Deliver the cert helpers to the puppet compiler pods

certmanager and sshsignhost are invoked server-side by generate() during catalog compilation but only exist on the legacy VM masters, so compiles on the k8s compilers fail with "No such file or directory".

- Stage a self-contained EL9 python runtime on the shared bins volume
- Render both helpers from their puppet-prod ERB templates on the code volume rather than copying the scripts here
- Mount a name-dispatching launcher at /usr/local/bin/{certmanager,sshsignhost}
- Mount kubernetes-auth configs at /opt/{certmanager,sshsignhost}/config.yaml

Needs terraform-vault #152 applied and the puppet-prod kubernetes-auth PR merged.

.woodpecker/vector-test.yaml

@@ -0,0 +1,29 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: vector-test
+    image: artifactapi.k8s.syd1.au.unkin.net/dockerhub/timberio/vector:0.57.0-debian
+    commands:
+      # Dummy creds + writable dirs so the full topologies build; the unit tests
+      # only exercise the transforms (sources are not started).
+      - export CLICKHOUSE_USER=ci CLICKHOUSE_PASSWORD=ci
+      - export NATS_PRODUCER_PASSWORD=ci NATS_CONSUMER_PASSWORD=ci
+      - mkdir -p /vector-data-dir /etc/vault-ca
+      - cp /etc/ssl/certs/ca-certificates.crt /etc/vault-ca/ca.crt
+      # Transform tier + VM ingest: unit-tested transforms.
+      - vector test apps/base/logging/vector/aggregator.yaml apps/base/logging/vector/aggregator-tests.yaml
+      - vector test apps/base/logging/vector/vm-ingest.yaml apps/base/logging/vector/vm-ingest-tests.yaml
+      # Agent has no transforms to unit-test; validate it builds. (The archiver
+      # leg is now the logarchiver service, not a Vector pipeline.)
+      - vector validate --no-environment apps/base/logging/vector/agent.yaml
+    backend_options:
+      kubernetes:
+        serviceAccountName: default
+        resources:
+          requests:
+            memory: 256Mi
+            cpu: 250m
+          limits:
+            memory: 1Gi
+            cpu: 1

apps/base/artifactapi/api-deployment.yaml

@@ -36,7 +36,7 @@ spec:
               mountPath: /combined-certs
       containers:
         - name: api
-          image: git.unkin.net/unkin/artifactapi:v3.12.0
+          image: git.unkin.net/unkin/artifactapi:v3.11.2
           imagePullPolicy: IfNotPresent
           ports:
             - containerPort: 8000

apps/base/artifactapi/ui-deployment.yaml

@@ -22,7 +22,7 @@ spec:
       automountServiceAccountToken: true
       containers:
         - name: ui
-          image: git.unkin.net/unkin/artifactapi-ui:v3.12.0
+          image: git.unkin.net/unkin/artifactapi-ui:v3.11.2
           imagePullPolicy: IfNotPresent
           ports:
             - containerPort: 80

apps/base/authentik/cnpg_cluster.yaml

@@ -64,12 +64,8 @@ spec:
       archive_mode: "on"
       archive_timeout: 5min
       dynamic_shared_memory_type: posix
-      effective_cache_size: 1536MB
+      effective_cache_size: 256MB
       full_page_writes: "on"
-      # Replicas report their oldest xmin to the primary, so multi-second reads on
-      # a hot standby stop exhausting max_standby_streaming_delay and being
-      # cancelled. Retained-dead-tuple cost is negligible on a ~155MB database.
-      hot_standby_feedback: "on"
       log_destination: csvlog
       log_directory: /controller/log
       log_filename: postgres
@@ -81,12 +77,7 @@ spec:
       max_parallel_workers: "16"
       max_replication_slots: "16"
       max_worker_processes: "16"
-      # A pg_stat_statements.* parameter is what makes CNPG treat the extension as
-      # managed and run CREATE EXTENSION in every database; preloading alone does
-      # not create it.
-      pg_stat_statements.max: "10000"
-      pg_stat_statements.track: top
-      shared_buffers: 512MB
+      shared_buffers: 128MB
       shared_memory_type: mmap
       ssl_max_protocol_version: TLSv1.3
       ssl_min_protocol_version: TLSv1.3
@@ -95,9 +86,6 @@ spec:
       wal_log_hints: "on"
       wal_receiver_timeout: 5s
       wal_sender_timeout: 5s
-    # CNPG merges this with the libraries it manages itself.
-    shared_preload_libraries:
-      - pg_stat_statements
     syncReplicaElectionConstraint:
       enabled: false
   primaryUpdateMethod: restart
@@ -117,16 +105,13 @@ spec:
     updateInterval: 30
   resources:
     limits:
-      # 500m is a 50ms CFS quota per 100ms period, exhausted by bursts even at
-      # ~0.01 cores average, so every query pays throttle latency.
-      cpu: "2"
+      cpu: 500m
       # 512Mi OOMKilled replicas under load (shared_buffers 128MB +
       # max_connections 200 leave no headroom) — see incident 2026-07-28.
-      # shared_buffers 512MB needs the same headroom multiple, hence 2Gi.
-      memory: 2Gi
-    requests:
-      cpu: 500m
       memory: 1Gi
+    requests:
+      cpu: 50m
+      memory: 512Mi
   smartShutdownTimeout: 180
   startDelay: 3600
   stopDelay: 1800

apps/base/authentik/httproute.yaml

@@ -37,22 +37,6 @@ spec:
       name: authentik
       sectionName: https
   rules:
-    - backendRefs:
-        - group: ""
-          kind: Service
-          name: authentik-server
-          port: 80
-          weight: 1
-      filters:
-        - type: URLRewrite
-          urlRewrite:
-            path:
-              type: ReplaceFullPath
-              replaceFullPath: /application/o/token/
-      matches:
-        - path:
-            type: Exact
-            value: /application/o/token
     - backendRefs:
         - group: ""
           kind: Service
@@ -102,22 +86,6 @@ spec:
       name: authentik-internal
       sectionName: https
   rules:
-    - backendRefs:
-        - group: ""
-          kind: Service
-          name: authentik-server
-          port: 80
-          weight: 1
-      filters:
-        - type: URLRewrite
-          urlRewrite:
-            path:
-              type: ReplaceFullPath
-              replaceFullPath: /application/o/token/
-      matches:
-        - path:
-            type: Exact
-            value: /application/o/token
     - backendRefs:
         - group: ""
           kind: Service

apps/base/authentik/kustomization.yaml

@@ -19,7 +19,6 @@ resources:
   - redis-deployment.yaml
   - redis-pvc.yaml
   - redis-service.yaml
-  - server-vmpodscrape.yaml
   - vaultauth.yaml
   - vaultstaticsecret.yaml
   - vmpodscrape.yaml

apps/base/authentik/server-vmpodscrape.yaml

@@ -1,16 +0,0 @@
----
-# Scrape the authentik server's django_prometheus endpoint (:9300). Picked up
-# by the observability VMAgent (selectAllByDefault).
-apiVersion: operator.victoriametrics.com/v1beta1
-kind: VMPodScrape
-metadata:
-  name: authentik-server
-  namespace: authentik
-spec:
-  selector:
-    matchLabels:
-      app.kubernetes.io/name: authentik
-      app.kubernetes.io/component: server
-  podMetricsEndpoints:
-    - port: metrics
-      path: /metrics

apps/base/bind-external/acme-unkin-net/a/kustomization.yaml

@@ -1,6 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - ns1.yaml

apps/base/bind-external/acme-unkin-net/a/ns1.yaml

@@ -1,16 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: acme-ns1-a
-  namespace: bind-external
-spec:
-  zoneRef: acme-unkin-net
-  name: ns1
-  type: A
-  ttl: 3600
-  values:
-    # Public address of this cluster's external BIND, same target as
-    # acme-ns1.unkin.net. Resolvers that cached the seeded ns1.acme.unkin.net
-    # NS name must still reach the zone.
-    - 103.216.191.185

apps/base/bind-external/acme-unkin-net/kustomization.yaml

@@ -1,11 +0,0 @@
----
-# Authoritative delegation records for acme.unkin.net. Without these the zone
-# only holds the operator's seed apex (NS ns1.acme.unkin.net glued to the
-# primary pod IP), which is unroutable off-cluster and goes stale on
-# reschedule. DNSRecords must live in the same namespace as their BindZone.
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - ns
-  - a

apps/base/bind-external/acme-unkin-net/ns/apex.yaml

@@ -1,16 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: acme-apex-ns
-  namespace: bind-external
-spec:
-  zoneRef: acme-unkin-net
-  # "@" is the zone apex.
-  name: "@"
-  type: NS
-  ttl: 3600
-  values:
-    # Matches the parent delegation in Google Cloud DNS. Out of zone, so the
-    # child needs no glue of its own.
-    - acme-ns1.unkin.net.

apps/base/bind-external/kustomization.yaml

@@ -7,5 +7,4 @@ resources:
   - cluster.yaml
   - tsigkey.yaml
   - zones.yaml
-  - acme-unkin-net
   - agent-dns-rolebinding.yaml

apps/base/bind-external/zones.yaml

@@ -17,14 +17,3 @@ spec:
   updateKeyRef: certmanager
   allowTransfer:
     - key certmanager
-  # Published apex NS. acme-ns1 is what the parent delegates to and glues; ns1 is
-  # in-zone, so its address is declared below or a reseed would glue it to the
-  # primary pod IP.
-  nameservers:
-    - acme-ns1.unkin.net.
-    - ns1.acme.unkin.net.
-  records:
-    - name: ns1
-      type: A
-      ttl: 3600
-      values: ["103.216.191.185"]

apps/base/bind-internal/authoritative/ceph-unkin-net/cname/dashboard.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: dashboard-ceph-cname
-  namespace: bind-internal
-spec:
-  zoneRef: ceph-unkin-net
-  name: dashboard
-  type: CNAME
-  ttl: 600
-  values:
-    # Ceph mgr dashboard, reached via lb1. Lets in-cluster clients (the
-    # cephrgw-operator) resolve dashboard.ceph.unkin.net.
-    - lb1.unkin.net.

apps/base/bind-internal/authoritative/ceph-unkin-net/cname/kustomization.yaml

@@ -1,7 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - dashboard.yaml
-  - s3.yaml

apps/base/bind-internal/authoritative/ceph-unkin-net/cname/s3.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: s3-ceph-cname
-  namespace: bind-internal
-spec:
-  zoneRef: ceph-unkin-net
-  name: s3
-  type: CNAME
-  ttl: 600
-  values:
-    # radosgw S3 endpoint. Points at the Consul service for now; the real
-    # target will be changed later.
-    - radosgw.service.consul.

