Merge branch 'master' into shahzeb/status-panel-lib-updates

common-lib/common/panels/disk/table/usage.libsonnet

@@ -237,13 +237,13 @@ base {
 
     + table.standardOptions.withOverrides([
       fieldOverride.byName.new('Mounted on')
-      + fieldOverride.byName.withProperty('custom.width', '260'),
+      + fieldOverride.byName.withProperty('custom.width', 260),
       fieldOverride.byName.new('Size')
-      + fieldOverride.byName.withProperty('custom.width', '80'),
+      + fieldOverride.byName.withProperty('custom.width', 80),
       fieldOverride.byName.new('Used')
-      + fieldOverride.byName.withProperty('custom.width', '80'),
+      + fieldOverride.byName.withProperty('custom.width', 80),
       fieldOverride.byName.new('Available')
-      + fieldOverride.byName.withProperty('custom.width', '80'),
+      + fieldOverride.byName.withProperty('custom.width', 80),
       fieldOverride.byName.new('Used, %')
       + fieldOverride.byName.withProperty(
         'custom.cellOptions', {

common-lib/common/variables/variables.libsonnet

@@ -13,6 +13,11 @@ local utils = import '../utils.libsonnet';
     prometheusDatasourceName=if enableLokiLogs then 'prometheus_datasource' else 'datasource',
     prometheusDatasourceLabel=if enableLokiLogs then 'Prometheus data source' else 'Data source',
   ): {
+       // strip trailing or starting comma if present:
+       // while trailing comma is accepted in PromQL expressions, starting comma is not.
+       // starting comma can be present in case of concatenation of empty filteringSelector with some extra selectors.
+       local _filteringSelector = std.stripChars(std.stripChars(filteringSelector, ' '), ','),
+
        local varMetricTemplate(varMetric, chainSelector) =
          // check if chainSelector is not empty string (case when filtering selector is empty):
          if std.type(varMetric) == 'array' && chainSelector != ''
@@ -57,24 +62,24 @@ local utils = import '../utils.libsonnet';
        // Use on dashboards where multiple entities can be selected, like fleet dashboards
        multiInstance:
          [root.datasources.prometheus]
-         + variablesFromLabels(groupLabels, instanceLabels, filteringSelector),
+         + variablesFromLabels(groupLabels, instanceLabels, _filteringSelector),
        // Use on dashboards where only single entity can be selected
        singleInstance:
          [root.datasources.prometheus]
-         + variablesFromLabels(groupLabels, instanceLabels, filteringSelector, multiInstance=false),
+         + variablesFromLabels(groupLabels, instanceLabels, _filteringSelector, multiInstance=false),
        queriesSelectorAdvancedSyntax:
          std.join(
            ',',
            std.filter(function(x) std.length(x) > 0, [
-             filteringSelector,
+             _filteringSelector,
              utils.labelsToPromQLSelectorAdvanced(groupLabels + instanceLabels),
            ])
          ),
        queriesSelector:
          std.join(
            ',',
            std.filter(function(x) std.length(x) > 0, [
-             filteringSelector,
+             _filteringSelector,
              utils.labelsToPromQLSelector(groupLabels + instanceLabels),
            ])
          ),

csp-mixin/alerts/azure-alerts.yml

@@ -13,7 +13,7 @@ groups:
         annotations:
           summary: 'VM CPU utilization is too high.'
           description: 'The VM {{ $labels.resourceName }} is under heavy load and may become unresponsive.'
-          dashboard_uid: '58f33c50e66c911b0ad8a25aa438a96e'
+          dashboard_url: '/a/grafana-csp-app/azure/dashboards/vm'
 
       - alert: AzureVMUnavailable
         expr: |
@@ -27,7 +27,7 @@ groups:
         annotations:
           summary: 'VM unavailable.'
           description: 'The VM {{ $labels.resourceName }} is not functioning or crashed, which may require immediate action.'
-          dashboard_uid: '58f33c50e66c911b0ad8a25aa438a96e'
+          dashboard_url: '/a/grafana-csp-app/azure/dashboards/vm'
 
