Host CPU is underutilized info
>>>
(min by (instance) (
rate
(
node_cpu_seconds_total
{mode="idle"}[1h]))) > 0.8
The query computes, for each instance, the smallest idle‑CPU rate among all its CPU cores over the past hour (min by (instance) (rate(node_cpu_seconds_total{mode="idle"}[1h]))). This value represents the fraction of time a core spends idle. The alert fires when that minimum idle fraction is greater than 0.8 – i.e., even the most‑busy core is idle more than 80 % of the time, meaning the overall CPU usage is below 20 % for a sustained period (e.g., a week).
Exporters Used
Related Alerts
Host CPU high iowait
Host CPU steal noisy neighbor
Host EDAC Correctable Errors detected
Host EDAC Uncorrectable Errors detected
Host Memory is underutilized
Host Network Bond Degraded
Host Network Receive Errors
Host Network Transmit Errors
Host OOM kill detected
Host clock not synchronising
Host clock skew
Host conntrack limit
Host context switching high
Host disk may fill in 24 hours
Host filesystem device error
Host high CPU load
Host inodes may fill in 24 hours
Host kernel version deviations
Host memory under memory pressure
Host node overtemperature alarm
Host out of disk space
Host out of inodes
Host out of memory
Host physical component too hot
Host software RAID disk failure
Host software RAID insufficient drives
Host swap is filling up
Host systemd service crashed
Host unusual disk IO
Host unusual disk read latency
Host unusual disk read rate
Host unusual disk write latency
Host unusual network throughput in
Host unusual network throughput out
NodeClockNotSynchronising
NodeClockSkewDetected
NodeFilesystemAlmostOutOfFiles
NodeFilesystemAlmostOutOfFiles
NodeFilesystemAlmostOutOfSpace
NodeFilesystemAlmostOutOfSpace
NodeFilesystemFilesFillingUp
NodeFilesystemFilesFillingUp
NodeFilesystemSpaceFillingUp
NodeFilesystemSpaceFillingUp
NodeHighNumberConntrackEntriesUsed
NodeNetworkReceiveErrs
NodeNetworkTransmitErrs
NodeRAIDDegraded
NodeRAIDDiskFailure
NodeTextFileCollectorScrapeError
ZFS offline pool