apps/base/bind-internal/authoritative/ceph-unkin-net/kustomization.yaml

@@ -1,6 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - cname

apps/base/bind-internal/authoritative/kustomization.yaml

@@ -2,15 +2,9 @@
 apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 
-# Individually-managed authoritative records live under <zone>/<type>/<record>.yaml.
-# DNSRecords must live in the same namespace as their BindZone (the operator
-# resolves zoneRef/clusterRef/updateKeyRef within the record's namespace), so
-# these sit alongside the zone in bind-internal, not in the app namespace.
 resources:
   - cluster.yaml
   - tsigkey.yaml
   - zones.yaml
-  - unkin-net
-  - main-unkin-net
-  - ceph-unkin-net
+  - records.yaml
   - acls.yaml

apps/base/bind-internal/authoritative/main-unkin-net/a/jellyfin.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: jellyfin-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: jellyfin
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/kustomization.yaml

@@ -1,12 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - jellyfin.yaml
-  - lidarr.yaml
-  - nzbget.yaml
-  - prowlarr.yaml
-  - radarr.yaml
-  - readarr.yaml
-  - sonarr.yaml

apps/base/bind-internal/authoritative/main-unkin-net/a/lidarr.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: lidarr-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: lidarr
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/nzbget.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: nzbget-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: nzbget
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/prowlarr.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: prowlarr-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: prowlarr
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/radarr.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: radarr-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: radarr
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/readarr.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: readarr-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: readarr
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/a/sonarr.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: sonarr-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: main-unkin-net
-  name: sonarr
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/main-unkin-net/kustomization.yaml

@@ -1,6 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - a

apps/base/bind-internal/authoritative/records.yaml

@@ -0,0 +1,164 @@
+# Individually-managed authoritative records for the unkin.net zone.
+# DNSRecords must live in the same namespace as their BindZone (the operator
+# resolves zoneRef/clusterRef/updateKeyRef within the record's namespace), so
+# these sit alongside the zone in bind-internal, not in the app namespace.
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  # "internal" in the name distinguishes this from the external DNS that
+  # Authentik will manage its own records from later.
+  name: identity-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: identity
+  type: A
+  ttl: 600
+  values:
+    # traefik-EXTERNAL (DMZ) gateway VIP; the authentik Gateway serves the
+    # identity.unkin.net hostname there.
+    - 198.18.199.0
+---
+# PRODUCTION CUTOVER RECORD — intentionally commented out.
+# git.unkin.net currently resolves to the LIVE VM forge (HAProxy VRRP VIP
+# 198.18.19.17), which holds every repo the estate depends on. Uncommenting this
+# repoints the whole org's git.unkin.net at the new k8s Gitea gateway VIP, so it
+# is the FINAL step of the forge migration — gated on the data migration (gitea
+# dump/restore + SECRET_KEY copy) in argocd-apps docs/gitea-migration.md.
+# NOTE: the live git.unkin.net answer is served by the puppet DNS master today
+# (profiles::dns::master, records from PuppetDB); this k8s apex zone holds only
+# SOA+NS + a few DNSRecords so far. Confirm the k8s bind cluster is the live
+# authority for unkin.net (or update the puppet record instead) before relying
+# on this CR at cutover.
+# ---
+# apiVersion: bind.unkin.net/v1alpha1
+# kind: DNSRecord
+# metadata:
+#   name: git-dns-internal
+#   namespace: bind-internal
+# spec:
+#   zoneRef: unkin-net
+#   name: git
+#   type: A
+#   ttl: 600
+#   values:
+#     # traefik-internal gateway VIP; the gitea Gateway serves git.unkin.net there.
+#     - 198.18.200.4
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: s3-ceph-cname
+  namespace: bind-internal
+spec:
+  zoneRef: ceph-unkin-net
+  name: s3
+  type: CNAME
+  ttl: 600
+  values:
+    # radosgw S3 endpoint. Points at the Consul service for now; the real
+    # target will be changed later.
+    - radosgw.service.consul.
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: dashboard-ceph-cname
+  namespace: bind-internal
+spec:
+  zoneRef: ceph-unkin-net
+  name: dashboard
+  type: CNAME
+  ttl: 600
+  values:
+    # Ceph mgr dashboard, reached via lb1. Lets in-cluster clients (the
+    # cephrgw-operator) resolve dashboard.ceph.unkin.net.
+    - lb1.unkin.net.
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: lb1-unkin-net
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: lb1
+  type: A
+  ttl: 600
+  values:
+    - 103.216.191.185
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: ghp-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: ghp
+  type: A
+  ttl: 600
+  values:
+    # traefik-internal gateway VIP; the ghp Gateway serves ghp.unkin.net there.
+    - 198.18.200.4
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: arrstack-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: arrstack
+  type: A
+  ttl: 600
+  values:
+    # traefik-EXTERNAL (DMZ) gateway VIP; the arrproxy Gateway serves the
+    # arrstack.unkin.net front door (oauth2-proxy) there.
+    - 198.18.199.0
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: logviewer-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: logviewer
+  type: A
+  ttl: 600
+  values:
+    # traefik-internal gateway VIP; the logviewer Gateway serves
+    # logviewer.unkin.net there.
+    - 198.18.200.4
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: cheeztv-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: cheeztv
+  type: A
+  ttl: 600
+  values:
+    # traefik-internal gateway VIP; the cheeztv Gateway serves cheeztv.unkin.net
+    # there.
+    - 198.18.200.4
+---
+apiVersion: bind.unkin.net/v1alpha1
+kind: DNSRecord
+metadata:
+  name: watchstate-dns-internal
+  namespace: bind-internal
+spec:
+  zoneRef: unkin-net
+  name: watchstate
+  type: A
+  ttl: 600
+  values:
+    # traefik-EXTERNAL (DMZ) gateway VIP; the watchstate-external Gateway serves
+    # the watchstate.unkin.net front door (oauth2-proxy) there.
+    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/arrstack.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: arrstack-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: arrstack
-  type: A
-  ttl: 600
-  values:
-    # traefik-EXTERNAL (DMZ) gateway VIP; the arrproxy Gateway serves the
-    # arrstack.unkin.net front door (oauth2-proxy) there.
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/auth.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: auth-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: auth
-  type: A
-  ttl: 600
-  values:
-    # traefik-internal gateway VIP; the kanidm Gateway serves auth.unkin.net there.
-    - 198.18.200.4

apps/base/bind-internal/authoritative/unkin-net/a/cheeztv.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: cheeztv-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: cheeztv
-  type: A
-  ttl: 600
-  values:
-    # traefik-internal gateway VIP; the cheeztv Gateway serves cheeztv.unkin.net
-    # there.
-    - 198.18.200.4

apps/base/bind-internal/authoritative/unkin-net/a/fafflix.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: fafflix-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: fafflix
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/ghp.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: ghp-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: ghp
-  type: A
-  ttl: 600
-  values:
-    # traefik-internal gateway VIP; the ghp Gateway serves ghp.unkin.net there.
-    - 198.18.200.4

apps/base/bind-internal/authoritative/unkin-net/a/git-edge.yaml

@@ -1,14 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: git-edge-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: git
-  type: A
-  ttl: 600
-  values:
-    # traefik-external haproxy edge Gateway VIP (TLS passthrough).
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/git.yaml

@@ -1,23 +0,0 @@
----
-# PRODUCTION CUTOVER RECORD — intentionally commented out.
-# git.unkin.net currently resolves to the haproxy edge (git-edge.yaml), which
-# forwards to the LIVE VM forge holding every repo. Uncommenting this
-# repoints the whole org's git.unkin.net at the new k8s Gitea gateway VIP, so it
-# is the FINAL step of the forge migration — gated on the data migration (gitea
-# dump/restore + SECRET_KEY copy) in argocd-apps docs/gitea-migration.md.
-# Uncomment this record AND its entry in kustomization.yaml (removing
-# git-edge.yaml there) to activate it.
-# ---
-# apiVersion: bind.unkin.net/v1alpha1
-# kind: DNSRecord
-# metadata:
-#   name: git-dns-internal
-#   namespace: bind-internal
-# spec:
-#   zoneRef: unkin-net
-#   name: git
-#   type: A
-#   ttl: 600
-#   values:
-#     # traefik-internal gateway VIP; the gitea Gateway serves git.unkin.net there.
-#     - 198.18.200.4

apps/base/bind-internal/authoritative/unkin-net/a/grafana.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: grafana-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: grafana
-  type: A
-  ttl: 600
-  values:
-    # traefik-EXTERNAL (DMZ) gateway VIP; the grafana-external Gateway serves
-    # grafana.unkin.net there.
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/identity.yaml

@@ -1,17 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  # "internal" in the name distinguishes this from the external DNS that
-  # Authentik will manage its own records from later.
-  name: identity-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: identity
-  type: A
-  ttl: 600
-  values:
-    # traefik-EXTERNAL (DMZ) gateway VIP; the authentik Gateway serves the
-    # identity.unkin.net hostname there.
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/kustomization.yaml

@@ -1,20 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - arrstack.yaml
-  - auth.yaml
-  - cheeztv.yaml
-  - fafflix.yaml
-  # PRODUCTION CUTOVER RECORD — see git.yaml. Uncomment together with the
-  # record itself, and remove git-edge.yaml.
-  # - git.yaml
-  - ghp.yaml
-  - git-edge.yaml
-  - grafana.yaml
-  - identity.yaml
-  - lb1.yaml
-  - logviewer.yaml
-  - vlogs.yaml
-  - watchstate.yaml

apps/base/bind-internal/authoritative/unkin-net/a/lb1.yaml

@@ -1,13 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: lb1-unkin-net
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: lb1
-  type: A
-  ttl: 600
-  values:
-    - 103.216.191.185

apps/base/bind-internal/authoritative/unkin-net/a/logviewer.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: logviewer-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: logviewer
-  type: A
-  ttl: 600
-  values:
-    # traefik-internal gateway VIP; the logviewer Gateway serves
-    # logviewer.unkin.net there.
-    - 198.18.200.4

apps/base/bind-internal/authoritative/unkin-net/a/vlogs.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: vlogs-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: vlogs
-  type: A
-  ttl: 600
-  values:
-    # traefik-EXTERNAL (DMZ) gateway VIP; the vlogs-external Gateway serves the
-    # vlogs.unkin.net front door (oauth2-proxy) there.
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/a/watchstate.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: bind.unkin.net/v1alpha1
-kind: DNSRecord
-metadata:
-  name: watchstate-dns-internal
-  namespace: bind-internal
-spec:
-  zoneRef: unkin-net
-  name: watchstate
-  type: A
-  ttl: 600
-  values:
-    # traefik-EXTERNAL (DMZ) gateway VIP; the watchstate-external Gateway serves
-    # the watchstate.unkin.net front door (oauth2-proxy) there.
-    - 198.18.199.0

apps/base/bind-internal/authoritative/unkin-net/kustomization.yaml

@@ -1,6 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - a

apps/base/bind-system/deployment.yaml

@@ -21,7 +21,7 @@ spec:
         runAsNonRoot: true
       containers:
         - name: operator
-          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/bind-operator:v0.3.1
+          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/bind-operator:v0.2.6
           args:
             - --metrics-bind-address=:8080
             - --health-probe-bind-address=:8081

