chore: Modernize f5-bigip-mixin

.gitignore

@@ -2,3 +2,4 @@
 vendor
 jsonnetfile.lock.json
 *.zip
+.worktrees

f5-bigip-mixin/alerts/alerts.libsonnet → f5-bigip-mixin/alerts.libsonnet

@@ -1,14 +1,14 @@
 {
-  prometheusAlerts+:: {
-    groups+: [
+  new(this): {
+    groups: [
       {
-        name: 'bigip-alerts',
+        name: this.config.uid + '-alerts',
         rules: [
           {
             alert: 'BigIPLowNodeAvailabilityStatus',
             expr: |||
               100 * (sum(bigip_node_status_availability_state) / clamp_min(count(bigip_node_status_availability_state), 1)) < %(alertsCriticalNodeAvailability)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'critical',
@@ -19,14 +19,14 @@
                 (
                   '{{ printf "%%.0f" $value }} percent of available nodes, ' +
                   'which is below the threshold of %(alertsCriticalNodeAvailability)s.'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'BigIPServerSideConnectionLimit',
             expr: |||
               max without(instance, job) (100 * bigip_node_serverside_cur_conns / clamp_min(bigip_node_serverside_max_conns, 1)) > %(alertsWarningServerSideConnectionLimit)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -37,14 +37,14 @@
                 (
                   '{{ 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,
+                ) % this.config,
             },
           },
           {
             alert: 'BigIPHighRequestRate',
             expr: |||
               max without(instance, job) (100 * rate(bigip_pool_tot_requests[10m]) / clamp_min(rate(bigip_pool_tot_requests[50m] offset 10m), 1)) > %(alertsCriticalHighRequestRate)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '10m',
             labels: {
               severity: 'warning',
@@ -55,14 +55,14 @@
                 (
                   '{{ printf "%%.0f" $value }} percent increase in requests on pool {{$labels.pool}}, ' +
                   'which is above the threshold of %(alertsCriticalHighRequestRate)s.'
-                ) % $._config,
+                ) % this.config,
             },
           },
           {
             alert: 'BigIPHighConnectionQueueDepth',
             expr: |||
               max without(instance, job) (100 * rate(bigip_pool_connq_depth[5m])) / clamp_min(rate(bigip_pool_connq_depth[50m] offset 10m), 1) > %(alertsCriticalHighConnectionQueueDepth)s
-            ||| % $._config,
+            ||| % this.config,
             'for': '5m',
             labels: {
               severity: 'warning',
@@ -73,7 +73,7 @@
                 (
                   '{{ printf "%%.0f" $value }} percent increase in connection queue depth on node {{$labels.pool}}, ' +
                   'which is above the threshold of %(alertsCriticalHighConnectionQueueDepth)s.'
-                ) % $._config,
+                ) % this.config,
             },
           },
         ],

f5-bigip-mixin/config.libsonnet

@@ -1,17 +1,40 @@
 {
-  _config+:: {
-    dashboardTags: ['f5-bigip-mixin'],
-    dashboardPeriod: 'now-30m',
-    dashboardTimezone: 'default',
-    dashboardRefresh: '1m',
+  local this = self,
 
-    // alerts thresholds
-    alertsCriticalNodeAvailability: 95,  // %
-    alertsWarningServerSideConnectionLimit: 80,  // %
-    alertsCriticalHighRequestRate: 150,  // %
-    alertsCriticalHighConnectionQueueDepth: 75,  // %
+  // Basic filtering
+  filteringSelector: 'job=~"$job", instance=~"$instance"',
+  groupLabels: ['job'],
+  instanceLabels: ['instance'],
 
-    enableLokiLogs: false,
-    filterSelector: 'job=~"syslog"',
+  // Dashboard settings
+  dashboardTags: ['f5-bigip-mixin'],
+  uid: 'f5-bigip',
+  dashboardNamePrefix: 'F5 BIG-IP',
+  dashboardRefresh: '1m',
+  dashboardPeriod: 'now-30m',
+  dashboardTimezone: 'default',
+
+  // Logs configuration
+  enableLokiLogs: true,
+  filterSelector: 'job=~"syslog"',
+  logLabels: ['job', 'host', 'syslog_facility', 'level'],
+  extraLogLabels: [],
+  logsVolumeGroupBy: 'level',
+  showLogsVolume: true,
+
+  // Alert thresholds
+  alertsCriticalNodeAvailability: 95,  // %
+  alertsWarningServerSideConnectionLimit: 80,  // %
+  alertsCriticalHighRequestRate: 150,  // %
+  alertsCriticalHighConnectionQueueDepth: 75,  // %
+
+  // Metrics source
+  metricsSource: 'prometheus',
+
+  signals+: {
+    system: (import './signals/system.libsonnet')(this),
+    virtualServers: (import './signals/virtual-servers.libsonnet')(this),
+    pools: (import './signals/pools.libsonnet')(this),
+    nodes: (import './signals/nodes.libsonnet')(this),
   },
 }