       - alert: AzureDatabaseHighDtuConsumption
         expr: |

csp-mixin/alerts/gcp-alerts.yml

@@ -11,6 +11,132 @@ groups:
           service: 'Compute Engine'
           namespace: cloud-provider-gcp
         annotations:
-          summary: 'CPU utilization is too high.'
+          summary: 'VM CPU utilization is too high.'
           description: 'The VM {{ $labels.instance_name }} is under heavy load and may become unresponsive.'
           dashboard_uid: 'f115fe73641347c43415535d77e2dc0f'
+
+      - alert: GcpCEHighIOLatency
+        expr: |
+          avg by (job,project_id,instance_id)(stackdriver_gce_instance_compute_googleapis_com_instance_disk_average_io_latency{job=~".+",project_id=~".+",instance_id=~".+"}) > 5000
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Compute Engine'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'VM IO latency is too high.'
+          description: 'Check {{ $labels.instance_id }} VM for I/O bottlenecks and upgrade to SSD if necessary.'
+          dashboard_uid: 'f115fe73641347c43415535d77e2dc0f'
+
+      - alert: GcpCloudSQLHighCpu
+        expr: |
+          100 * avg by (job,project_id,instance,database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_cpu_utilization{job=~".+",project_id=~".+",instance=~".+", database_id=~".+"}) > 90
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'Database CPU utilization is too high.'
+          description: 'Check {{ $labels.database_id }} database for high CPU queries and optimize them, or scale up the instance if sustained high usage.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpCloudSQLMemoryUsage
+        expr: |
+          100 * avg by (job,project_id,instance,database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_memory_utilization{job=~".+",project_id=~".+",instance=~".+", database_id=~".+"}) > 85
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'Database memory utilization is too high.'
+          description: 'Review high-memory queries or add more memory to the {{ $labels.database_id }} instance.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpCloudSQLDiskUsage
+        expr: |
+          100 * avg by (job,project_id,instance,database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_disk_utilization{job=~".+",project_id=~".+",instance=~".+", database_id=~".+"}) > 85
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'Database disk utilization is too high.'
+          description: 'Delete or archive unused data, or increase disk size to the {{ $labels.database_id }} database.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpCloudSQLActiveConnections
+        expr: |
+          avg by (job,project_id,instance, database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_mysql_threads{thread_kind="THREADS_CONNECTED", job=~".+",project_id=~".+",instance=~".+", database_id=~".+"}) > 0.9 * avg by (job,project_id,instance, database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_mysql_max_connections{job=~".+",project_id=~".+",instance=~".+", database_id=~".+"})
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'Too many database active connections.'
+          description: 'Investigate connection pooling settings and connection management in your {{ $labels.database_id }} application database.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpCloudSQLAbortedConnections
+        expr: |
+          sum by(job, instance, project_id)(rate(stackdriver_cloudsql_database_cloudsql_googleapis_com_database_mysql_aborted_connects_count[5m])) > 5
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: critical
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'More than 5 MySQL failed connections in 5 minutes.'
+          description: 'Verify credentials and network settings; check for firewall rules blocking connections for the {{ $labels.database_id }} database.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpCloudSQLLagSecondsBehindMaster
+        expr: |
+          avg by (job,project_id,instance, database_id) (stackdriver_cloudsql_database_cloudsql_googleapis_com_database_mysql_replication_seconds_behind_master) > 5
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: warning
+          service: 'Cloud SQL'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'More than 5 seconds lag between database read replica and primary.'
+          description: 'Check {{ $labels.database_id }} database for network latency between primary and replica; adjust configurations to optimize replication.'
+          dashboard_uid: 'cc710d49022fdd69bed0e992891863e9'
+
+      - alert: GcpPubSubNumUndeliveredMessages
+        expr: |
+          avg by (job,project_id,instance)(stackdriver_pubsub_subscription_pubsub_googleapis_com_subscription_num_undelivered_messages{job=~".+",project_id=~".+",instance=~".+"}) > 1000
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: warning
+          service: 'Pub/Sub'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'More than 1000 unacknowledged messages for a PubSub subscription.'
+          description: 'Scale up subscribers or adjust message processing capacity for the {{ $labels.instance }} instance.'
+          dashboard_uid: '2abad1eb5e4873b95e9176e7ef10a30c'
+
+      - alert: GcpPubSubUnackedMessageAge
+        expr: |
+          avg by (job,project_id,instance)(stackdriver_pubsub_subscription_pubsub_googleapis_com_subscription_oldest_unacked_message_age{job=~".+",project_id=~".+",instance=~".+"}) > 60
+        for: 5m
+        keep_firing_for: 10m
+        labels:
+          severity: warning
+          service: 'Pub/Sub'
+          namespace: cloud-provider-gcp
+        annotations:
+          summary: 'Unacknowledged messages for more than 60 seconds for a PubSub subscription.'
+          description: 'Investigate {{ $labels.instance }} instance and speed up message processing; ensure consumers can handle the load.'
+          dashboard_uid: '2abad1eb5e4873b95e9176e7ef10a30c'