apps/base/bind-system/kustomization.yaml

@@ -6,7 +6,7 @@ resources:
   - namespace.yaml
   # CRDs are pulled from the bind-operator repo at the matching tag rather than
   # vendored here, so they never drift from the operator.
-  - https://git.unkin.net/unkin/bind-operator/raw/tag/v0.3.1/config/crd/install.yaml
+  - https://git.unkin.net/unkin/bind-operator/raw/tag/v0.2.6/config/crd/install.yaml
   - rbac.yaml
   - agent-dns-rbac.yaml
   - deployment.yaml

apps/base/cert-manager/certificate_wildcard-ceph-unkin-net.yaml

@@ -1,26 +0,0 @@
----
-# Let's Encrypt *.ceph.unkin.net wildcard for the haproxy edge (ceph dashboard).
-# DNS-01 needs the delegated _acme-challenge.ceph.unkin.net CNAME in the public
-# unkin.net zone.
-# _acme-challenge.ceph.unkin.net. CNAME _acme-challenge.ceph.acme.unkin.net.
-apiVersion: cert-manager.io/v1
-kind: Certificate
-metadata:
-  name: wildcard-ceph-unkin-net
-  namespace: cert-manager
-spec:
-  secretName: wildcard-ceph-unkin-net-tls
-  secretTemplate:
-    annotations:
-      reflector.v1.k8s.emberstack.com/reflection-allowed: "true"
-      reflector.v1.k8s.emberstack.com/reflection-allowed-namespaces: "haproxy"
-      reflector.v1.k8s.emberstack.com/reflection-auto-enabled: "true"
-      reflector.v1.k8s.emberstack.com/reflection-auto-namespaces: "haproxy"
-  privateKey:
-    size: 4096
-  dnsNames:
-    - "*.ceph.unkin.net"
-  issuerRef:
-    name: letsencrypt
-    kind: ClusterIssuer
-    group: cert-manager.io

apps/base/cert-manager/certificate_wildcard-main-unkin-net.yaml

@@ -1,26 +0,0 @@
----
-# Let's Encrypt *.main.unkin.net wildcard for the haproxy edge (pve, arr stack,
-# jellyfin, stalwart webadmin/autoconfig). DNS-01 needs the delegated
-# _acme-challenge.main.unkin.net CNAME in the public unkin.net zone.
-# _acme-challenge.main.unkin.net. CNAME _acme-challenge.main.acme.unkin.net.
-apiVersion: cert-manager.io/v1
-kind: Certificate
-metadata:
-  name: wildcard-main-unkin-net
-  namespace: cert-manager
-spec:
-  secretName: wildcard-main-unkin-net-tls
-  secretTemplate:
-    annotations:
-      reflector.v1.k8s.emberstack.com/reflection-allowed: "true"
-      reflector.v1.k8s.emberstack.com/reflection-allowed-namespaces: "haproxy"
-      reflector.v1.k8s.emberstack.com/reflection-auto-enabled: "true"
-      reflector.v1.k8s.emberstack.com/reflection-auto-namespaces: "haproxy"
-  privateKey:
-    size: 4096
-  dnsNames:
-    - "*.main.unkin.net"
-  issuerRef:
-    name: letsencrypt
-    kind: ClusterIssuer
-    group: cert-manager.io

apps/base/cert-manager/certificate_wildcard-unkin-net.yaml

@@ -14,9 +14,9 @@ spec:
   secretTemplate:
     annotations:
       reflector.v1.k8s.emberstack.com/reflection-allowed: "true"
-      reflector.v1.k8s.emberstack.com/reflection-allowed-namespaces: "cheeztv,arrstack,authentik,gitea,watchstate,mediamark,repospawner,haproxy,vlogs,grafana"
+      reflector.v1.k8s.emberstack.com/reflection-allowed-namespaces: "cheeztv,arrstack,authentik,gitea,watchstate,mediamark,repospawner"
       reflector.v1.k8s.emberstack.com/reflection-auto-enabled: "true"
-      reflector.v1.k8s.emberstack.com/reflection-auto-namespaces: "cheeztv,arrstack,authentik,gitea,watchstate,mediamark,repospawner,haproxy,vlogs,grafana"
+      reflector.v1.k8s.emberstack.com/reflection-auto-namespaces: "cheeztv,arrstack,authentik,gitea,watchstate,mediamark,repospawner"
   privateKey:
     size: 4096
   dnsNames:

apps/base/cert-manager/kustomization.yaml

@@ -12,5 +12,3 @@ resources:
   - clusterissuer_letsencrypt.yaml
   - clusterissuer_letsencrypt-staging.yaml
   - certificate_wildcard-unkin-net.yaml
-  - certificate_wildcard-main-unkin-net.yaml
-  - certificate_wildcard-ceph-unkin-net.yaml

apps/base/cheeztv/plugin-configmap.yaml

@@ -26,7 +26,7 @@ data:
           </key>
           <value>
             <PluginConfiguration>
-              <OidEndpoint>https://identity.unkin.net/application/o/jellyfin/</OidEndpoint>
+              <OidEndpoint>https://identity.k8s.syd1.au.unkin.net/application/o/jellyfin/</OidEndpoint>
               <OidClientId>jellyfin</OidClientId>
               <OidSecret>@@CLIENT_SECRET@@</OidSecret>
               <Enabled>true</Enabled>

apps/base/cheeztv/service.yaml

@@ -13,4 +13,6 @@ spec:
       targetPort: http
   selector:
     app: cheeztv
+  # Pin each client to one replica to reduce transcode-session churn/takeover.
+  sessionAffinity: ClientIP
   type: ClusterIP

apps/base/cheeztv/statefulset.yaml

@@ -4,8 +4,6 @@ kind: StatefulSet
 metadata:
   name: cheeztv
   namespace: cheeztv
-  annotations:
-    configmap.reloader.stakater.com/auto: "true"
 spec:
   # HA: two replicas coordinate transcode session ownership through Valkey and
   # resume each other's HLS segments off the shared RWX transcode PVC. Stable
@@ -164,7 +162,7 @@ spec:
               readOnly: true
       containers:
         - name: cheeztv
-          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/jellyfin-ha:v0.4.0
+          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/jellyfin-ha:v0.2.0
           imagePullPolicy: IfNotPresent
           ports:
             - name: http

apps/base/clickhouse-system/kustomization.yaml

@@ -3,4 +3,4 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 
 resources:
-  - apex.yaml
+  - namespace.yaml

apps/base/clickhouse-system/namespace.yaml

@@ -0,0 +1,7 @@
+---
+apiVersion: v1
+kind: Namespace
+metadata:
+  labels:
+    app.kubernetes.io/name: clickhouse-system
+  name: clickhouse-system

apps/base/dhcp-system/deployment.yaml

@@ -21,7 +21,7 @@ spec:
         runAsNonRoot: true
       containers:
         - name: operator
-          image: git.unkin.net/unkin/kea-operator:v0.1.6
+          image: git.unkin.net/unkin/kea-operator:v0.1.5
           args:
             - --metrics-bind-address=:8080
             - --health-probe-bind-address=:8081

apps/base/fafflix/plugin-configmap.yaml

@@ -26,7 +26,7 @@ data:
           </key>
           <value>
             <PluginConfiguration>
-              <OidEndpoint>https://identity.unkin.net/application/o/jellyfin/</OidEndpoint>
+              <OidEndpoint>https://identity.k8s.syd1.au.unkin.net/application/o/jellyfin/</OidEndpoint>
               <OidClientId>jellyfin</OidClientId>
               <OidSecret>@@CLIENT_SECRET@@</OidSecret>
               <Enabled>true</Enabled>

apps/base/fafflix/service.yaml

@@ -13,4 +13,6 @@ spec:
       targetPort: http
   selector:
     app: fafflix
+  # Pin each client to one replica to reduce transcode-session churn/takeover.
+  sessionAffinity: ClientIP
   type: ClusterIP

apps/base/fafflix/statefulset.yaml

@@ -4,8 +4,6 @@ kind: StatefulSet
 metadata:
   name: fafflix
   namespace: fafflix
-  annotations:
-    configmap.reloader.stakater.com/auto: "true"
 spec:
   # HA: two replicas coordinate transcode session ownership through Valkey and
   # resume each other's HLS segments off the shared RWX transcode PVC. Stable
@@ -164,7 +162,7 @@ spec:
               readOnly: true
       containers:
         - name: fafflix
-          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/jellyfin-ha:v0.4.0
+          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/jellyfin-ha:v0.2.0
           imagePullPolicy: IfNotPresent
           ports:
             - name: http

apps/base/gitea/gateway.yaml

@@ -1,6 +1,11 @@
 ---
-# HTTPS front for the k8s Gitea on git.k8s.syd1.au.unkin.net (external-dns
-# k8s.syd1 zone). git.unkin.net stays on the VM haproxy until cutover.
+# HTTPS front for the k8s Gitea, served on two names:
+#   git.unkin.net              — canonical/production (apex, bind-operator zone;
+#                                DNS flip is the gated cutover step, see the doc)
+#   git.k8s.syd1.au.unkin.net  — admin/backup route (external-dns k8s.syd1 zone),
+#                                same dual-name pattern as identity.unkin.net.
+# The cert-manager Certificate (vault-issuer) takes CN git.unkin.net and gets a
+# DNS SAN for each TLS listener hostname automatically.
 apiVersion: gateway.networking.k8s.io/v1
 kind: Gateway
 metadata:
@@ -12,13 +17,36 @@ metadata:
     traefik.io/instance: internal
   annotations:
     cert-manager.io/cluster-issuer: vault-issuer
-    cert-manager.io/common-name: git.k8s.syd1.au.unkin.net
+    cert-manager.io/common-name: git.unkin.net
     cert-manager.io/private-key-size: "4096"
+    # Only the k8s admin route is published by external-dns (it owns just the
+    # k8s.syd1.au.unkin.net zone). git.unkin.net lives in the apex zone and is
+    # flipped at cutover — NOT managed here.
     external-dns.alpha.kubernetes.io/hostname: git.k8s.syd1.au.unkin.net
     external-dns.alpha.kubernetes.io/target: 198.18.200.4
 spec:
   gatewayClassName: traefik-internal
   listeners:
+    - name: http-primary
+      port: 80
+      protocol: HTTP
+      hostname: git.unkin.net
+      allowedRoutes:
+        namespaces:
+          from: Same
+    - name: https-primary
+      port: 443
+      protocol: HTTPS
+      hostname: git.unkin.net
+      allowedRoutes:
+        namespaces:
+          from: Same
+      tls:
+        mode: Terminate
+        certificateRefs:
+          - group: ""
+            kind: Secret
+            name: gitea-tls
     - name: http-admin
       port: 80
       protocol: HTTP