f5-bigip-mixin/dashboards.libsonnet

@@ -0,0 +1,225 @@
+local g = import './g.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 = 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;
+
+    {
+      // Cluster Overview Dashboard
+      'f5-bigip-cluster-overview.json':
+        g.dashboard.new(prefix + ' cluster overview')
+        + g.dashboard.withDescription('Overview of F5 BIG-IP cluster health and top metrics.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.clusterOverviewRow,
+              ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            // k (top count) variable
+            g.dashboard.variable.custom.new('k', values=['5', '10', '20', '50'])
+            + g.dashboard.variable.custom.generalOptions.withLabel('Top node count')
+            + g.dashboard.variable.custom.generalOptions.withCurrent('5'),
+            // bigip_partition variable
+            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{job=~"$job", instance=~"$instance", partition=~"$bigip_partition"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '-cluster-overview',
+          tags,
+          links { clusterOverview+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+
+      // Node Overview Dashboard
+      'f5-bigip-node-overview.json':
+        g.dashboard.new(prefix + ' node overview')
+        + g.dashboard.withDescription('Detailed view of F5 BIG-IP node metrics and status.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.nodeStatusRow,
+                this.grafana.rows.nodeMetricsRow,
+              ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            // bigip_node variable
+            g.dashboard.variable.query.new('bigip_node')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('node', 'bigip_node_status_availability_state{job=~"$job", instance=~"$instance"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP node')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+            // bigip_partition variable
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_node_status_availability_state{job=~"$job", instance=~"$instance", node=~"$bigip_node"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '-node-overview',
+          tags,
+          links { nodeOverview+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+
+      // Pool Overview Dashboard
+      'f5-bigip-pool-overview.json':
+        g.dashboard.new(prefix + ' pool overview')
+        + g.dashboard.withDescription('Detailed view of F5 BIG-IP pool metrics and status.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.poolStatusRow,
+                this.grafana.rows.poolMetricsRow,
+              ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            // bigip_pool variable
+            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{job=~"$job", instance=~"$instance"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP pool')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+            // bigip_partition variable
+            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{job=~"$job", instance=~"$instance", pool=~"$bigip_pool"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '-pool-overview',
+          tags,
+          links { poolOverview+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+
+      // Virtual Server Overview Dashboard
+      'f5-bigip-virtual-server-overview.json':
+        g.dashboard.new(prefix + ' virtual server overview')
+        + g.dashboard.withDescription('Detailed view of F5 BIG-IP virtual server metrics and status.')
+        + g.dashboard.withPanels(
+          g.util.panel.resolveCollapsedFlagOnRows(
+            g.util.grid.wrapPanels(
+              [
+                this.grafana.rows.virtualServerStatusRow,
+                this.grafana.rows.virtualServerClientsideMetricsRow,
+                this.grafana.rows.virtualServerEphemeralMetricsRow,
+              ]
+            )
+          )
+        )
+        + root.applyCommon(
+          vars.multiInstance + [
+            // bigip_virtual_server variable
+            g.dashboard.variable.query.new('bigip_virtual_server')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('virtual_server', 'bigip_vs_status_availability_state{job=~"$job", instance=~"$instance"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP virtual server')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+            // bigip_partition variable
+            g.dashboard.variable.query.new('bigip_partition')
+            + g.dashboard.variable.query.withDatasourceFromVariable(vars.datasources.prometheus)
+            + g.dashboard.variable.query.queryTypes.withLabelValues('partition', 'bigip_vs_status_availability_state{job=~"$job", instance=~"$instance", virtual_server=~"$bigip_virtual_server"}')
+            + g.dashboard.variable.query.generalOptions.withLabel('BIG-IP partition')
+            + g.dashboard.variable.query.selectionOptions.withMulti(true)
+            + g.dashboard.variable.query.selectionOptions.withIncludeAll(true, '.+')
+            + g.dashboard.variable.query.refresh.onLoad()
+            + g.dashboard.variable.query.refresh.onTime(),
+          ],
+          uid + '-virtual-server-overview',
+          tags,
+          links { virtualServerOverview+:: {} },
+          annotations,
+          timezone,
+          refresh,
+          period
+        ),
+    }
+    +
+    if this.config.enableLokiLogs then
+      {
+        'f5-bigip-logs.json':
+          logslib.new(
+            prefix + ' logs',
+            datasourceName=vars.datasources.loki.name,
+            datasourceRegex=vars.datasources.loki.regex,
+            filterSelector=this.config.filterSelector,
+            labels=this.config.logLabels,
+            formatParser=null,
+            showLogsVolume=this.config.showLogsVolume,
+            logsVolumeGroupBy=this.config.logsVolumeGroupBy,
+          )
+          {
+            dashboards+:
+              {
+                logs+:
+                  g.dashboard.withUid(uid + '-logs')
+                  + g.dashboard.withTags(tags)
+                  + g.dashboard.withRefresh(refresh)
+                  + g.dashboard.withLinks(std.objectValues(links { logs+:: {} }))
+                  + g.panel.logs.options.withShowTime(false),
+              },
+          }.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)),
+}