Set severity to NodeCPUHighUsage to info

Signed-off-by: Vitaly Zhuravlev <v-zhuravlev@users.noreply.github.com>
pull/2644/head
Vitaly Zhuravlev 2 years ago
parent 6bdc1d9c98
commit b7dfb32bfc

@ -312,15 +312,17 @@
{ {
alert: 'NodeCPUHighUsage', alert: 'NodeCPUHighUsage',
expr: ||| expr: |||
sum without(mode) (avg without (cpu) (rate(node_cpu_seconds_total{%(nodeExporterSelector)s, mode!="idle"}[2m]))) > 0.8 sum without(mode) (avg without (cpu) (rate(node_cpu_seconds_total{%(nodeExporterSelector)s, mode!="idle"}[2m]))) * 100 > %(cpuHighUsageThreshold)d
||| % $._config, ||| % $._config,
'for': '15m', 'for': '15m',
labels: { labels: {
severity: 'warning', severity: 'info',
}, },
annotations: { annotations: {
summary: 'High CPU usage.', summary: 'High CPU usage.',
description: 'CPU usage at {{ $labels.instance }} has been above 80% for the last 15 minutes, is currently at {{ printf "%.2f" $value }}%.', description: |||
CPU usage at {{ $labels.instance }} has been above %(cpuHighUsageThreshold)d%% for the last 15 minutes, is currently at {{ printf "%%.2f" $value }}%%.
||| % $._config,
}, },
}, },
{ {
@ -336,7 +338,7 @@
annotations: { annotations: {
summary: 'System saturated, load per core is very high.', summary: 'System saturated, load per core is very high.',
description: ||| description: |||
System load per core at {{ $labels.instance }} has been above %(systemSaturationPerCoreThreshold)d for the last 15 minutes, is currently at {{ printf "%.2f" $value }}. System load per core at {{ $labels.instance }} has been above %(systemSaturationPerCoreThreshold)d for the last 15 minutes, is currently at {{ printf "%%.2f" $value }}.
This might indicate this instance resources saturation and can cause it becoming unresponsive. This might indicate this instance resources saturation and can cause it becoming unresponsive.
||| % $._config, ||| % $._config,
}, },
@ -353,7 +355,7 @@
annotations: { annotations: {
summary: 'Memory major page faults are occurring at very high rate.', summary: 'Memory major page faults are occurring at very high rate.',
description: ||| description: |||
Memory major pages are occurring at very high rate at {{ $labels.instance }}, %(memoryMajorPagesFaultsThreshold)d major page faults per second for the last 15 minutes, is currently at {{ printf "%.2f" $value }}. Memory major pages are occurring at very high rate at {{ $labels.instance }}, %(memoryMajorPagesFaultsThreshold)d major page faults per second for the last 15 minutes, is currently at {{ printf "%%.2f" $value }}.
Please check that there is enough memory available at this instance. Please check that there is enough memory available at this instance.
||| % $._config, ||| % $._config,
}, },
@ -370,7 +372,7 @@
annotations: { annotations: {
summary: 'Host is running out of memory.', summary: 'Host is running out of memory.',
description: ||| description: |||
Memory is filling up at {{ $labels.instance }}, has been above %(memoryHighUtilizationThreshold)d%% for the last 15 minutes, is currently at {{ printf "%.2f" $value }}%. Memory is filling up at {{ $labels.instance }}, has been above %(memoryHighUtilizationThreshold)d%% for the last 15 minutes, is currently at {{ printf "%%.2f" $value }}%%.
||| % $._config, ||| % $._config,
}, },
}, },
@ -386,7 +388,7 @@
annotations: { annotations: {
summary: 'Disk IO queue is high.', summary: 'Disk IO queue is high.',
description: ||| description: |||
Disk IO queue (aqu-sq) is high on {{ $labels.device }} at {{ $labels.instance }}, has been above %(diskIOSaturationThreshold)d for the last 15 minutes, is currently at {{ printf "%.2f" $value }}. Disk IO queue (aqu-sq) is high on {{ $labels.device }} at {{ $labels.instance }}, has been above %(diskIOSaturationThreshold)d for the last 15 minutes, is currently at {{ printf "%%.2f" $value }}.
This symptom might indicate disk saturation. This symptom might indicate disk saturation.
||| % $._config, ||| % $._config,
}, },

@ -43,7 +43,9 @@
// just a warning for K8s nodes. // just a warning for K8s nodes.
nodeCriticalSeverity: 'critical', nodeCriticalSeverity: 'critical',
// CPU utilization (%) on which to trigger the
// 'NodeCPUHighUsage' alert.
cpuHighUsageThreshold: 90,
// Load average 1m (per core) on which to trigger the // Load average 1m (per core) on which to trigger the
// 'NodeSystemSaturation' alert. // 'NodeSystemSaturation' alert.
systemSaturationPerCoreThreshold: 2, systemSaturationPerCoreThreshold: 2,

Loading…
Cancel
Save