apps/base/gitea/httproute.yaml

@@ -9,8 +9,13 @@ metadata:
     app.kubernetes.io/instance: gitea
 spec:
   hostnames:
+    - git.unkin.net
     - git.k8s.syd1.au.unkin.net
   parentRefs:
+    - group: gateway.networking.k8s.io
+      kind: Gateway
+      name: gitea
+      sectionName: http-primary
     - group: gateway.networking.k8s.io
       kind: Gateway
       name: gitea
@@ -36,8 +41,13 @@ metadata:
     app.kubernetes.io/instance: gitea
 spec:
   hostnames:
+    - git.unkin.net
     - git.k8s.syd1.au.unkin.net
   parentRefs:
+    - group: gateway.networking.k8s.io
+      kind: Gateway
+      name: gitea
+      sectionName: https-primary
     - group: gateway.networking.k8s.io
       kind: Gateway
       name: gitea

apps/base/grafana/dashboards/k8up.yaml

@@ -1,15 +0,0 @@
----
-# k8up dashboard (M0NsTeRRR/grafana-dashboards), datasource resolved to the
-# in-cluster VictoriaMetrics. Backed by the k8up VMServiceScrape.
-apiVersion: grafana.integreatly.org/v1beta1
-kind: GrafanaDashboard
-metadata:
-  name: k8up
-  namespace: grafana
-spec:
-  instanceSelector:
-    matchLabels:
-      dashboards: "grafana"
-  resyncPeriod: 5m
-  allowCrossNamespaceImport: false
-  gzipJson: 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

apps/base/grafana/gateway.yaml

@@ -37,36 +37,3 @@ spec:
           - group: ""
             kind: Secret
             name: grafana-tls
----
-# Public grafana.unkin.net via the external Traefik; TLS uses the reflected
-# Let's Encrypt *.unkin.net wildcard, DNS lives in the bind-operator zone.
-apiVersion: gateway.networking.k8s.io/v1
-kind: Gateway
-metadata:
-  labels:
-    traefik.io/instance: external
-  name: grafana-external
-  namespace: grafana
-spec:
-  gatewayClassName: traefik-external
-  listeners:
-    - allowedRoutes:
-        namespaces:
-          from: Same
-      hostname: grafana.unkin.net
-      name: http
-      port: 80
-      protocol: HTTP
-    - allowedRoutes:
-        namespaces:
-          from: Same
-      hostname: grafana.unkin.net
-      name: https
-      port: 443
-      protocol: HTTPS
-      tls:
-        certificateRefs:
-          - group: ""
-            kind: Secret
-            name: wildcard-unkin-net-tls
-        mode: Terminate

apps/base/grafana/grafana.yaml

@@ -9,16 +9,8 @@ metadata:
 spec:
   deployment:
     spec:
-      replicas: 3
       template:
         spec:
-          topologySpreadConstraints:
-            - maxSkew: 1
-              topologyKey: kubernetes.io/hostname
-              whenUnsatisfiable: ScheduleAnyway
-              labelSelector:
-                matchLabels:
-                  app: grafana
           containers:
             - name: grafana
               env:
@@ -39,11 +31,11 @@ spec:
                   cpu: 100m
                   memory: 256Mi
                 limits:
-                  cpu: "2"
-                  memory: 4Gi
+                  cpu: "1"
+                  memory: 1Gi
   config:
     server:
-      root_url: "https://grafana.unkin.net"
+      root_url: "https://grafana.k8s.syd1.au.unkin.net"
     database:
       type: "postgres"
       host: "postgres-pooler-rw.grafana.svc.cluster.local:5432"
@@ -69,8 +61,3 @@ spec:
       # to akR-global-admin members (and direct members) via terraform-authentik.
       role_attribute_path: "contains(ak_groups[*], 'akP-grafana-admin') && 'Admin' || 'Viewer'"
       role_attribute_strict: "false"
-    # Gossip between replicas so each alert notification is sent once.
-    unified_alerting:
-      ha_peers: "grafana-alerting.grafana.svc.cluster.local:9094"
-      ha_listen_address: "${POD_IP}:9094"
-      ha_advertise_address: "${POD_IP}:9094"

apps/base/grafana/grafanadatasource.yaml

@@ -20,22 +20,3 @@ spec:
     jsonData:
       timeInterval: "15s"
       httpMethod: "POST"
----
-apiVersion: grafana.integreatly.org/v1beta1
-kind: GrafanaDatasource
-metadata:
-  name: victorialogs
-  namespace: grafana
-spec:
-  instanceSelector:
-    matchLabels:
-      dashboards: "grafana"
-  plugins:
-    - name: victoriametrics-logs-datasource
-      version: 0.32.0
-  datasource:
-    name: "VictoriaLogs"
-    type: "victoriametrics-logs-datasource"
-    uid: "victorialogs"
-    access: "proxy"
-    url: "http://vlselect-logs.logging.svc.cluster.local:9471"

apps/base/grafana/httproute.yaml

@@ -53,58 +53,3 @@ spec:
         - path:
             type: PathPrefix
             value: /
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: HTTPRoute
-metadata:
-  name: grafana-external-http-redirect
-  namespace: grafana
-  labels:
-    app.kubernetes.io/name: grafana
-    app.kubernetes.io/instance: grafana
-spec:
-  hostnames:
-    - grafana.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: grafana-external
-      sectionName: http
-  rules:
-    - filters:
-        - type: RequestRedirect
-          requestRedirect:
-            scheme: https
-            statusCode: 301
-      matches:
-        - path:
-            type: PathPrefix
-            value: /
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: HTTPRoute
-metadata:
-  name: grafana-external
-  namespace: grafana
-  labels:
-    app.kubernetes.io/name: grafana
-    app.kubernetes.io/instance: grafana
-spec:
-  hostnames:
-    - grafana.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: grafana-external
-      sectionName: https
-  rules:
-    - backendRefs:
-        - group: ""
-          kind: Service
-          name: grafana-service
-          port: 3000
-          weight: 1
-      matches:
-        - path:
-            type: PathPrefix
-            value: /

apps/base/grafana/kustomization.yaml

@@ -18,7 +18,6 @@ resources:
   - dashboards/frr-ospf-route-metrics.yaml
   - dashboards/gitea.yaml
   - dashboards/haproxy.yaml
-  - dashboards/k8up.yaml
   - dashboards/media-dashboard.yaml
   - dashboards/node-exporter-full.yaml
   - dashboards/nzbget.yaml

apps/base/haproxy/configmap.yaml

@@ -1,256 +0,0 @@
----
-apiVersion: v1
-kind: ConfigMap
-metadata:
-  name: haproxy-config
-  namespace: haproxy
-data:
-  certificate.list: |
-    # First entry is the default cert for non-matching SNI.
-    /etc/haproxy/certs/unkin-net/tls.crt
-    /etc/haproxy/certs/main-unkin-net/tls.crt
-    /etc/haproxy/certs/ceph-unkin-net/tls.crt
-
-  fe_https.map: |
-    sonarr.main.unkin.net be_sonarr
-    radarr.main.unkin.net be_radarr
-    lidarr.main.unkin.net be_lidarr
-    readarr.main.unkin.net be_readarr
-    prowlarr.main.unkin.net be_prowlarr
-    nzbget.main.unkin.net be_nzbget
-    jellyfin.main.unkin.net be_jellyfin
-    fafflix.unkin.net be_jellyfin
-    git.unkin.net be_gitea
-    dashboard.ceph.unkin.net be_ceph_dashboard
-    auth.unkin.net be_k8s_kanidm
-
-  haproxy.cfg: |
-    global
-      log stdout format raw local0
-      log stdout format raw local1 notice
-      maxconn  4000
-      hard-stop-after  2m
-      ssl-default-bind-ciphers  EECDH+AESGCM:EDH+AESGCM:AES256+EECDH:AES256+EDH
-      ssl-default-bind-options  ssl-min-ver TLSv1.2 ssl-max-ver TLSv1.3
-      ssl-default-server-ciphers  kEECDH+aRSA+AES:kRSA+AES:+AES256:RC4-SHA:!kEDH:!LOW:!EXP:!MD5:!aNULL:!eNULL
-      ssl-default-server-options  no-sslv3
-      stats  timeout 30s
-      stats  socket /var/lib/haproxy/stats
-      stats  socket /var/lib/haproxy/admin.sock mode 660 level admin
-      tune.ssl.default-dh-param  2048
-
-    defaults
-      log  global
-      maxconn  5000
-      mode  http
-      option  httplog
-      option  dontlognull
-      option  http-server-close
-      option  forwardfor except 127.0.0.0/8
-      option  redispatch
-      retries  3
-      stats  enable
-      timeout  http-request 10s
-      timeout  queue 1m
-      timeout  connect 10s
-      timeout  client 5m
-      timeout  server 5m
-      timeout  http-keep-alive 10s
-      timeout  check 10s
-
-    frontend fe_https
-      bind 0.0.0.0:443 ssl crt-list /usr/local/etc/haproxy/certificate.list ciphers EECDH+AESGCM:EDH+AESGCM:AES256+EECDH:AES256+EDH force-tlsv12
-      mode http
-      description Global HTTPS Frontend
-      http-request set-header X-Forwarded-Proto https
-      http-request set-header X-Real-IP %[src]
-      http-response set-header X-Content-Type-Options nosniff
-      http-response set-header X-XSS-Protection 1;mode=block
-      use_backend %[req.hdr(host),lower,map(/usr/local/etc/haproxy/fe_https.map,be_default)]
-
-    frontend fe_metrics
-      bind 0.0.0.0:8405
-      mode http
-      description Metrics Frontend
-      http-request set-header X-Forwarded-Proto https
-      http-request set-header X-Real-IP %[src]
-      http-request use-service prometheus-exporter if { path /metrics }
-
-    backend be_ceph_dashboard
-      description Backend for Ceph Dashboard from Mgr instances
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-check expect status 200
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 9443 }
-      http-reuse always
-      option httpchk GET /
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      stick-table type ip size 200k expire 30m
-      server prodnxsr0009 198.18.23.9:9443 check cookie prodnxsr0009 fall 2 inter 2s rise 3 ssl verify none
-      server prodnxsr0010 198.18.23.10:9443 check cookie prodnxsr0010 fall 2 inter 2s rise 3 ssl verify none
-      server prodnxsr0011 198.18.23.11:9443 check cookie prodnxsr0011 fall 2 inter 2s rise 3 ssl verify none
-      server prodnxsr0012 198.18.23.12:9443 check cookie prodnxsr0012 fall 2 inter 2s rise 3 ssl verify none
-      server prodnxsr0013 198.18.23.13:9443 check cookie prodnxsr0013 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_default
-      description Backend for unmatched HTTP traffic
-      balance roundrobin
-      cookie SRVNAME insert
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      option httpchk GET /
-      option forwardfor
-
-    backend be_gitea
-      description Backend for gitea cluster
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      stick on src
-      stick-table type ip size 200k expire 30m
-      server ausyd1nxvm2080 198.18.26.18:443 check cookie ausyd1nxvm2080 fall 2 inter 2s rise 3 ssl verify none
-      server ausyd1nxvm2081 198.18.27.117:443 check cookie ausyd1nxvm2081 fall 2 inter 2s rise 3 ssl verify none
-      server ausyd1nxvm2082 198.18.28.71:443 check cookie ausyd1nxvm2082 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_jellyfin
-      description Backend for au-syd1 jellyfin
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2051 198.18.25.164:443 check cookie ausyd1nxvm2051 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_k8s_kanidm
-      description Backend for Kanidm (auth.unkin.net via Kubernetes internal Traefik)
-      balance roundrobin
-      http-reuse always
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      redirect scheme https if !{ ssl_fc }
-      option httpchk
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      http-check connect ssl sni auth.unkin.net
-      http-check send meth GET uri /status ver HTTP/1.1 hdr Host auth.unkin.net
-      http-check expect status 200
-      server k8s-traefik-internal 198.18.200.4:443 ssl verify none check inter 2s rise 3 fall 2 sni str(auth.unkin.net)
-
-    backend be_lidarr
-      description Backend for au-syd1 lidarr
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2048 198.18.28.165:443 check cookie ausyd1nxvm2048 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_nzbget
-      description Backend for au-syd1 nzbget
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2045 198.18.25.44:443 check cookie ausyd1nxvm2045 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_prowlarr
-      description Backend for au-syd1 prowlarr
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2050 198.18.25.66:443 check cookie ausyd1nxvm2050 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_radarr
-      description Backend for au-syd1 radarr
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2047 198.18.27.131:443 check cookie ausyd1nxvm2047 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_readarr
-      description Backend for au-syd1 readarr
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2049 198.18.29.32:443 check cookie ausyd1nxvm2049 fall 2 inter 2s rise 3 ssl verify none
-
-    backend be_sonarr
-      description Backend for au-syd1 sonarr
-      balance roundrobin
-      cookie SRVNAME insert indirect nocache
-      http-request set-header X-Forwarded-Port %[dst_port]
-      http-request add-header X-Forwarded-Proto https if { dst_port 443 }
-      http-reuse always
-      option httpchk GET /consul/health
-      option forwardfor
-      option http-keep-alive
-      option prefer-last-server
-      redirect scheme https if !{ ssl_fc }
-      server ausyd1nxvm2046 198.18.26.161:443 check cookie ausyd1nxvm2046 fall 2 inter 2s rise 3 ssl verify none
-
-    # The `peers au-syd1-prod` section is dropped: peer names must be static and a
-    # Deployment cannot provide them. Behind the external Traefik's TLS
-    # passthrough `src` is a Traefik pod, so X-Real-IP, forwardfor and the
-    # `stick on src` tables all key on that; the SRVNAME cookie carries real
-    # session persistence. Traefik cannot emit PROXY protocol to a TLSRoute
-    # backend, so there is nothing to bind `accept-proxy` to.
-
-    listen health
-      bind 0.0.0.0:8404
-      mode http
-      monitor-uri /healthz
-
-    listen stats
-      bind 127.0.0.1:9090
-      mode http
-      stats uri /
-      stats auth admin:admin

