From 58e290d461557309d051c2268707dd60701572aa Mon Sep 17 00:00:00 2001
From: Monty Dawson <wgd23@cam.ac.uk>
Date: Tue, 15 Jun 2021 14:01:08 +0100
Subject: [PATCH] Surface the alerting percentage variable.
---
main.tf | 7 ++++---
variables.tf | 13 +++++++++++++
2 files changed, 17 insertions(+), 3 deletions(-)
diff --git a/main.tf b/main.tf
index bcba828..4c82a33 100644
--- a/main.tf
+++ b/main.tf
@@ -248,9 +248,10 @@ module "uptime_monitoring" {
uptime_check = {
# Accept either e.g. "60s" or 60 for timeout and periods for compatibility
# with previous releases.
- timeout = tonumber(trimsuffix(var.alerting_uptime_timeout, "s"))
- period = tonumber(trimsuffix(var.alerting_uptime_period, "s"))
- path = var.monitoring_path
+ timeout = tonumber(trimsuffix(var.alerting_uptime_timeout, "s"))
+ period = tonumber(trimsuffix(var.alerting_uptime_period, "s"))
+ path = var.monitoring_path
+ success_threshold_percent = var.alerting_success_threshold_percent
alert_enabled = var.alerting_enabled
}
diff --git a/variables.tf b/variables.tf
index 0541a9d..22a4191 100644
--- a/variables.tf
+++ b/variables.tf
@@ -122,6 +122,19 @@ variable "alerting_uptime_period" {
description = "Frequency of uptime checks"
}
+variable "alerting_success_threshold_percent" {
+ type = number
+ default = 75
+ description = <<EOT
+If the percentage of successful uptime checks within the given uptime period falls below
+this, an alert will be triggered. Set to 100 to trigger an alert if any uptime check fails,
+set to a lower number to tolerate failures without alerting.
+
+Experience has taught us that uptime checks can fail semi-regularly due to transient
+problems outside our control, therefore we allow some leeway before triggering an alert.
+EOT
+}
+
variable "alerting_enabled" {
type = bool
default = true
--
GitLab