chore: Apache HBase modernization

apache-hbase-mixin/alerts/alerts.libsonnet → apache-hbase-mixin/alerts.libsonnet

@@ -1,14 +1,14 @@
 {
-  prometheusAlerts+:: {
-    groups+: [
+  new(this): {
+    groups: [
       {
         name: 'apache-hbase-alerts',
         rules: [
           {
             alert: 'HBaseHighHeapMemUsage',
             expr: |||
-              100 * sum without(context, hostname, processname) (jvm_metrics_mem_heap_used_m{%(filterSelector)s} / clamp_min(jvm_metrics_mem_heap_committed_m{%(filterSelector)s}, 1))  > %(alertsHighHeapMemUsage)s
-            ||| % $._config,
+              100 * sum without(context, hostname, processname) (jvm_metrics_mem_heap_used_m{%(filteringSelector)s} / clamp_min(jvm_metrics_mem_heap_committed_m{%(filteringSelector)s}, 1))  > %(alertsHighHeapMemUsage)s
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -18,14 +18,14 @@
               description:
                 (
                   'The heap memory usage for the JVM on instance {{$labels.instance}} in cluster {{$labels.hbase_cluster}} is {{printf "%%.0f" $value}} percent, which is above the threshold of %(alertsHighHeapMemUsage)s percent'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'HBaseDeadRegionServer',
             expr: |||
               server_num_dead_region_servers > %(alertsDeadRegionServer)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -35,14 +35,14 @@
               description:
                 (
                   '{{$value}} RegionServer(s) in cluster {{$labels.hbase_cluster}} are unresponsive, which is above the threshold of %(alertsDeadRegionServer)s. The name(s) of the dead RegionServer(s) are {{$labels.deadregionservers}}'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'HBaseOldRegionsInTransition',
             expr: |||
               100 * assignment_manager_rit_count_over_threshold / clamp_min(assignment_manager_rit_count, 1) > %(alertsOldRegionsInTransition)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -52,14 +52,14 @@
               description:
                 (
                   '{{printf "%%.0f" $value}} percent of RegionServers in transition in cluster {{$labels.hbase_cluster}} are transitioning for longer than expected, which is above the threshold of %(alertsOldRegionsInTransition)s percent'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'HBaseHighMasterAuthFailRate',
             expr: |||
               100 * rate(master_authentication_failures[5m]) / (clamp_min(rate(master_authentication_successes[5m]), 1) + clamp_min(rate(master_authentication_failures[5m]), 1)) > %(alertsHighMasterAuthFailRate)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -69,14 +69,14 @@
               description:
                 (
                   '{{printf "%%.0f" $value}} percent of authentication attempts to the master are failing in cluster {{$labels.hbase_cluster}}, which is above the threshold of %(alertsHighMasterAuthFailRate)s percent'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'HBaseHighRSAuthFailRate',
             expr: |||
               100 * rate(region_server_authentication_failures[5m]) / (clamp_min(rate(region_server_authentication_successes[5m]), 1) + clamp_min(rate(region_server_authentication_failures[5m]), 1)) > %(alertsHighRSAuthFailRate)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -86,7 +86,7 @@
               description:
                 (
                   '{{printf "%%.0f" $value}} percent of authentication attempts to the RegionServer {{$labels.instance}} are failing in cluster {{$labels.hbase_cluster}}, which is above the threshold of %(alertsHighRSAuthFailRate)s percent'
-                ) % $._config,
+                ) % this.config,
             },
           },
         ],

apache-hbase-mixin/config.libsonnet

@@ -1,20 +1,35 @@
 {
-  _config+:: {
-    filterSelector: 'job="integrations/apache-hbase"',
+  local this = self,
+  filteringSelector: 'job="integrations/apache-hbase"',
+  groupLabels: ['job', 'hbase_cluster'],
+  instanceLabels: ['instance'],
+  logLabels: ['job', 'hbase_cluster', 'instance'],
 
-    dashboardTags: ['apache-hbase-mixin'],
-    dashboardPeriod: 'now-30m',
-    dashboardTimezone: 'default',
-    dashboardRefresh: '1m',
+  dashboardTags: [self.uid + '-mixin'],
+  uid: 'apache-hbase',
+  dashboardNamePrefix: 'Apache HBase',
+  dashboardPeriod: 'now-30m',
+  dashboardTimezone: 'default',
+  dashboardRefresh: '1m',
+  metricsSource: ['prometheus', 'prometheusv2'],
 
