From 5f1278eb61501f031e63bd79e0c98b3f57f5cd34 Mon Sep 17 00:00:00 2001
From: ale <ale@incal.net>
Date: Wed, 24 Feb 2021 15:34:10 +0000
Subject: [PATCH] Replace the "target" field with "host" on prober metrics
---
.../templates/prometheus.yml.j2 | 22 +++++++++----------
.../templates/rules/alerts_base.conf.yml | 6 ++---
.../templates/rules/rules_base.conf.yml | 12 +++++-----
3 files changed, 19 insertions(+), 21 deletions(-)
diff --git a/roles/float-infra-prometheus/templates/prometheus.yml.j2 b/roles/float-infra-prometheus/templates/prometheus.yml.j2
index e2c21a4c..d1209843 100644
--- a/roles/float-infra-prometheus/templates/prometheus.yml.j2
+++ b/roles/float-infra-prometheus/templates/prometheus.yml.j2
@@ -108,7 +108,7 @@ scrape_configs:
- ping
relabel_configs:
- source_labels: [__address__]
- target_label: target
+ target_label: host
regex: "([^.]*).*"
replacement: "${1}"
- source_labels: [__address__]
@@ -117,7 +117,7 @@ scrape_configs:
target_label: instance
- target_label: __address__
replacement: {{ prober_host }}.prometheus.{{ domain }}:9115
- - target_label: host
+ - target_label: prober_host
replacement: {{ prober_host }}
static_configs:
- targets:
@@ -137,7 +137,7 @@ scrape_configs:
- http_base
relabel_configs:
- source_labels: [__address__]
- target_label: target
+ target_label: host
regex: "https?://([^.:/]*).*"
replacement: "${1}"
- source_labels: [__address__]
@@ -146,7 +146,7 @@ scrape_configs:
target_label: instance
- target_label: __address__
replacement: {{ prober_host }}.prometheus.{{ domain }}:9115
- - target_label: host
+ - target_label: prober_host
replacement: {{ prober_host }}
static_configs:
- targets:
@@ -165,7 +165,7 @@ scrape_configs:
module: [dns_toplevel]
relabel_configs:
- source_labels: [__address__]
- target_label: target
+ target_label: host
regex: "([^.]*).*"
replacement: "${1}"
- source_labels: [__address__]
@@ -174,7 +174,7 @@ scrape_configs:
target_label: instance
- target_label: __address__
replacement: {{ prober_host }}.prometheus.{{ domain }}:9115
- - target_label: host
+ - target_label: prober_host
replacement: {{ prober_host }}
static_configs:
- targets:
@@ -191,7 +191,7 @@ scrape_configs:
{# Additional blackbox probers #}
{% for p in prometheus_additional_blackbox_probers | default([]) | sort(attribute='name') %}
-{% for host in groups[p.service] | sort %}
+{% for prober_host in groups[p.service] | sort %}
- job_name: "prober_{{ p.name }}_{{ loop.index }}"
metrics_path: "/probe"
params:
@@ -200,16 +200,16 @@ scrape_configs:
relabel_configs:
- source_labels: [__address__]
target_label: target
- regex: "(?:https?://)?([^.:/]*).*"
+ regex: "{{ p.target_label_regex | default('([^:]*):.*') }}"
replacement: "${1}"
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
- replacement: {{ host }}.{{ p.service }}.{{ domain }}:{{ p.port }}
- - target_label: host
- replacement: {{ host }}
+ replacement: {{ prober_host }}.{{ p.service }}.{{ domain }}:{{ p.port }}
+ - target_label: prober_host
+ replacement: {{ prober_host }}
static_configs:
- targets:
{#
diff --git a/roles/float-infra-prometheus/templates/rules/alerts_base.conf.yml b/roles/float-infra-prometheus/templates/rules/alerts_base.conf.yml
index ab5bbb00..6e8a5541 100644
--- a/roles/float-infra-prometheus/templates/rules/alerts_base.conf.yml
+++ b/roles/float-infra-prometheus/templates/rules/alerts_base.conf.yml
@@ -72,9 +72,9 @@ groups:
severity: page
scope: host
annotations:
- summary: 'Probe {{ $labels.probe }}@{{ $labels.target }} is failing'
- description: 'Probe {{ $labels.probe }} ({{ $labels.zone }}) is failing
- for target {{ $labels.target }} (success ratio {{ $value }}).'
+ summary: 'Probe {{ $labels.probe }}@{{ $labels.host }} is failing'
+ description: 'Probe {{ $labels.probe }} ({{ $labels.host }}) is failing
+ for target {{ $labels.host }} (success ratio {{ $value }}).'
runbook: '[[ alert_runbook_fmt | format("ProbeFailure") ]]'
- alert: ProbeFailure
diff --git a/roles/float-infra-prometheus/templates/rules/rules_base.conf.yml b/roles/float-infra-prometheus/templates/rules/rules_base.conf.yml
index 1dd30827..456c1ecb 100644
--- a/roles/float-infra-prometheus/templates/rules/rules_base.conf.yml
+++ b/roles/float-infra-prometheus/templates/rules/rules_base.conf.yml
@@ -11,9 +11,9 @@ groups:
# Sum prober metrics over the probers (hosts), producing
# an aggregation by target.
- record: target:probe_success:count
- expr: count(probe_success) by (probe,probeset,zone,target)
+ expr: count(probe_success) by (probe,probeset,zone,host)
- record: target:probe_success:sum
- expr: sum(probe_success) by (probe,probeset,zone,target)
+ expr: sum(probe_success) by (probe,probeset,zone,host)
- record: target:probe_success:ratio
expr: target:probe_success:sum / target:probe_success:count
@@ -25,9 +25,7 @@ groups:
- record: probe:probe_success:ratio
expr: probe:probe_success:sum / probe:probe_success:count
- # Special metric for the ping probe. The label_replace() sets
- # the host to the value of the target label (instead of the host
- # running the prober). The 'bool' qualifier makes the greater-than
- # operation not act as a filter.
+ # Special metric for the ping probe.
+ # The 'bool' qualifier makes the greater-than operation not act as a filter.
- record: host_reachable
- expr: label_replace(target:probe_success:ratio{probe="ping"} > bool 0.6, "host", "$1", "target", "(.*)")
+ expr: target:probe_success:ratio{probe="ping"} > bool 0.6
--
GitLab