apps/base/haproxy/deployment.yaml

@@ -1,148 +0,0 @@
----
-apiVersion: apps/v1
-kind: Deployment
-metadata:
-  name: haproxy
-  namespace: haproxy
-  annotations:
-    reloader.stakater.com/auto: "true"
-spec:
-  replicas: 3
-  selector:
-    matchLabels:
-      app: haproxy
-  strategy:
-    type: RollingUpdate
-    rollingUpdate:
-      maxUnavailable: 1
-  template:
-    metadata:
-      labels:
-        app: haproxy
-    spec:
-      automountServiceAccountToken: false
-      terminationGracePeriodSeconds: 150
-      affinity:
-        podAntiAffinity:
-          requiredDuringSchedulingIgnoredDuringExecution:
-            - labelSelector:
-                matchLabels:
-                  app: haproxy
-              topologyKey: kubernetes.io/hostname
-      securityContext:
-        runAsNonRoot: true
-        runAsUser: 99
-        runAsGroup: 99
-        seccompProfile:
-          type: RuntimeDefault
-      containers:
-        - name: haproxy
-          image: haproxy:3.2.24-alpine
-          imagePullPolicy: IfNotPresent
-          command:
-            - haproxy
-            - -W
-            - -db
-            - -f
-            - /usr/local/etc/haproxy/haproxy.cfg
-          securityContext:
-            allowPrivilegeEscalation: false
-            readOnlyRootFilesystem: true
-            capabilities:
-              drop: [ALL]
-              # fe_https binds the privileged port 443 as uid 99, and the
-              # dst_port ACLs need the real port.
-              add: [NET_BIND_SERVICE]
-          ports:
-            - name: https
-              containerPort: 443
-              protocol: TCP
-            - name: health
-              containerPort: 8404
-              protocol: TCP
-            - name: metrics
-              containerPort: 8405
-              protocol: TCP
-            - name: stats
-              containerPort: 9090
-              protocol: TCP
-          lifecycle:
-            preStop:
-              exec:
-                # SIGUSR1 to the master soft-stops the workers; hard-stop-after
-                # caps the drain. Wait so kubelet holds SIGTERM until it is done.
-                command:
-                  - /bin/sh
-                  - -c
-                  - kill -s USR1 1; while kill -0 1 2>/dev/null; do sleep 1; done
-          livenessProbe:
-            httpGet:
-              path: /healthz
-              port: health
-            initialDelaySeconds: 15
-            periodSeconds: 30
-            timeoutSeconds: 5
-            failureThreshold: 3
-          readinessProbe:
-            httpGet:
-              path: /healthz
-              port: health
-            initialDelaySeconds: 5
-            periodSeconds: 5
-            timeoutSeconds: 5
-            failureThreshold: 3
-          resources:
-            requests:
-              cpu: 200m
-              memory: 256Mi
-            limits:
-              cpu: 2
-              memory: 1Gi
-          volumeMounts:
-            - name: config
-              mountPath: /usr/local/etc/haproxy
-              readOnly: true
-            - name: cert-unkin-net
-              mountPath: /etc/haproxy/certs/unkin-net
-              readOnly: true
-            - name: cert-main-unkin-net
-              mountPath: /etc/haproxy/certs/main-unkin-net
-              readOnly: true
-            - name: cert-ceph-unkin-net
-              mountPath: /etc/haproxy/certs/ceph-unkin-net
-              readOnly: true
-            - name: run
-              mountPath: /var/lib/haproxy
-      volumes:
-        - name: config
-          configMap:
-            name: haproxy-config
-        # ssl-load-extra-files loads <crtfile>.key by default, so the key is
-        # projected next to the cert as tls.crt.key.
-        - name: cert-unkin-net
-          secret:
-            secretName: wildcard-unkin-net-tls
-            items:
-              - key: tls.crt
-                path: tls.crt
-              - key: tls.key
-                path: tls.crt.key
-        - name: cert-main-unkin-net
-          secret:
-            secretName: wildcard-main-unkin-net-tls
-            items:
-              - key: tls.crt
-                path: tls.crt
-              - key: tls.key
-                path: tls.crt.key
-        - name: cert-ceph-unkin-net
-          secret:
-            secretName: wildcard-ceph-unkin-net-tls
-            items:
-              - key: tls.crt
-                path: tls.crt
-              - key: tls.key
-                path: tls.crt.key
-        - name: run
-          emptyDir: {}
-      restartPolicy: Always

apps/base/haproxy/gateway.yaml

@@ -1,31 +0,0 @@
----
-# External (DMZ) front for the haproxy edge on the traefik-external LB VIP
-# 198.18.199.0. The :443 listener is TLS Passthrough: haproxy owns the three
-# wildcard certs and terminates behind Traefik, so there are no certificateRefs
-# here. Listener hostnames are deliberately unset and the routes carry the
-# explicit hostname list instead; allowedRoutes Same keeps other namespaces off
-# these listeners.
-apiVersion: gateway.networking.k8s.io/v1
-kind: Gateway
-metadata:
-  name: haproxy
-  namespace: haproxy
-  labels:
-    traefik.io/instance: external
-spec:
-  gatewayClassName: traefik-external
-  listeners:
-    - name: http
-      port: 80
-      protocol: HTTP
-      allowedRoutes:
-        namespaces:
-          from: Same
-    - name: https-passthrough
-      port: 443
-      protocol: TLS
-      tls:
-        mode: Passthrough
-      allowedRoutes:
-        namespaces:
-          from: Same

apps/base/haproxy/httproute.yaml

@@ -1,36 +0,0 @@
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: HTTPRoute
-metadata:
-  name: haproxy-http-redirect
-  namespace: haproxy
-  labels:
-    app: haproxy
-spec:
-  hostnames:
-    - sonarr.main.unkin.net
-    - radarr.main.unkin.net
-    - lidarr.main.unkin.net
-    - readarr.main.unkin.net
-    - prowlarr.main.unkin.net
-    - nzbget.main.unkin.net
-    - jellyfin.main.unkin.net
-    - fafflix.unkin.net
-    - git.unkin.net
-    - dashboard.ceph.unkin.net
-    - auth.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: haproxy
-      sectionName: http
-  rules:
-    - filters:
-        - type: RequestRedirect
-          requestRedirect:
-            scheme: https
-            statusCode: 301
-      matches:
-        - path:
-            type: PathPrefix
-            value: /

apps/base/haproxy/kustomization.yaml

@@ -1,15 +0,0 @@
----
-apiVersion: kustomize.config.k8s.io/v1beta1
-kind: Kustomization
-
-resources:
-  - namespace.yaml
-  - configmap.yaml
-  - deployment.yaml
-  - service.yaml
-  - gateway.yaml
-  - tlsroute.yaml
-  - httproute.yaml
-  - pdb.yaml
-  - vpa.yaml
-  - vmpodscrape.yaml

apps/base/haproxy/namespace.yaml

@@ -1,5 +0,0 @@
----
-apiVersion: v1
-kind: Namespace
-metadata:
-  name: haproxy

apps/base/haproxy/pdb.yaml

@@ -1,11 +0,0 @@
----
-apiVersion: policy/v1
-kind: PodDisruptionBudget
-metadata:
-  name: haproxy
-  namespace: haproxy
-spec:
-  maxUnavailable: 1
-  selector:
-    matchLabels:
-      app: haproxy

apps/base/haproxy/service.yaml

@@ -1,19 +0,0 @@
----
-apiVersion: v1
-kind: Service
-metadata:
-  name: haproxy
-  namespace: haproxy
-spec:
-  type: ClusterIP
-  # Reached only by the external Traefik's TLS-passthrough TLSRoute, so the
-  # peer address here is a Traefik pod, not the client. sessionAffinity is
-  # deliberately absent: keyed on ClientIP it would pin whole Traefik pods,
-  # not clients. Backend persistence rests on the per-backend SRVNAME cookie.
-  selector:
-    app: haproxy
-  ports:
-    - name: https
-      port: 443
-      protocol: TCP
-      targetPort: https

