chore: F5 mixin modernization

f5-bigip-mixin/alerts.libsonnet

@@ -0,0 +1,85 @@
+{
+  new(this): {
+    local config = this.config,
+
+    groups: [
+      {
+        name: 'f5-bigip-alerts',
+        rules: [
+          {
+            alert: 'BigIPLowNodeAvailabilityStatus',
+            expr: |||
+              100 * (sum(bigip_node_status_availability_state{%(filteringSelector)s}) / clamp_min(count(bigip_node_status_availability_state{%(filteringSelector)s}), 1)) < %(alertsCriticalNodeAvailability)s
+            ||| % config,
+            'for': '5m',
+            labels: {
+              severity: 'critical',
+            },
+            annotations: {
+              summary: 'Detecting a significant number of unavailable nodes which can causes potential downtime or degraded performance.',
+              description:
+                (
+                  '{{ printf "%%.0f" $value }} percent of available nodes, ' +
+                  'which is below the threshold of %(alertsCriticalNodeAvailability)s.'
+                ) % config,
+            },
+          },
+          {
+            alert: 'BigIPServerSideConnectionLimit',
+            expr: |||
+              max without(instance, job) (100 * bigip_node_serverside_cur_conns{%(filteringSelector)s} / clamp_min(bigip_node_serverside_max_conns{%(filteringSelector)s}, 1)) > %(alertsWarningServerSideConnectionLimit)s
+            ||| % config,
+            'for': '5m',
+            labels: {
+              severity: 'warning',
+            },
+            annotations: {
+              summary: 'Approaching the connection limit may lead to rejecting new connections, impacting availability.',
+              description:
+                (
+                  '{{ printf "%%.0f" $value }} percent of the max number of connections in use on node {{$labels.node}}, ' +
+                  'which is above the threshold of %(alertsWarningServerSideConnectionLimit)s percent.'
+                ) % config,
+            },
+          },
+          {
+            alert: 'BigIPHighRequestRate',
+            expr: |||
+              max without(instance, job) (100 * rate(bigip_pool_tot_requests{%(filteringSelector)s}[10m]) / clamp_min(rate(bigip_pool_tot_requests{%(filteringSelector)s}[50m] offset 10m), 1)) > %(alertsCriticalHighRequestRate)s
+            ||| % config,
+            'for': '10m',
+            labels: {
+              severity: 'warning',
+            },
+            annotations: {
+              summary: 'An unexpected spike in requests might indicate an issue like a DDoS attack or unexpected high load.',
+              description:
+                (
+                  '{{ printf "%%.0f" $value }} percent increase in requests on pool {{$labels.pool}}, ' +
+                  'which is above the threshold of %(alertsCriticalHighRequestRate)s.'
+                ) % config,
+            },
+          },
+          {
+            alert: 'BigIPHighConnectionQueueDepth',
+            expr: |||
+              max without(instance, job) (100 * rate(bigip_pool_connq_depth{%(filteringSelector)s}[5m])) / clamp_min(rate(bigip_pool_connq_depth{%(filteringSelector)s}[50m] offset 10m), 1) > %(alertsCriticalHighConnectionQueueDepth)s
+            ||| % config,
+            'for': '5m',
+            labels: {
+              severity: 'warning',
+            },
+            annotations: {
+              summary: 'A sudden spike or sustained high queue depth may indicate a bottleneck in handling incoming connections.',
+              description:
+                (
+                  '{{ printf "%%.0f" $value }} percent increase in connection queue depth on node {{$labels.pool}}, ' +
+                  'which is above the threshold of %(alertsCriticalHighConnectionQueueDepth)s.'
+                ) % config,
+            },
+          },
+        ],
+      },
+    ],
+  },
+}

f5-bigip-mixin/config.libsonnet

@@ -1,17 +1,31 @@
 {
-  _config+:: {
-    dashboardTags: ['f5-bigip-mixin'],
-    dashboardPeriod: 'now-30m',
-    dashboardTimezone: 'default',
-    dashboardRefresh: '1m',
+  local this = self,
+  filteringSelector: 'job="integrations/f5-bigip"',
+  groupLabels: ['job'],
+  instanceLabels: ['instance'],
+  uid: 'f5-bigip',
+  dashboardNamePrefix: 'F5 BIG-IP',
+  dashboardTags: [self.uid + '-mixin'],
+  dashboardPeriod: 'now-30m',
+  dashboardTimezone: 'default',
+  dashboardRefresh: '1m',
 
