked by procedures each day. Some checks run once and others, like the disk capacity, run for 24 hours and exit.
See www.mspbuilder.com for more info.