apps/base/haproxy/tlsroute.yaml

@@ -1,33 +0,0 @@
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: TLSRoute
-metadata:
-  name: haproxy
-  namespace: haproxy
-  labels:
-    app: haproxy
-spec:
-  hostnames:
-    - sonarr.main.unkin.net
-    - radarr.main.unkin.net
-    - lidarr.main.unkin.net
-    - readarr.main.unkin.net
-    - prowlarr.main.unkin.net
-    - nzbget.main.unkin.net
-    - jellyfin.main.unkin.net
-    - fafflix.unkin.net
-    - git.unkin.net
-    - dashboard.ceph.unkin.net
-    - auth.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: haproxy
-      sectionName: https-passthrough
-  rules:
-    - backendRefs:
-        - group: ""
-          kind: Service
-          name: haproxy
-          port: 443
-          weight: 1

apps/base/haproxy/vmpodscrape.yaml

@@ -1,13 +0,0 @@
----
-apiVersion: operator.victoriametrics.com/v1beta1
-kind: VMPodScrape
-metadata:
-  name: haproxy
-  namespace: haproxy
-spec:
-  selector:
-    matchLabels:
-      app: haproxy
-  podMetricsEndpoints:
-    - port: metrics
-      path: /metrics

apps/base/haproxy/vpa.yaml

@@ -1,13 +0,0 @@
----
-apiVersion: autoscaling.k8s.io/v1
-kind: VerticalPodAutoscaler
-metadata:
-  name: haproxy-vpa
-  namespace: haproxy
-spec:
-  targetRef:
-    apiVersion: apps/v1
-    kind: Deployment
-    name: haproxy
-  updatePolicy:
-    updateMode: "Off"

apps/base/k8up-system/kustomization.yaml

@@ -4,4 +4,3 @@ kind: Kustomization
 
 resources:
   - namespace.yaml
-  - vmservicescrape.yaml

apps/base/k8up-system/vmservicescrape.yaml

@@ -1,15 +0,0 @@
----
-# Scrape the k8up operator metrics (k8up-metrics Service, port http :8080).
-apiVersion: operator.victoriametrics.com/v1beta1
-kind: VMServiceScrape
-metadata:
-  name: k8up
-  namespace: k8up-system
-spec:
-  selector:
-    matchLabels:
-      app.kubernetes.io/name: k8up
-      app.kubernetes.io/instance: k8up
-  endpoints:
-    - port: http
-      path: /metrics

apps/base/logging/cephrgw.yaml

@@ -0,0 +1,43 @@
+---
+# S3 bucket (Ceph RGW) for the long-term raw-log archive, provisioned by the
+# in-estate cephrgw-operator. The archiver Vector deployment writes here.
+apiVersion: ceph.unkin.net/v1alpha1
+kind: ObjectStoreUser
+metadata:
+  name: logs-archive-owner
+  namespace: logging
+spec:
+  displayName: "Logging raw-archive bucket owner"
+  maxBuckets: 5
+  quota:
+    enabled: true
+    # 5 TiB soft cap; real retention is enforced RGW-side by a bucket lifecycle
+    # policy (see PR notes) — the operator does not manage lifecycle.
+    maxSizeBytes: 5497558138880
+---
+apiVersion: ceph.unkin.net/v1alpha1
+kind: Bucket
+metadata:
+  name: logs-archive
+  namespace: logging
+spec:
+  bucketName: logs-archive
+  ownerRef: logs-archive-owner
+  versioning: false
+  tags:
+    app: logging
+    purpose: raw-log-archive
+  # Keep the bucket (and its objects) if this CR is ever deleted.
+  retainOnDelete: true
+---
+apiVersion: ceph.unkin.net/v1alpha1
+kind: BucketAccess
+metadata:
+  name: logs-archive-writer
+  namespace: logging
+spec:
+  bucketRef: logs-archive
+  level: read-write
+  # Operator writes AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY (+ RGW_UID,
+  # S3_ENDPOINT, BUCKET_NAME) into this Secret; the archiver consumes it.
+  secretName: logs-archive-s3

apps/base/logging/clickhouseinstallation.yaml

@@ -0,0 +1,97 @@
+---
+apiVersion: clickhouse.altinity.com/v1
+kind: ClickHouseInstallation
+metadata:
+  name: logs
+  namespace: logging
+spec:
+  defaults:
+    templates:
+      dataVolumeClaimTemplate: data-volume
+      serviceTemplate: chi-service
+      podTemplate: clickhouse
+  configuration:
+    users:
+      # Password hash is sourced from the Vault-synced clickhouse-credentials
+      # Secret; the plaintext never lands in git or the ClickHouse config.
+      vector/password_sha256_hex:
+        valueFrom:
+          secretKeyRef:
+            name: clickhouse-credentials
+            key: password_sha256_hex
+      vector/networks/ip:
+        - "::/0"
+      vector/profile: default
+      vector/quota: default
+      # Allow the vector user to create the logs database/table (bootstrap Job)
+      # and to INSERT. Restrict the built-in default user to loopback only.
+      vector/access_management: "1"
+      default/networks/ip:
+        - "127.0.0.1"
+        - "::1"
+      # Read-only user for CLI tools + the logviewer UI. Hash sourced from the
+      # Vault-synced clickhouse-logreader Secret, same mechanism as vector.
+      # Scoped to the logs database only (unlike vector, which bootstraps it).
+      logreader/password_sha256_hex:
+        valueFrom:
+          secretKeyRef:
+            name: clickhouse-logreader
+            key: password_sha256_hex
+      logreader/networks/ip:
+        - "::/0"
+      logreader/profile: readonly
+      logreader/quota: default
+      logreader/allow_databases/database:
+        - "logs"
+    profiles:
+      default/max_memory_usage: "10000000000"
+      default/max_execution_time: "120"
+      readonly/readonly: "2"
+      readonly/max_memory_usage: "10000000000"
+      readonly/max_execution_time: "120"
+    clusters:
+      - name: logs
+        layout:
+          shardsCount: 1
+          replicasCount: 1
+  templates:
+    volumeClaimTemplates:
+      - name: data-volume
+        spec:
+          storageClassName: cephrbd-fast-delete
+          accessModes:
+            - ReadWriteOnce
+          resources:
+            requests:
+              # 3d TTL on logs.raw. At ~130 GiB/day raw, ClickHouse LZ4/ZSTD
+              # (~6x on log text) stores ~20-25 GiB/day => ~60-75 GiB/3d, plus
+              # merge headroom (~2x peak). logs.raw is the only table. 150Gi
+              # gives comfortable headroom; long-term data lives in S3, not here.
+              storage: 150Gi
+    serviceTemplates:
+      - name: chi-service
+        generateName: "clickhouse-{chi}"
+        spec:
+          type: ClusterIP
+          ports:
+            - name: http
+              port: 8123
+            - name: tcp
+              port: 9000
+    podTemplates:
+      - name: clickhouse
+        spec:
+          securityContext:
+            fsGroup: 101
+            runAsUser: 101
+            runAsGroup: 101
+          containers:
+            - name: clickhouse
+              image: docker.io/clickhouse/clickhouse-server:24.8
+              resources:
+                requests:
+                  cpu: 500m
+                  memory: 2Gi
+                limits:
+                  cpu: "2"
+                  memory: 8Gi

apps/base/logging/configmap_logarchiver.yaml

@@ -0,0 +1,40 @@
+---
+# logarchiver non-secret config. Secrets (NATS/S3/ClickHouse creds) and the
+# subject filter come from env; everything else uses the binary's built-in
+# defaults, which already target this stack. ack_wait MUST exceed batch.max_age
+# so unacked messages in an open batch are not redelivered mid-batch.
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: logarchiver-config
+  namespace: logging
+data:
+  config.yaml: |
+    nats:
+      ack_wait: 5m
+      fetch_batch: 512
+    batch:
+      max_bytes: 67108864
+      max_events: 200000
+      max_age: 2m
+    # Pin the proven RGW endpoint/bucket; ignore the secret's S3_ENDPOINT/BUCKET_NAME
+    # (AWS creds still come from the secret env). endpoint_env/bucket_env off.
+    s3:
+      endpoint: "https://s3.ceph.unkin.net"
+      bucket: "logs-archive"
+      region: "us-east-1"
+      path_style: true
+      ca_file: /etc/vault-ca/ca.crt
+      endpoint_env: ""
+      bucket_env: ""
+    crypto:
+      key_name: logarchive
+      pubkey_source: vault
+      vault:
+        address: "https://vault.service.consul:8200"
+        mount: gpg
+        auth_method: kubernetes
+        k8s_mount: k8s/au/syd1
+        k8s_role: logging_logarchiver
+        k8s_jwt_path: /var/run/secrets/vault/token
+        ca_file: /etc/vault-ca/ca.crt

apps/base/logging/deployment_logarchiver.yaml

@@ -0,0 +1,134 @@
+---
+# logarchiver — replaces the vector-archiver leg. Independent JetStream durable
+# consumer (archiver) that seals raw logs to S3 as zstd + OpenPGP objects and
+# indexes each object in ClickHouse (logs.archive_index). Acks only after the
+# object is in S3 AND indexed. Reuses the same NATS/S3/CA wiring the Vector
+# archiver used; the OpenPGP public key is delivered as a mounted file.
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: logarchiver
+  namespace: logging
+  annotations:
+    configmap.reloader.stakater.com/auto: "true"
+    secret.reloader.stakater.com/reload: "vault-ca-cert"
+  labels:
+    app.kubernetes.io/name: logarchiver
+    app.kubernetes.io/component: archiver
+spec:
+  replicas: 1
+  strategy:
+    type: Recreate
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: logarchiver
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: logarchiver
+        vector.dev/exclude: "true"
+    spec:
+      # Dedicated SA whose projected vault-audience token authenticates the
+      # k8s-auth login used to fetch the logarchive public key from the gpg engine.
+      serviceAccountName: logarchiver
+      automountServiceAccountToken: false
+      securityContext:
+        runAsNonRoot: true
+        runAsUser: 65532
+        runAsGroup: 65532
+        seccompProfile:
+          type: RuntimeDefault
+      containers:
+        - name: logarchiver
+          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/logarchiver:v0.1.0
+          imagePullPolicy: IfNotPresent
+          args: ["run"]
+          securityContext:
+            allowPrivilegeEscalation: false
+            readOnlyRootFilesystem: true
+            capabilities:
+              drop:
+                - ALL
+          ports:
+            - containerPort: 9090
+              name: metrics
+              protocol: TCP
+          env:
+            - name: LOGARCHIVER_CONFIG
+              value: /etc/logarchiver/config.yaml
+            # Server-side subject filter; must match the archiver consumer's filter.
+            - name: ARCHIVE_SUBJECTS
+              value: "logs.k8s.vault.>"
+            - name: NATS_CONSUMER_PASSWORD
+              valueFrom:
+                secretKeyRef:
+                  name: nats-auth
+                  key: consumer_password
+            - name: CLICKHOUSE_USER
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-credentials
+                  key: username
+            - name: CLICKHOUSE_PASSWORD
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-credentials
+                  key: password
+          # S3 creds + S3_ENDPOINT + BUCKET_NAME from the cephrgw BucketAccess Secret.
+          envFrom:
+            - secretRef:
+                name: logs-archive-s3
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: metrics
+            initialDelaySeconds: 15
+            periodSeconds: 30
+            timeoutSeconds: 5
+            failureThreshold: 3
+          readinessProbe:
+            httpGet:
+              path: /healthz
+              port: metrics
+            initialDelaySeconds: 5
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          resources:
+            requests:
+              cpu: 100m
+              memory: 256Mi
+            limits:
+              cpu: "1"
+              memory: 1Gi
+          volumeMounts:
+            - name: config
+              mountPath: /etc/logarchiver/config.yaml
+              subPath: config.yaml
+              readOnly: true
+            - name: vault-token
+              mountPath: /var/run/secrets/vault
+              readOnly: true
+            - name: vault-ca-cert
+              mountPath: /etc/vault-ca/ca.crt
+              subPath: ca.crt
+              readOnly: true
+            - name: tmp
+              mountPath: /tmp
+      volumes:
+        - name: config
+          configMap:
+            name: logarchiver-config
+        # Projected SA token with audience "vault" for the gpg-engine k8s login.
+        - name: vault-token
+          projected:
+            sources:
+              - serviceAccountToken:
+                  path: token
+                  audience: vault
+                  expirationSeconds: 600
+        - name: vault-ca-cert
+          secret:
+            secretName: vault-ca-cert
+        - name: tmp
+          emptyDir: {}

