feat: sysdig update (#12)

Author: arcshiftsolutions

.github/workflows/merge.yml

@@ -80,6 +80,15 @@ jobs:
   # unset or monitoring/alerts/ is empty — adoption is gradual. Alert
   # templates live in monitoring/alerts/ in this repo; add or remove files
   # there to customize the alert set.
+  #
+  # This repo is backend-only and deploys a single Deployment named
+  # <app>-<zone> (nr-user-lookup-api-prod) — there is no component segment in
+  # the name. So `components` is pinned to one entry (the action defaults to
+  # `frontend,backend`, which would create a phantom frontend alert set), and
+  # the queries in monitoring/alerts/*.json target __APP__-prod directly
+  # rather than the action's usual __APP__-__COMPONENT__-prod. The value below
+  # only shapes the alert names (<app>-backend-<rule>); don't reintroduce
+  # __COMPONENT__ into the queries.
   monitor-prod:
     name: Sysdig Monitor (PROD)
     needs: [deploy-prod]
@@ -95,6 +104,7 @@ jobs:
           sysdig_api_token: ${{ secrets.SYSDIG_API_TOKEN }}
           oc_namespace: ${{ secrets.oc_namespace }}
           app: ${{ github.event.repository.name }}
+          components: backend
 
   promote:
     name: Promote Images

monitoring/alerts/crashloop.json

@@ -1,12 +1,12 @@
 {
   "type": "PROMETHEUS",
   "name": "__ALERT_NAME__",
-  "description": "Container in __NAMESPACE__/__COMPONENT__ has been in CrashLoopBackOff for over __DURATION_MINUTES__ minutes.",
+  "description": "Container in __NAMESPACE__/__APP__-prod has been in CrashLoopBackOff for over __DURATION_MINUTES__ minutes.",
   "severity": "high",
   "enabled": true,
   "config": {
     "duration": 0,
-    "query": "max by (kube_deployment_name, kube_pod_name) (kube_pod_container_status_waiting_reason{reason=\"CrashLoopBackOff\", kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"}) > 0"
+    "query": "max by (kube_deployment_name, kube_pod_name) (kube_pod_container_status_waiting_reason{reason=\"CrashLoopBackOff\", kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"}) > 0"
   },
   "notificationChannelConfigList": [
     { "type": "EMAIL", "channelId": 0 }

monitoring/alerts/nopods.json

@@ -1,12 +1,12 @@
 {
   "type": "PROMETHEUS",
   "name": "__ALERT_NAME__",
-  "description": "__NAMESPACE__/__COMPONENT__ has had no available pods for over __DURATION_MINUTES__ minutes. Fires whatever the cause — crash, failed rollout, scaled to zero, or the Deployment no longer reporting at all.",
+  "description": "__NAMESPACE__/__APP__-prod has had no available pods for over __DURATION_MINUTES__ minutes. Fires whatever the cause — crash, failed rollout, scaled to zero, or the Deployment no longer reporting at all.",
   "severity": "high",
   "enabled": true,
   "config": {
     "duration": 0,
-    "query": "kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"} == 0 or absent(kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"})"
+    "query": "kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"} == 0 or absent(kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"})"
   },
   "notificationChannelConfigList": [
     { "type": "EMAIL", "channelId": 0 }

monitoring/alerts/replicas.json

@@ -1,12 +1,12 @@
 {
   "type": "PROMETHEUS",
   "name": "__ALERT_NAME__",
-  "description": "Available replicas for __NAMESPACE__/__COMPONENT__ have been below desired for over __DURATION_MINUTES__ minutes.",
+  "description": "Available replicas for __NAMESPACE__/__APP__-prod have been below desired for over __DURATION_MINUTES__ minutes.",
   "severity": "high",
   "enabled": true,
   "config": {
     "duration": 0,
-    "query": "kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"} < on (kube_namespace_name, kube_deployment_name) kube_deployment_spec_replicas{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"}"
+    "query": "kube_deployment_status_replicas_available{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"} < on (kube_namespace_name, kube_deployment_name) kube_deployment_spec_replicas{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"}"
   },
   "notificationChannelConfigList": [
     { "type": "EMAIL", "channelId": 0 }

monitoring/alerts/restarts.json

@@ -1,12 +1,12 @@
 {
   "type": "PROMETHEUS",
   "name": "__ALERT_NAME__",
-  "description": "Container in __NAMESPACE__/__COMPONENT__ has restarted more than once per minute for over __DURATION_MINUTES__ minutes.",
+  "description": "Container in __NAMESPACE__/__APP__-prod has restarted more than once per minute for over __DURATION_MINUTES__ minutes.",
   "severity": "medium",
   "enabled": true,
   "config": {
     "duration": 0,
-    "query": "sum by (kube_deployment_name, kube_pod_name) (rate(kube_pod_container_status_restarts_total{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-__COMPONENT__-prod\"}[5m])) * 60 > 1"
+    "query": "sum by (kube_deployment_name, kube_pod_name) (rate(kube_pod_container_status_restarts_total{kube_namespace_name=\"__NAMESPACE__\", kube_deployment_name=\"__APP__-prod\"}[5m])) * 60 > 1"
   },
   "notificationChannelConfigList": [
     { "type": "EMAIL", "channelId": 0 }