-    // alerts thresholds
-    alertsCriticalNodeAvailability: 95,  // %
-    alertsWarningServerSideConnectionLimit: 80,  // %
-    alertsCriticalHighRequestRate: 150,  // %
-    alertsCriticalHighConnectionQueueDepth: 75,  // %
+  // alerts thresholds
+  alertsCriticalNodeAvailability: 95,  // %
+  alertsWarningServerSideConnectionLimit: 80,  // %
+  alertsCriticalHighRequestRate: 150,  // %
+  alertsCriticalHighConnectionQueueDepth: 75,  // %
 
-    enableLokiLogs: false,
-    filterSelector: 'job=~"syslog"',
+  enableLokiLogs: true,
+  extraLogLabels: [],
+  showLogsVolume: true,
+
+  // metrics source for signals
+  metricsSource: 'prometheus',
+  signals: {
+    cluster: (import './signals/cluster.libsonnet')(this),
+    node: (import './signals/node.libsonnet')(this),
+    pool: (import './signals/pool.libsonnet')(this),
+    virtualServer: (import './signals/virtualserver.libsonnet')(this),
   },
 }

f5-bigip-mixin/dashboards.libsonnet

@@ -0,0 +1,206 @@
+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;
+
+    {
+      'bigip-cluster-overview.json':
+        g.dashboard.new(prefix + ' cluster overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.clusterOverviewRow,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.custom.new('k', values=['5', '10', '20', '50'])
+            + g.dashboard.variable.query.selectionOptions.withMulti(false)
+            + g.dashboard.variable.custom.generalOptions.withCurrent('5')
+            + g.dashboard.variable.custom.generalOptions.withLabel('Top node count')
+            + g.dashboard.variable.query.refresh.onTime(),
+
+
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_node_status_availability_state{' + this.config.filteringSelector + '}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.refresh.onTime(),
+
+
+          ],
+          uid + '_cluster_overview',
+          tags,
+          links { f5BigipClusterOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+
+      'bigip-node-overview.json':
+        g.dashboard.new(prefix + ' node overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.nodeOverviewRow,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.query.new('bigip_node')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('node', 'bigip_node_status_availability_state{' + this.config.filteringSelector + '}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP node')
+            + g.dashboard.variable.query.refresh.onTime(),
+
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_node_status_availability_state{' + this.config.filteringSelector + ', node=~"$bigip_node"}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '_node_overview',
+          tags,
+          links { f5BigipNodeOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+
+      'bigip-pool-overview.json':
+        g.dashboard.new(prefix + ' pool overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.poolOverviewRow,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.query.new('bigip_pool')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('pool', 'bigip_pool_status_availability_state{' + this.config.filteringSelector + '}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP pool')
+            + g.dashboard.variable.query.refresh.onTime(),
+
+
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_pool_status_availability_state{' + this.config.filteringSelector + ', pool=~"$bigip_pool"}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '_pool_overview',
+          tags,
+          links { f5BigipPoolOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+
+      'bigip-virtual-server-overview.json':
+        g.dashboard.new(prefix + ' virtual server overview')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.virtualServerOverviewRow,
+              ]
+            )
+          )
+        ) + root.applyCommon(
+          vars.multiInstance + [
+            g.dashboard.variable.query.new('bigip_vs')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('vs', 'bigip_vs_status_availability_state{' + this.config.filteringSelector + '}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP virtual server')
+            + g.dashboard.variable.query.refresh.onTime(),
+
+
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_vs_status_availability_state{' + this.config.filteringSelector + ', vs=~"$bigip_vs"}')
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '_virtual_server_overview',
+          tags,
+          links { f5BigipVirtualServerOverview:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period,
+        ),
+    } + if this.config.enableLokiLogs then {
+      'bigip-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.groupLabels + this.config.extraLogLabels,
+          formatParser=null,
+          showLogsVolume=this.config.showLogsVolume,
+        )
+        {
+          dashboards+:
+            {
+              logs+:
+                root.applyCommon(super.logs.templating.list, uid=uid + '-logs', tags=tags, links=links { f5BigipLogs+:: {} }, annotations=annotations, timezone=timezone, refresh=refresh, period=period),
+            },
+          panels+:
+            {
+              logs+:
+                {
+                  logsPanel+:
+                    g.panel.logs.options.withShowTime(true),
+                },
+            },
+        }.dashboards.logs,
+    } else {},
+
+  applyCommon(vars, uid, tags, links, annotations, timezone, refresh, period):
+    g.dashboard.withTags(tags)
+    + g.dashboard.withUid(uid)
+    + g.dashboard.withLinks(links)
+    + g.dashboard.withTimezone(timezone)
+    + g.dashboard.withRefresh(refresh)
+    + g.dashboard.time.withFrom(period)
+    + g.dashboard.withVariables(vars)
+    + g.dashboard.withAnnotations(annotations),
+}