apps/base/logging/gateway.yaml

@@ -1,13 +1,16 @@
 ---
-# Log ingestion endpoint for puppet-managed VMs (and any non-k8s client):
-# fronts the VLCluster vlinsert service over TLS at a name VMs can resolve.
+# Log ingestion endpoint for puppet-managed VMs (and any non-k8s client).
+# Reuses the internal Traefik gateway + cert-manager + external-dns pattern so
+# VMs reach the Vector aggregator's HTTP source over TLS at a DNS name they can
+# resolve. The puppet-side Vector rollout ships NDJSON to
+# https://logs-ingest.k8s.syd1.au.unkin.net/ (a later task).
 apiVersion: gateway.networking.k8s.io/v1
 kind: Gateway
 metadata:
   name: logs-ingest
   namespace: logging
   labels:
-    app.kubernetes.io/name: victorialogs
+    app.kubernetes.io/name: vector-aggregator
     app.kubernetes.io/component: ingest
     traefik.io/instance: internal
   annotations:
@@ -39,43 +42,3 @@ spec:
           - group: ""
             kind: Secret
             name: logs-ingest-tls
----
-# Unauthenticated LogsQL query endpoint fronting the VLCluster vlselect service.
-apiVersion: gateway.networking.k8s.io/v1
-kind: Gateway
-metadata:
-  name: vlselect
-  namespace: logging
-  labels:
-    app.kubernetes.io/name: victorialogs
-    app.kubernetes.io/component: query
-    traefik.io/instance: internal
-  annotations:
-    cert-manager.io/cluster-issuer: vault-issuer
-    cert-manager.io/common-name: vlselect.k8s.syd1.au.unkin.net
-    cert-manager.io/private-key-size: "4096"
-    external-dns.alpha.kubernetes.io/hostname: vlselect.k8s.syd1.au.unkin.net
-    external-dns.alpha.kubernetes.io/target: 198.18.200.4
-spec:
-  gatewayClassName: traefik-internal
-  listeners:
-    - name: http
-      port: 80
-      protocol: HTTP
-      hostname: vlselect.k8s.syd1.au.unkin.net
-      allowedRoutes:
-        namespaces:
-          from: Same
-    - name: https
-      port: 443
-      protocol: HTTPS
-      hostname: vlselect.k8s.syd1.au.unkin.net
-      allowedRoutes:
-        namespaces:
-          from: Same
-      tls:
-        mode: Terminate
-        certificateRefs:
-          - group: ""
-            kind: Secret
-            name: vlselect-tls

apps/base/logging/httproute.yaml

@@ -5,7 +5,7 @@ metadata:
   name: logs-ingest-http-redirect
   namespace: logging
   labels:
-    app.kubernetes.io/name: victorialogs
+    app.kubernetes.io/name: vector-aggregator
     app.kubernetes.io/component: ingest
 spec:
   hostnames:
@@ -32,7 +32,7 @@ metadata:
   name: logs-ingest
   namespace: logging
   labels:
-    app.kubernetes.io/name: victorialogs
+    app.kubernetes.io/name: vector-aggregator
     app.kubernetes.io/component: ingest
 spec:
   hostnames:
@@ -46,63 +46,8 @@ spec:
     - backendRefs:
         - group: ""
           kind: Service
-          name: vlinsert-logs
-          port: 9481
-          weight: 1
-      matches:
-        - path:
-            type: PathPrefix
-            value: /
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: HTTPRoute
-metadata:
-  name: vlselect-http-redirect
-  namespace: logging
-  labels:
-    app.kubernetes.io/name: victorialogs
-    app.kubernetes.io/component: query
-spec:
-  hostnames:
-    - vlselect.k8s.syd1.au.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: vlselect
-      sectionName: http
-  rules:
-    - filters:
-        - type: RequestRedirect
-          requestRedirect:
-            scheme: https
-            statusCode: 301
-      matches:
-        - path:
-            type: PathPrefix
-            value: /
----
-apiVersion: gateway.networking.k8s.io/v1
-kind: HTTPRoute
-metadata:
-  name: vlselect
-  namespace: logging
-  labels:
-    app.kubernetes.io/name: victorialogs
-    app.kubernetes.io/component: query
-spec:
-  hostnames:
-    - vlselect.k8s.syd1.au.unkin.net
-  parentRefs:
-    - group: gateway.networking.k8s.io
-      kind: Gateway
-      name: vlselect
-      sectionName: https
-  rules:
-    - backendRefs:
-        - group: ""
-          kind: Service
-          name: vlselect-logs
-          port: 9471
+          name: vector-vm-ingest
+          port: 8080
           weight: 1
       matches:
         - path:

apps/base/logging/job_clickhouse-schema.yaml

@@ -0,0 +1,131 @@
+---
+# Declarative ClickHouse schema bootstrap. Runs as an ArgoCD PostSync hook so it
+# executes after the ClickHouseInstallation is reconciled, and re-runs on every
+# sync (idempotent CREATE ... IF NOT EXISTS). Edit the DDL here to evolve the
+# schema; the Vector aggregator writes to logs.raw with skip_unknown_fields, so
+# adding columns is backward-compatible.
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: clickhouse-schema
+  namespace: logging
+  annotations:
+    argocd.argoproj.io/hook: PostSync
+    argocd.argoproj.io/hook-delete-policy: BeforeHookCreation
+  labels:
+    app.kubernetes.io/name: clickhouse-schema
+    app.kubernetes.io/component: bootstrap
+spec:
+  backoffLimit: 20
+  activeDeadlineSeconds: 1800
+  ttlSecondsAfterFinished: 3600
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: clickhouse-schema
+        vector.dev/exclude: "true"
+    spec:
+      restartPolicy: OnFailure
+      securityContext:
+        runAsNonRoot: true
+        runAsUser: 101
+        runAsGroup: 101
+      containers:
+        - name: clickhouse-schema
+          image: docker.io/clickhouse/clickhouse-server:24.8
+          securityContext:
+            allowPrivilegeEscalation: false
+            readOnlyRootFilesystem: true
+            capabilities:
+              drop:
+                - ALL
+          env:
+            - name: HOME
+              value: /tmp
+            - name: CLICKHOUSE_USER
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-credentials
+                  key: username
+            - name: CLICKHOUSE_PASSWORD
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-credentials
+                  key: password
+          command:
+            - /bin/bash
+            - -ec
+            - |
+              host=clickhouse-logs.logging.svc.cluster.local
+              echo "Waiting for ClickHouse at ${host}:9000 ..."
+              until clickhouse-client --host "$host" --port 9000 \
+                  --user "$CLICKHOUSE_USER" --password "$CLICKHOUSE_PASSWORD" \
+                  --query "SELECT 1" >/dev/null 2>&1; do
+                echo "  not ready, retrying in 5s"; sleep 5
+              done
+              echo "Applying schema ..."
+              clickhouse-client --host "$host" --port 9000 \
+                --user "$CLICKHOUSE_USER" --password "$CLICKHOUSE_PASSWORD" \
+                --multiquery <<'EOSQL'
+              CREATE DATABASE IF NOT EXISTS logs;
+
+              CREATE TABLE IF NOT EXISTS logs.raw
+              (
+                  timestamp  DateTime64(3)           DEFAULT now64(3),
+                  host       LowCardinality(String)  DEFAULT '',
+                  source     LowCardinality(String)  DEFAULT '',
+                  namespace  LowCardinality(String)  DEFAULT '',
+                  pod        String                  DEFAULT '',
+                  container  LowCardinality(String)  DEFAULT '',
+                  stream     LowCardinality(String)  DEFAULT '',
+                  severity   LowCardinality(String)  DEFAULT '',
+                  message    String                  DEFAULT '',
+                  labels     Map(LowCardinality(String), String),
+                  fields     Map(LowCardinality(String), String)
+              )
+              ENGINE = MergeTree
+              PARTITION BY toDate(timestamp)
+              ORDER BY (source, namespace, host, timestamp)
+              TTL toDateTime(timestamp) + INTERVAL 3 DAY
+              SETTINGS index_granularity = 8192;
+
+              -- One row per archived S3 object (written by logarchiver). No TTL:
+              -- the index must outlive logs.raw so the long-term S3 archive stays
+              -- searchable. Keep in sync with logarchiver internal/index/ddl.go.
+              CREATE TABLE IF NOT EXISTS logs.archive_index
+              (
+                  object_key       String,
+                  bucket           LowCardinality(String),
+                  subject          LowCardinality(String),
+                  hosts            Array(LowCardinality(String)),
+                  min_ts           DateTime64(3),
+                  max_ts           DateTime64(3),
+                  event_count      UInt64,
+                  raw_bytes        UInt64,
+                  stored_bytes     UInt64,
+                  compression      LowCardinality(String),
+                  cipher           LowCardinality(String),
+                  container_format LowCardinality(String),
+                  key_name         LowCardinality(String),
+                  key_fingerprint  String,
+                  created_at       DateTime64(3) DEFAULT now64(3),
+                  INDEX idx_hosts hosts TYPE bloom_filter GRANULARITY 1
+              )
+              ENGINE = MergeTree
+              PARTITION BY toYYYYMM(min_ts)
+              ORDER BY (subject, min_ts, object_key);
+              EOSQL
+              echo "Schema applied."
+          resources:
+            requests:
+              cpu: 50m
+              memory: 128Mi
+            limits:
+              cpu: 500m
+              memory: 512Mi
+          volumeMounts:
+            - name: tmp
+              mountPath: /tmp
+      volumes:
+        - name: tmp
+          emptyDir: {}