jvm-mixin/.lint

@@ -1,13 +1,8 @@
 exclusions:
-  panel-title-description-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
+  panel-datasource-rule:
+    entries:
+      - panel: GC duration
+      - panel: Allocated/promoted
   panel-units-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
-  template-datasource-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
-  template-instance-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
-  template-job-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
-  template-on-time-change-reload-rule:
-    reason: "mixtool upgrade made this rule stricter. TODO: Fix errors and remove the warning exclusion"
+    entries:
+      - panel: Process files open

jvm-mixin/config.libsonnet

@@ -1,5 +1,5 @@
 {
-  filteringSelector: 'job!=""',
+  filteringSelector: '',
   groupLabels: ['job'],
   instanceLabels: ['instance'],
   uid: 'jvm',

jvm-observ-lib/config.libsonnet

@@ -1,6 +1,6 @@
 {
   local this = self,
-  filteringSelector: 'job!=""',
+  filteringSelector: '',  // set to apply static filters to all queries and alerts, i.e. job="bar"
   groupLabels: ['job'],
   instanceLabels: ['instance'],
   uid: 'jvm',

kafka-observ-lib/config.libsonnet

@@ -1,6 +1,6 @@
 {
   local this = self,
-  filteringSelector: 'job!=""',
+  filteringSelector: '',  // set to apply static filters to all queries and alerts, i.e. job="integrations/kafka"
   zookeeperfilteringSelector: this.filteringSelector,
   groupLabels: ['kafka_cluster'],  // label(s) that defines kafka cluster
   instanceLabels: ['instance'],  // label(s) that defines single broker

kafka-observ-lib/signals/consumerGroup.libsonnet

@@ -2,7 +2,7 @@ local commonlib = import 'common-lib/common/main.libsonnet';
 
 function(this)
   {
-    filteringSelector: std.join(',', [this.filteringSelector, this.topicsFilteringSelector, this.consumerGroupFilteringSelector]),
+    filteringSelector: std.join(',', [this.topicsFilteringSelector, this.consumerGroupFilteringSelector, this.filteringSelector]),
     groupLabels: this.groupLabels,
     instanceLabels: ['topic', 'consumergroup'],  // this.instanceLabels is ommitted, as it would point to kafka_exporter instance.
     aggLevel: 'group',

kafka-observ-lib/signals/topic.libsonnet

@@ -2,7 +2,7 @@ local commonlib = import 'common-lib/common/main.libsonnet';
 
 function(this)
   {
-    filteringSelector: std.join(',', [this.filteringSelector, this.topicsFilteringSelector]),
+    filteringSelector: std.join(',', [this.topicsFilteringSelector, this.filteringSelector]),
     groupLabels: this.groupLabels,
     instanceLabels: ['topic'],  // this.instanceLabels is ommitted, as it would point to kafka_exporter instance.
     aggLevel: 'group',

kafka-observ-lib/signals/totalTime.libsonnet

@@ -3,7 +3,7 @@ local commonlib = import 'common-lib/common/main.libsonnet';
 // TotalTimeMs metric
 function(this)
   {
-    filteringSelector: this.filteringSelector + ', quantile="%s"' % this.totalTimeMsQuantile,
+    filteringSelector: ('quantile="%s"' % this.totalTimeMsQuantile) + ',' + this.filteringSelector,
     groupLabels: this.groupLabels,
     instanceLabels: this.instanceLabels,
     aggLevel: if this.totalTimeMetricsRepeat then 'instance' else 'group',

kafka-observ-lib/signals/zookeeperClient.libsonnet

@@ -2,7 +2,7 @@ local commonlib = import 'common-lib/common/main.libsonnet';
 
 function(this)
   {
-    filteringSelector: this.filteringSelector + ', quantile="%s"' % this.zookeeperClientQuantile,
+    filteringSelector: ('quantile="%s"' % this.zookeeperClientQuantile) + ',' + this.filteringSelector,
     groupLabels: this.groupLabels,
     instanceLabels: this.instanceLabels,
     aggLevel: 'instance',

