feat: webarm: normalize service alerts
This commit is contained in:
@@ -1,58 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "amzebs-mysql-service-down-alert-uid";
|
||||
title = "MySQL Service Down on amzebs-01";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"mysql.service\", instance=\"amzebs-01:9100\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "MySQL service is down on amzebs-01";
|
||||
summary = "MySQL Service Down on amzebs-01";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
host = "amzebs-01";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,58 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "amzebs-nginx-service-down-alert-uid";
|
||||
title = "Nginx Service Down on amzebs-01";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"nginx.service\", instance=\"amzebs-01:9100\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "Nginx service is down on amzebs-01";
|
||||
summary = "Nginx Service Down on amzebs-01";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
host = "amzebs-01";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,58 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "amzebs-phpfpm-service-down-alert-uid";
|
||||
title = "PHP-FPM Service Down on amzebs-01";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=~\"phpfpm-.*\\\\.service\", instance=\"amzebs-01:9100\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "min";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "One or more PHP-FPM services are down on amzebs-01";
|
||||
summary = "PHP-FPM Service Down on amzebs-01";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
host = "amzebs-01";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,26 +1,6 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
let
|
||||
giteaDownAlertRules = (import ./gitea_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
giteaRunnerDownAlertRules = (import ./gitea_runner_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
postfixDownAlertRules = (import ./postfix_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
dovecotDownAlertRules = (import ./dovecot_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
openldapDownAlertRules = (import ./openldap_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
wireguardDownAlertRules = (import ./wireguard_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
|
||||
# amzebs-01 service alerts
|
||||
ambebsMysqlDownAlertRules = (import ./amzebs_mysql_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
ambebsNginxDownAlertRules = (import ./amzebs_nginx_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
ambebsPhpfpmDownAlertRules = (import ./amzebs_phpfpm_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
|
||||
allServiceRules = giteaDownAlertRules
|
||||
++ giteaRunnerDownAlertRules
|
||||
++ postfixDownAlertRules
|
||||
++ dovecotDownAlertRules
|
||||
++ openldapDownAlertRules
|
||||
++ wireguardDownAlertRules
|
||||
++ ambebsMysqlDownAlertRules
|
||||
++ ambebsNginxDownAlertRules
|
||||
++ ambebsPhpfpmDownAlertRules;
|
||||
servicesDownAlertRules = (import ./services_down.nix { inherit lib pkgs config; }).grafanaAlertRuleDefinitions;
|
||||
in
|
||||
{
|
||||
services.grafana.provision.alerting.rules.settings.groups = [
|
||||
@@ -28,7 +8,7 @@ in
|
||||
name = "Service Alerts";
|
||||
folder = "Service Monitoring";
|
||||
interval = "1m";
|
||||
rules = allServiceRules;
|
||||
rules = servicesDownAlertRules;
|
||||
}
|
||||
];
|
||||
}
|
||||
}
|
||||
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "dovecot-service-down-alert-uid";
|
||||
title = "Dovecot Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"dovecot.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "Dovecot service is down on {{ $labels.instance }}";
|
||||
summary = "Dovecot Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "gitea-service-down-alert-uid";
|
||||
title = "Gitea Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"container@git.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "Gitea service is down on {{ $labels.instance }}";
|
||||
summary = "Gitea Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "gitea-runner-service-down-alert-uid";
|
||||
title = "Gitea Runner Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"microvm@git-runner-1.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "Gitea Runner service is down on {{ $labels.instance }}";
|
||||
summary = "Gitea Runner Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "openldap-service-down-alert-uid";
