From b485aab7e71c1625cfc27e0f92c9509f42378458 Mon Sep 17 00:00:00 2001
From: Daniel Baumann <daniel.baumann@progress-linux.org>
Date: Sun, 5 May 2024 13:19:16 +0200
Subject: Adding upstream version 1.45.3+dfsg.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
---
 src/health/health.d/load.conf | 67 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 67 insertions(+)
 create mode 100644 src/health/health.d/load.conf

(limited to 'src/health/health.d/load.conf')

diff --git a/src/health/health.d/load.conf b/src/health/health.d/load.conf
new file mode 100644
index 000000000..7b0e18b84
--- /dev/null
+++ b/src/health/health.d/load.conf
@@ -0,0 +1,67 @@
+# you can disable an alarm notification by setting the 'to' line to: silent
+
+# Calculate the base trigger point for the load average alarms.
+# This is the maximum number of CPU's in the system over the past 1
+# minute, with a special case for a single CPU of setting the trigger at 2.
+      alarm: load_cpu_number
+         on: system.load
+      class: Utilization
+       type: System
+  component: Load
+host labels: _os=linux
+       calc: ($active_processors == nan or $active_processors == 0) ? (nan) : ( ($active_processors < 2) ? ( 2 ) : ( $active_processors ) )
+      units: cpus
+      every: 1m
+       info: Number of active CPU cores in the system
+
+# Send alarms if the load average is unusually high.
+# These intentionally _do not_ calculate the average over the sampled
+# time period because the values being checked already are averages.
+
+      alarm: load_average_15
+         on: system.load
+      class: Utilization
+       type: System
+  component: Load
+host labels: _os=linux
+     lookup: max -1m unaligned of load15
+       calc: ($load_cpu_number == nan) ? (nan) : ($this)
+      units: load
+      every: 1m
+       warn: ($this * 100 / $load_cpu_number) > (($status >= $WARNING) ? 175 : 200)
+      delay: down 15m multiplier 1.5 max 1h
+    summary: Host load average (15 minutes)
+       info: System load average for the past 15 minutes
+         to: silent
+
+      alarm: load_average_5
+         on: system.load
+      class: Utilization
+       type: System
+  component: Load
+host labels: _os=linux
+     lookup: max -1m unaligned of load5
+       calc: ($load_cpu_number == nan) ? (nan) : ($this)
+      units: load
+      every: 1m
+       warn: ($this * 100 / $load_cpu_number) > (($status >= $WARNING) ? 350 : 400)
+      delay: down 15m multiplier 1.5 max 1h
+    summary: System load average (5 minutes)
+       info: System load average for the past 5 minutes
+         to: silent
+
+      alarm: load_average_1
+         on: system.load
+      class: Utilization
+       type: System
+  component: Load
+host labels: _os=linux
+     lookup: max -1m unaligned of load1
+       calc: ($load_cpu_number == nan) ? (nan) : ($this)
+      units: load
+      every: 1m
+       warn: ($this * 100 / $load_cpu_number) > (($status >= $WARNING) ? 700 : 800)
+      delay: down 15m multiplier 1.5 max 1h
+    summary: System load average (1 minute)
+       info: System load average for the past 1 minute
+         to: silent
-- 
cgit v1.2.3