summaryrefslogtreecommitdiffstats
path: root/health/health.d/ml.conf
diff options
context:
space:
mode:
Diffstat (limited to 'health/health.d/ml.conf')
-rw-r--r--health/health.d/ml.conf53
1 files changed, 53 insertions, 0 deletions
diff --git a/health/health.d/ml.conf b/health/health.d/ml.conf
new file mode 100644
index 0000000..6836ce7
--- /dev/null
+++ b/health/health.d/ml.conf
@@ -0,0 +1,53 @@
+# below are some examples of using the `anomaly-bit` option to define alerts based on anomaly
+# rates as opposed to raw metric values. You can read more about the anomaly-bit and Netdata's
+# native anomaly detection here:
+# https://learn.netdata.cloud/docs/agent/ml#anomaly-bit---100--anomalous-0--normal
+
+# examples below are commented, you would need to uncomment and adjust as desired to enable them.
+
+# node level anomaly rate example
+# https://learn.netdata.cloud/docs/agent/ml#node-anomaly-rate
+# if node level anomaly rate is between 1-5% then warning (pick your own threshold that works best via tial and error).
+# if node level anomaly rate is above 5% then critical (pick your own threshold that works best via tial and error).
+# template: ml_1min_node_ar
+# on: anomaly_detection.anomaly_rate
+# os: linux
+# hosts: *
+# lookup: average -1m foreach anomaly_rate
+# calc: $this
+# units: %
+# every: 30s
+# warn: $this > (($status >= $WARNING) ? (1) : (5))
+# crit: $this > (($status == $CRITICAL) ? (5) : (100))
+# info: rolling 1min node level anomaly rate
+
+# alert per dimension example
+# if anomaly rate is between 5-20% then warning (pick your own threshold that works best via tial and error).
+# if anomaly rate is above 20% then critical (pick your own threshold that works best via tial and error).
+# template: ml_5min_cpu_dims
+# on: system.cpu
+# os: linux
+# hosts: *
+# lookup: average -5m anomaly-bit foreach *
+# calc: $this
+# units: %
+# every: 30s
+# warn: $this > (($status >= $WARNING) ? (5) : (20))
+# crit: $this > (($status == $CRITICAL) ? (20) : (100))
+# info: rolling 5min anomaly rate for each system.cpu dimension
+
+# alert per chart example
+# if anomaly rate is between 5-20% then warning (pick your own threshold that works best via tial and error).
+# if anomaly rate is above 20% then critical (pick your own threshold that works best via tial and error).
+# template: ml_5min_cpu_chart
+# on: system.cpu
+# os: linux
+# hosts: *
+# lookup: average -5m anomaly-bit of *
+# calc: $this
+# units: %
+# every: 30s
+# warn: $this > (($status >= $WARNING) ? (5) : (20))
+# crit: $this > (($status == $CRITICAL) ? (20) : (100))
+# info: rolling 5min anomaly rate for system.cpu chart
+