2019-04-23 17:35:12 +02:00

108 lines
4.0 KiB
HCL

### Elasticsearch cluster status monitor ###
/* Note about the query
- If aws.es.cluster_statusred is 1 --> query value (= 2.1) > 2 : critical
- If aws.es.cluster_statusyellow is 1 --> 1 < query value (=1.1) < 2 : warning
Workaround : in the query, we add "0.1" to the result and we use the comparator ">=". No alert was triggered without that. */
resource "datadog_monitor" "es_cluster_status" {
count = "${var.es_cluster_status_enabled == "true" ? 1 : 0}"
name = "[${var.environment}] ElasticSearch cluster status is not green"
message = "${coalesce(var.es_cluster_status_message, var.message)}"
type = "metric alert"
query = <<EOF
max(${var.es_cluster_status_timeframe}): (
avg:aws.es.cluster_statusred${module.filter-tags.query_alert} by {region,name} * 2 +
(avg:aws.es.cluster_statusyellow${module.filter-tags.query_alert} by {region,name} + 0.1)
) >= 2
EOF
thresholds {
warning = 1
critical = 2
}
notify_no_data = true
evaluation_delay = "${var.evaluation_delay}"
renotify_interval = 0
notify_audit = false
timeout_h = 0
include_tags = true
locked = false
require_full_window = false
new_host_delay = "${var.new_host_delay}"
silenced = "${var.es_cluster_status_silenced}"
tags = ["env:${var.environment}", "type:cloud", "provider:aws", "resource:elasticsearch", "team:claranet", "created-by:terraform", "${var.es_cluster_status_extra_tags}"]
}
### Elasticsearch cluster free storage space monitor ###
resource "datadog_monitor" "es_free_space_low" {
count = "${var.diskspace_enabled == "true" ? 1 : 0}"
name = "[${var.environment}] ElasticSearch cluster free storage space {{#is_alert}}{{{comparator}}} {{threshold}}% ({{value}}%){{/is_alert}}{{#is_warning}}{{{comparator}}} {{warn_threshold}}% ({{value}}%){{/is_warning}}"
message = "${coalesce(var.diskspace_message, var.message)}"
type = "metric alert"
query = <<EOF
${var.diskspace_time_aggregator}(${var.diskspace_timeframe}): (
avg:aws.es.free_storage_space${module.filter-tags.query_alert} by {region,name} /
(${var.es_cluster_volume_size}*1000) * 100
) < ${var.diskspace_threshold_critical}
EOF
thresholds {
warning = "${var.diskspace_threshold_warning}"
critical = "${var.diskspace_threshold_critical}"
}
notify_no_data = true
evaluation_delay = "${var.evaluation_delay}"
renotify_interval = 0
notify_audit = false
timeout_h = 0
include_tags = true
locked = false
require_full_window = false
new_host_delay = "${var.new_host_delay}"
silenced = "${var.diskspace_silenced}"
tags = ["env:${var.environment}", "type:cloud", "provider:aws", "resource:elasticsearch", "team:claranet", "created-by:terraform", "${var.diskspace_extra_tags}"]
}
### Elasticsearch cluster CPU monitor ###
resource "datadog_monitor" "es_cpu_90_15min" {
count = "${var.cpu_enabled == "true" ? 1 : 0}"
name = "[${var.environment}] ElasticSearch cluster CPU high {{#is_alert}}{{{comparator}}} {{threshold}}% ({{value}}%){{/is_alert}}{{#is_warning}}{{{comparator}}} {{warn_threshold}}% ({{value}}%){{/is_warning}}"
message = "${coalesce(var.cpu_message, var.message)}"
type = "metric alert"
query = <<EOF
${var.cpu_time_aggregator}(${var.cpu_timeframe}): (
avg:aws.es.cpuutilization${module.filter-tags.query_alert} by {region,name}
) > ${var.cpu_threshold_critical}
EOF
thresholds {
warning = "${var.cpu_threshold_warning}"
critical = "${var.cpu_threshold_critical}"
}
notify_no_data = true
evaluation_delay = "${var.evaluation_delay}"
renotify_interval = 0
notify_audit = false
timeout_h = 0
include_tags = true
locked = false
require_full_window = false
new_host_delay = "${var.new_host_delay}"
silenced = "${var.cpu_silenced}"
tags = ["env:${var.environment}", "type:cloud", "provider:aws", "resource:elasticsearch", "team:claranet", "created-by:terraform", "${var.cpu_extra_tags}"]
}