rpi.carlosedp.cluster-monit.../elasticsearch_exporter.jsonnet
2019-08-08 19:39:08 -03:00

189 lines
6.4 KiB
Plaintext

local k = import 'ksonnet/ksonnet.beta.4/k.libsonnet';
{
_config+:: {
namespace: 'monitoring',
replicas: 1,
versions+:: {
elasticExporter: '1.0.2',
},
imageRepos+:: {
elasticExporter: 'justwatch/elasticsearch_exporter',
},
prometheus+:: {
namespaces+:: ['logging'],
},
// Add custom dashboards
grafanaDashboards+:: {
'elasticsearch-dashboard.json': (import 'grafana-dashboards/elasticsearch-dashboard.json'),
'fluentd-dashboard.json': (import 'grafana-dashboards/fluentd-dashboard.json'),
},
},
elasticExporter+:: {
deployment:
local deployment = k.apps.v1beta2.deployment;
local container = k.apps.v1beta2.deployment.mixin.spec.template.spec.containersType;
local containerPort = container.portsType;
local podLabels = { 'k8s-app': 'elasticsearch-exporter' };
local elasticExporter =
container.new('elasticsearch-exporter',
$._config.imageRepos.elasticExporter + ':' + $._config.versions.elasticExporter) +
container.withCommand([
'/bin/elasticsearch_exporter',
'-es.uri=http://elasticsearch.logging.svc:9200',
'-es.timeout=60s',
'-es.all=true',
]) +
container.withPorts(containerPort.newNamed(9108, 'es-metrics')) +
container.mixin.securityContext.capabilities.withDrop(['SETPCAP', 'MKNOD', 'AUDIT_WRITE', 'CHOWN', 'NET_RAW', 'DAC_OVERRIDE', 'FOWNER', 'FSETID', 'KILL', 'SETGID', 'SETUID', 'NET_BIND_SERVICE', 'SYS_CHROOT', 'SETFCAP']) +
container.mixin.securityContext.withRunAsNonRoot(true) +
container.mixin.securityContext.withRunAsUser(1000) +
container.mixin.securityContext.withReadOnlyRootFilesystem(true) +
container.mixin.resources.withRequests({ memory: '64Mi', cpu: '25m' }) +
container.mixin.resources.withLimits({ memory: '128Mi', cpu: '100m' }) +
container.mixin.livenessProbe.httpGet.withPath('/health') +
container.mixin.livenessProbe.httpGet.withPort(9108) +
container.mixin.livenessProbe.withInitialDelaySeconds(30) +
container.mixin.livenessProbe.withTimeoutSeconds(10) +
container.mixin.readinessProbe.httpGet.withPath('/health') +
container.mixin.readinessProbe.httpGet.withPort(9108) +
container.mixin.readinessProbe.withInitialDelaySeconds(30) +
container.mixin.readinessProbe.withTimeoutSeconds(10);
local c = [elasticExporter];
deployment.new('elasticsearch-exporter', $._config.replicas, c, podLabels) +
deployment.mixin.metadata.withNamespace($._config.namespace) +
deployment.mixin.metadata.withLabels(podLabels) +
deployment.mixin.spec.selector.withMatchLabels(podLabels) +
deployment.mixin.spec.strategy.withType('RollingUpdate') +
deployment.mixin.spec.strategy.rollingUpdate.withMaxSurge(1) +
deployment.mixin.spec.strategy.rollingUpdate.withMaxUnavailable(0) +
deployment.mixin.spec.template.spec.withRestartPolicy('Always'),
service:
local service = k.core.v1.service;
local servicePort = k.core.v1.service.mixin.spec.portsType;
local elasticExporterPorts = servicePort.newNamed('es-metrics', 9108, 'es-metrics');
service.new('elasticsearch-exporter', $.elasticExporter.deployment.spec.selector.matchLabels, elasticExporterPorts) +
service.mixin.metadata.withNamespace($._config.namespace) +
service.mixin.metadata.withLabels({ 'k8s-app': 'elasticsearch-exporter' }),
serviceMonitorElastic:
{
apiVersion: 'monitoring.coreos.com/v1',
kind: 'ServiceMonitor',
metadata: {
name: 'elasticsearch-exporter',
namespace: $._config.namespace,
labels: {
'k8s-app': 'elasticsearch-exporter',
},
},
spec: {
jobLabel: 'k8s-app',
selector: {
matchLabels: {
'k8s-app': 'elasticsearch-exporter',
},
},
endpoints: [
{
port: 'es-metrics',
scheme: 'http',
interval: '30s',
},
],
namespaceSelector: {
matchNames: [
'monitoring',
],
},
},
},
serviceMonitorFluentd:
{
apiVersion: 'monitoring.coreos.com/v1',
kind: 'ServiceMonitor',
metadata: {
name: 'fluentd-es',
namespace: $._config.namespace,
labels: {
'k8s-app': 'fluentd-es',
},
},
spec: {
jobLabel: 'k8s-app',
selector: {
matchLabels: {
'k8s-app': 'fluentd-es',
},
},
endpoints: [
{
port: 'metrics',
scheme: 'http',
interval: '30s',
},
],
namespaceSelector: {
matchNames: [
'logging',
],
},
},
},
},
// Add Prometheus monitoring rules for ElasticSearch
prometheusRules+:: {
groups+: [
{
name: 'elasticsearch-k8s-rules',
rules: [
{
expr: '100 * (elasticsearch_filesystem_data_size_bytes - elasticsearch_filesystem_data_free_bytes) / elasticsearch_filesystem_data_size_bytes',
record: 'elasticsearch_filesystem_data_used_percent',
},
{
expr: '100 - elasticsearch_filesystem_data_used_percent',
record: 'elasticsearch_filesystem_data_free_percent',
},
{
expr: 'elasticsearch_cluster_health_number_of_nodes < 3',
alert: 'ElasticsearchTooFewNodesRunning',
'for': '5m',
annotations: {
message: 'There are only {{$value}} < 3 ElasticSearch nodes running',
summary: 'ElasticSearch running on less than 3 nodes',
},
labels: {
severity: 'critical',
},
},
{
expr: 'elasticsearch_jvm_memory_used_bytes{area="heap"} / elasticsearch_jvm_memory_max_bytes{area="heap"} > 0.9',
alert: 'ElasticsearchHeapTooHigh',
'for': '15m',
annotations: {
message: 'The heap usage is over 90% for 15m',
summary: 'ElasticSearch node {{$labels.node}} heap usage is high',
},
labels: {
severity: 'critical',
},
},
],
},
],
},
}