logs-lib/logs/variables.libsonnet

@@ -9,6 +9,9 @@ function(
   labels,
 )
   {
+    // strip trailing or starting comma if present that are not accepted in LoqQL
+    // starting comma can be present in case of concatenation of empty filteringSelector with some extra selectors.
+    local _filteringSelector = std.stripChars(std.stripChars(filterSelector, ' '), ','),
     local this = self,
     local variablesFromLabels(labels, filterSelector) =
       local chainVarProto(chainVar) =
@@ -34,7 +37,7 @@ function(
       ;
       [
         chainVarProto(chainVar)
-        for chainVar in utils.chainLabels(labels, [filterSelector])
+        for chainVar in utils.chainLabels(labels, [_filteringSelector])
       ],
 
     datasource:
@@ -48,12 +51,15 @@ function(
 
     toArray:
       [self.datasource]
-      + variablesFromLabels(labels, filterSelector)
+      + variablesFromLabels(labels, _filteringSelector)
       + [self.regex_search],
 
     queriesSelector:
-      '%s,%s' % [
-        filterSelector,
-        utils.labelsToPromQLSelector(labels),
-      ],
+      std.join(
+        ',',
+        std.filter(function(x) std.length(x) > 0, [
+          _filteringSelector,
+          utils.labelsToPromQLSelector(labels),
+        ])
+      ),
   }

openstack-mixin/panels.libsonnet

@@ -273,8 +273,8 @@ local utils = commonlib.utils;
       )
       + gauge.queryOptions.withTargetsMixin(t.vCPUUsed)
       + gauge.standardOptions.withUnit('percent')
-      + gauge.standardOptions.withMin('0')
-      + gauge.standardOptions.withMax('150')
+      + gauge.standardOptions.withMin(0)
+      + gauge.standardOptions.withMax(150)
       + gauge.standardOptions.thresholds.withSteps([
         gauge.standardOptions.threshold.step.withValue(0) +
         gauge.standardOptions.threshold.step.withColor('green'),
@@ -288,8 +288,8 @@ local utils = commonlib.utils;
       )
       + gauge.queryOptions.withTargetsMixin(t.RAMUsed)
       + gauge.standardOptions.withUnit('percent')
-      + gauge.standardOptions.withMin('0')
-      + gauge.standardOptions.withMax('150')
+      + gauge.standardOptions.withMin(0)
+      + gauge.standardOptions.withMax(150)
       + gauge.standardOptions.thresholds.withSteps([
         gauge.standardOptions.threshold.step.withValue(0) +
         gauge.standardOptions.threshold.step.withColor('green'),

process-observ-lib/config.libsonnet

@@ -6,7 +6,7 @@
   // 'instanceLabels' - one or more labels that can be used to identify single entity of instances. In simple cases, can be 'instance' or 'pod'.
   // 'uid' - UID to prefix all dashboards original uids
   local this = self,
-  filteringSelector: 'job!=""',
+  filteringSelector: '',  // set to apply static filters to all queries and alerts, i.e. job="bar"
   groupLabels: ['job'],
   instanceLabels: ['instance'],
   dashboardTags: [self.uid],

windows-mixin/config.libsonnet

@@ -5,7 +5,7 @@
     // labels to identify single windows host:
     instanceLabels: ['instance'],
     // selector to include in all queries(including alerts)
-    filteringSelector: 'job=~".*windows.*"',
+    filteringSelector: '',
     // prefix all dashboards uids and alert groups
     uid: 'windows',
     // prefix dashboards titles

windows-observ-lib/alerts.libsonnet

@@ -80,7 +80,7 @@
           {
             alert: 'WindowsCPUHighUsage',
             expr: |||
-              100 - (avg without (mode, core) (rate(windows_cpu_time_total{%(filteringSelector)s, mode="idle"}[2m])) * 100) > %(alertsCPUThresholdWarning)s
+              100 - (avg without (mode, core) (rate(windows_cpu_time_total{mode="idle", %(filteringSelector)s}[2m])) * 100) > %(alertsCPUThresholdWarning)s
             ||| % this.config,
             'for': '15m',
             keep_firing_for: '5m',
@@ -133,7 +133,7 @@
           {
             alert: 'WindowsServiceNotHealthy',
             expr: |||
-              windows_service_status{%(filteringSelector)s, status!~"starting|stopping|ok"} > 0
+              windows_service_status{status!~"starting|stopping|ok", %(filteringSelector)s} > 0
             ||| % this.config,
             'for': '5m',
             labels: {
@@ -150,7 +150,7 @@
           {
             alert: 'WindowsDiskDriveNotHealthy',
             expr: |||
-              windows_disk_drive_status{%(filteringSelector)s, status="OK"} != 1
+              windows_disk_drive_status{status="OK", %(filteringSelector)s} != 1
             ||| % this.config,
             'for': '5m',
             labels: {