-    // alerts thresholds
-    alertsHighHeapMemUsage: 80,  // percentage
-    alertsHighNonHeapMemUsage: 80,  // percentage
-    alertsDeadRegionServer: 0,  // count
-    alertsOldRegionsInTransition: 50,  // percentage
-    alertsHighMasterAuthFailRate: 35,  // percentage
-    alertsHighRSAuthFailRate: 35,  // percentage
+  // Logging configuration
+  enableLokiLogs: true,
+  extraLogLabels: ['level'],
+  logsVolumeGroupBy: 'level',
+  showLogsVolume: true,
 
-    enableLokiLogs: true,
+  // Alerts thresholds
+  alertsHighHeapMemUsage: 80,  // percentage
+  alertsHighNonHeapMemUsage: 80,  // percentage
+  alertsDeadRegionServer: 0,  // count
+  alertsOldRegionsInTransition: 50,  // percentage
+  alertsHighMasterAuthFailRate: 35,  // percentage
+  alertsHighRSAuthFailRate: 35,  // percentage
+
+  // Signals configuration
+  signals+: {
+    cluster: (import './signals/cluster.libsonnet')(this),
+    regionserver: (import './signals/regionserver.libsonnet')(this),
   },
 }

apache-hbase-mixin/dashboards.libsonnet

@@ -0,0 +1,105 @@
+local g = import './g.libsonnet';
+local commonlib = import 'common-lib/common/main.libsonnet';
+local logslib = import 'logs-lib/logs/main.libsonnet';
+
+{
+  local root = self,
+  new(this)::
+    local prefix = this.config.dashboardNamePrefix;
+    local links = this.grafana.links;
+    local tags = this.config.dashboardTags;
+    local uid = g.util.string.slugify(this.config.uid);
+    local vars = this.grafana.variables;
+    local annotations = this.grafana.annotations;
+    local refresh = this.config.dashboardRefresh;
+    local period = this.config.dashboardPeriod;
+    local timezone = this.config.dashboardTimezone;
+    {
+      'apache-hbase-cluster-overview.json':
+        g.dashboard.new(prefix + ' cluster overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.clusterOverview,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance,
+          uid + '-cluster-overview',
+          tags,
+          links { clusterOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+
+      'apache-hbase-regionserver-overview.json':
+        g.dashboard.new(prefix + ' RegionServer overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.regionServerOverview,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance,
+          uid + '-regionserver-overview',
+          tags,
+          links { regionServerOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+    }
+    +
+    if this.config.enableLokiLogs then
+      {
+        'apache-hbase-logs.json':
+          logslib.new(
+            prefix + ' logs',
+            datasourceName=this.grafana.variables.datasources.loki.name,
+            datasourceRegex=this.grafana.variables.datasources.loki.regex,
+            filterSelector=this.config.filteringSelector,
+            labels=this.config.logLabels + this.config.extraLogLabels,
+            formatParser=null,
+            showLogsVolume=this.config.showLogsVolume,
+            logsVolumeGroupBy=this.config.logsVolumeGroupBy,
+          )
+          {
+            dashboards+:
+              {
+                logs+:
+                  root.applyCommon(super.logs.templating.list, uid=uid + '_logs', tags=tags, links=links { logs+:: {} }, annotations=annotations, timezone=timezone, refresh=refresh, period=period),
+              },
+            panels+:
+              {
+                logs+:
+                  g.panel.logs.options.withEnableLogDetails(true)
+                  + g.panel.logs.options.withShowTime(false)
+                  + g.panel.logs.options.withWrapLogMessage(false),
+              },
+            variables+: {
+              toArray+: [
+                this.grafana.variables.datasources.prometheus { hide: 2 },
+              ],
+            },
+          }.dashboards.logs,
+      }
+    else {},
+
+  applyCommon(vars, uid, tags, links, annotations, timezone, refresh, period):
+    g.dashboard.withTags(tags)
+    + g.dashboard.withUid(uid)
+    + g.dashboard.withLinks(std.objectValues(links))
+    + g.dashboard.withTimezone(timezone)
+    + g.dashboard.withRefresh(refresh)
+    + g.dashboard.time.withFrom(period)
+    + g.dashboard.withVariables(vars)
+    + g.dashboard.withAnnotations(std.objectValues(annotations)),
+}