|
||||
title = "OpenLDAP Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"openldap.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "OpenLDAP service is down on {{ $labels.instance }}";
|
||||
summary = "OpenLDAP Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "postfix-service-down-alert-uid";
|
||||
title = "Postfix Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"postfix.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "Postfix service is down on {{ $labels.instance }}";
|
||||
summary = "Postfix Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
@@ -0,0 +1,90 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
let
|
||||
# Add services here - each entry generates an alert rule
|
||||
# instance = which node exporter to query (hostname:9100)
|
||||
monitoredServices = [
|
||||
{ name = "AI-Mailer"; service = "ai-mailer.service"; instance = "fw:9100"; }
|
||||
{ name = "Postfix"; service = "postfix.service"; instance = "mail:9100"; }
|
||||
{ name = "Dovecot"; service = "dovecot.service"; instance = "mail:9100"; }
|
||||
{ name = "OpenLDAP"; service = "openldap.service"; instance = "mail:9100"; }
|
||||
{ name = "Gitea"; service = "container@git.service"; instance = "fw:9100"; }
|
||||
{ name = "Gitea Runner"; service = "microvm@git-runner-1.service"; instance = "fw:9100"; }
|
||||
{ name = "WireGuard"; service = "wireguard-wg_cloonar.service"; instance = "mail:9100"; }
|
||||
{ name = "MySQL"; service = "mysql.service"; instance = "amzebs-01:9100"; }
|
||||
{ name = "Nginx"; service = "nginx.service"; instance = "amzebs-01:9100"; }
|
||||
{ name = "PHP-FPM"; service = "phpfpm-.*\\.service"; instance = "amzebs-01:9100"; }
|
||||
];
|
||||
|
||||
# Extract host from instance (e.g., "fw:9100" -> "fw")
|
||||
getHost = instance: lib.head (lib.splitString ":" instance);
|
||||
|
||||
# Generate a unique UID from service name
|
||||
mkUid = name: "${lib.toLower (lib.replaceStrings [" " "@" "."] ["-" "-" "-"] name)}-down-uid";
|
||||
|
||||
# Check if service pattern uses regex (contains special chars)
|
||||
isRegex = svc: lib.hasInfix ".*" svc || lib.hasInfix "\\" svc;
|
||||
|
||||
# Build the PromQL expression
|
||||
mkExpr = svc:
|
||||
let
|
||||
nameMatch = if isRegex svc.service
|
||||
then "name=~\"${svc.service}\""
|
||||
else "name=\"${svc.service}\"";
|
||||
in "node_systemd_unit_state{state=\"active\", ${nameMatch}, instance=\"${svc.instance}\"} OR on() vector(0)";
|
||||
|
||||
mkServiceAlert = svc: {
|
||||
uid = mkUid svc.name;
|
||||
title = "${svc.name} Service Down on ${getHost svc.instance}";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = mkExpr svc;
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "${svc.name} service is down on ${getHost svc.instance}";
|
||||
summary = "${svc.name} Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
host = getHost svc.instance;
|
||||
};
|
||||
};
|
||||
in {
|
||||
grafanaAlertRuleDefinitions = map mkServiceAlert monitoredServices;
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
{ lib, pkgs, config, ... }:
|
||||
{
|
||||
grafanaAlertRuleDefinitions = [
|
||||
{
|
||||
uid = "wireguard-service-down-alert-uid";
|
||||
title = "WireGuard Service Down";
|
||||
condition = "C";
|
||||
data = [
|
||||
{
|
||||
refId = "A";
|
||||
relativeTimeRange = {
|
||||
from = 300;
|
||||
to = 0;
|
||||
};
|
||||
datasourceUid = "vm-datasource-uid";
|
||||
model = {
|
||||
editorMode = "code";
|
||||
expr = "node_systemd_unit_state{state=\"active\", name=\"wireguard-wg_cloonar.service\"} OR on() vector(0)";
|
||||
hide = false;
|
||||
intervalMs = 1000;
|
||||
legendFormat = "__auto";
|
||||
maxDataPoints = 43200;
|
||||
range = true;
|
||||
refId = "A";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "B";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "reduce";
|
||||
expression = "A";
|
||||
reducer = "last";
|
||||
};
|
||||
}
|
||||
{
|
||||
refId = "C";
|
||||
datasourceUid = "__expr__";
|
||||
model = {
|
||||
type = "math";
|
||||
expression = "$B < 1";
|
||||
};
|
||||
}
|
||||
];
|
||||
noDataState = "Alerting";
|
||||
execErrState = "Alerting";
|
||||
for = "5m";
|
||||
annotations = {
|
||||
description = "WireGuard service is down on {{ $labels.instance }}";
|
||||
summary = "WireGuard Service Down";
|
||||
};
|
||||
labels = {
|
||||
severity = "critical";
|
||||
};
|
||||
}
|
||||
];
|
||||
}
|
||||
Reference in New Issue
Block a user