apps/base/logging/kustomization.yaml

@@ -5,7 +5,46 @@ kind: Kustomization
 resources:
   - namespace.yaml
   - vaultauth.yaml
-  - vlcluster.yaml
-  - vlagent.yaml
+  - vaultstaticsecret.yaml
+  - clickhouseinstallation.yaml
+  - job_clickhouse-schema.yaml
+  - nats-bootstrap-job.yaml
+  - cephrgw.yaml
   - gateway.yaml
   - httproute.yaml
+  - serviceaccount_logarchiver.yaml
+  - configmap_logarchiver.yaml
+  - deployment_logarchiver.yaml
+  - logviewer
+
+# Vector pipelines are the single source of truth (also validated by
+# `vector test` in CI). Mounted into each tier via `existingConfigMaps`.
+configMapGenerator:
+  # Tunable JetStream stream limits (the nats-bootstrap Job reads these and does
+  # create-or-update). Hash suffix is INTENTIONALLY left on: editing a value
+  # renames the ConfigMap, which rewrites the Job's env reference, which changes
+  # the PostSync hook Job's spec and forces Argo to re-run it -> new limits apply.
+  # Sizing assumes ~1500 events/s avg @ ~1 KiB/event with S2 compression (~4x):
+  # ~33 GiB/day compressed -> ~100 GiB/3d per replica. max_bytes 130 GiB sits
+  # under the 180Gi/node PVC (see values-nats.yaml). Raising retention beyond the
+  # PVC requires bumping BOTH max_bytes here and fileStore PVC size in values.
+  - name: nats-stream-limits
+    literals:
+      - max_age=72h
+      - max_bytes=139586437120
+      - dupe_window=2m
+  - name: vector-agent-config
+    files:
+      - agent.yaml=vector/agent.yaml
+    options:
+      disableNameSuffixHash: true
+  - name: vector-aggregator-config
+    files:
+      - aggregator.yaml=vector/aggregator.yaml
+    options:
+      disableNameSuffixHash: true
+  - name: vector-vm-ingest-config
+    files:
+      - vm-ingest.yaml=vector/vm-ingest.yaml
+    options:
+      disableNameSuffixHash: true

apps/base/logging/logviewer/deployment.yaml

@@ -0,0 +1,80 @@
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: logviewer
+  namespace: logging
+  annotations:
+    secret.reloader.stakater.com/reload: "clickhouse-logreader"
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: logviewer
+  strategy:
+    type: Recreate
+  template:
+    metadata:
+      labels:
+        app: logviewer
+    spec:
+      serviceAccountName: default
+      automountServiceAccountToken: false
+      securityContext:
+        runAsNonRoot: true
+        runAsUser: 65532
+        runAsGroup: 65532
+        fsGroup: 65532
+        seccompProfile:
+          type: RuntimeDefault
+      containers:
+        - name: logviewer
+          image: artifactapi.k8s.syd1.au.unkin.net/docker-internal/logviewer:v0.1.0
+          imagePullPolicy: IfNotPresent
+          ports:
+            - containerPort: 8080
+              name: http
+              protocol: TCP
+          env:
+            - name: CH_URL
+              value: http://clickhouse-logs.logging.svc.cluster.local:8123
+            - name: CH_USER
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-logreader
+                  key: username
+            - name: CH_PASSWORD
+              valueFrom:
+                secretKeyRef:
+                  name: clickhouse-logreader
+                  key: password
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: http
+            initialDelaySeconds: 10
+            periodSeconds: 30
+            timeoutSeconds: 5
+            failureThreshold: 3
+          readinessProbe:
+            httpGet:
+              path: /healthz
+              port: http
+            initialDelaySeconds: 5
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          securityContext:
+            allowPrivilegeEscalation: false
+            readOnlyRootFilesystem: true
+            capabilities:
+              drop:
+                - ALL
+          resources:
+            requests:
+              cpu: 100m
+              memory: 128Mi
+            limits:
+              cpu: 500m
+              memory: 512Mi
+      restartPolicy: Always

apps/base/logging/logviewer/gateway.yaml

@@ -0,0 +1,38 @@
+---
+# Internal front for the logviewer UI (cf. mediamover/pdbmux).
+apiVersion: gateway.networking.k8s.io/v1
+kind: Gateway
+metadata:
+  labels:
+    traefik.io/instance: internal
+  annotations:
+    cert-manager.io/cluster-issuer: vault-issuer
+    cert-manager.io/common-name: logviewer.unkin.net
+    cert-manager.io/private-key-size: "4096"
+    external-dns.alpha.kubernetes.io/hostname: logviewer.unkin.net
+    external-dns.alpha.kubernetes.io/target: 198.18.200.4
+  name: logviewer
+  namespace: logging
+spec:
+  gatewayClassName: traefik-internal
+  listeners:
+    - allowedRoutes:
+        namespaces:
+          from: Same
+      hostname: logviewer.unkin.net
+      name: http
+      port: 80
+      protocol: HTTP
+    - allowedRoutes:
+        namespaces:
+          from: Same
+      hostname: logviewer.unkin.net
+      name: https
+      port: 443
+      protocol: HTTPS
+      tls:
+        certificateRefs:
+          - group: ""
+            kind: Secret
+            name: logviewer-tls
+        mode: Terminate

apps/base/logging/logviewer/httproute.yaml

@@ -2,15 +2,15 @@
 apiVersion: gateway.networking.k8s.io/v1
 kind: HTTPRoute
 metadata:
-  name: vlogs-http-redirect
-  namespace: vlogs
+  name: logviewer-http-redirect
+  namespace: logging
 spec:
   hostnames:
-    - vlogs.unkin.net
+    - logviewer.unkin.net
   parentRefs:
     - group: gateway.networking.k8s.io
       kind: Gateway
-      name: vlogs-external
+      name: logviewer
       sectionName: http
   rules:
     - filters:
@@ -26,33 +26,21 @@ spec:
 apiVersion: gateway.networking.k8s.io/v1
 kind: HTTPRoute
 metadata:
-  name: vlogs
-  namespace: vlogs
+  name: logviewer
+  namespace: logging
 spec:
   hostnames:
-    - vlogs.unkin.net
+    - logviewer.unkin.net
   parentRefs:
     - group: gateway.networking.k8s.io
       kind: Gateway
-      name: vlogs-external
+      name: logviewer
       sectionName: https
   rules:
-    # Exact / outranks the PathPrefix catch-all; target needs the trailing slash
-    - filters:
-        - type: RequestRedirect
-          requestRedirect:
-            path:
-              type: ReplaceFullPath
-              replaceFullPath: /select/vmui/
-            statusCode: 302
-      matches:
-        - path:
-            type: Exact
-            value: /
     - backendRefs:
         - group: ""
           kind: Service
-          name: vlogs-oauth2
+          name: logviewer-oauth2
           port: 80
           weight: 1
       matches:

apps/base/logging/logviewer/kustomization.yaml

@@ -3,9 +3,8 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 
 resources:
-  - namespace.yaml
-  - vaultauth.yaml
   - vaultstaticsecret.yaml
+  - deployment.yaml
   - oauth2-proxy-configmap.yaml
   - oauth2-proxy-deployment.yaml
   - service.yaml

apps/base/logging/logviewer/oauth2-proxy-configmap.yaml

@@ -1,27 +1,31 @@
 ---
+# Non-secret oauth2-proxy configuration (client_id/secret/cookie_secret come
+# from the logviewer-oauth-credentials Secret). Single auth front for the
+# logviewer UI: everything requires an Authentik session in akP-logviewer-admin.
 apiVersion: v1
 kind: ConfigMap
 metadata:
-  name: vlogs-oauth2-env
-  namespace: vlogs
+  name: logviewer-oauth2-env
+  namespace: logging
 data:
   OAUTH2_PROXY_HTTP_ADDRESS: "0.0.0.0:4180"
   OAUTH2_PROXY_PROVIDER: "oidc"
-  OAUTH2_PROXY_OIDC_ISSUER_URL: "https://identity.unkin.net/application/o/vlogs/"
-  OAUTH2_PROXY_REDIRECT_URL: "https://vlogs.unkin.net/oauth2/callback"
-  OAUTH2_PROXY_UPSTREAMS: "http://vlselect-logs.logging.svc.cluster.local:9471/"
+  # Authentik logviewer app discovery issuer (terraform-authentik PR #21).
+  OAUTH2_PROXY_OIDC_ISSUER_URL: "https://identity.unkin.net/application/o/logviewer/"
+  OAUTH2_PROXY_REDIRECT_URL: "https://logviewer.unkin.net/oauth2/callback"
+  OAUTH2_PROXY_UPSTREAMS: "http://logviewer.logging.svc.cluster.local:8080/"
   OAUTH2_PROXY_SCOPE: "openid email profile ak_groups"
   # Populate session.Groups from the Authentik ak_groups claim.
   OAUTH2_PROXY_OIDC_GROUPS_CLAIM: "ak_groups"
-  OAUTH2_PROXY_ALLOWED_GROUPS: "akP-vlogs-admin"
+  OAUTH2_PROXY_ALLOWED_GROUPS: "akP-logviewer-admin"
   OAUTH2_PROXY_PASS_USER_HEADERS: "true"
   OAUTH2_PROXY_EMAIL_DOMAINS: "*"
   # Authentik hardcodes email_verified=false in the id_token; authorization is
   # enforced via ak_groups, so accepting the unverified email is safe.
   OAUTH2_PROXY_INSECURE_OIDC_ALLOW_UNVERIFIED_EMAIL: "true"
   OAUTH2_PROXY_COOKIE_SECURE: "true"
-  OAUTH2_PROXY_COOKIE_DOMAINS: "vlogs.unkin.net"
-  OAUTH2_PROXY_WHITELIST_DOMAINS: "vlogs.unkin.net"
+  OAUTH2_PROXY_COOKIE_DOMAINS: "logviewer.unkin.net"
+  OAUTH2_PROXY_WHITELIST_DOMAINS: "logviewer.unkin.net"
   OAUTH2_PROXY_REVERSE_PROXY: "true"
   OAUTH2_PROXY_PROVIDER_CA_FILES: "/etc/ssl/combined/ca-certificates.crt"
   OAUTH2_PROXY_CODE_CHALLENGE_METHOD: "S256"

apps/base/logging/logviewer/oauth2-proxy-deployment.yaml

@@ -2,16 +2,16 @@
 apiVersion: apps/v1
 kind: Deployment
 metadata:
-  name: vlogs-oauth2
-  namespace: vlogs
+  name: logviewer-oauth2
+  namespace: logging
   annotations:
     configmap.reloader.stakater.com/auto: "true"
-    secret.reloader.stakater.com/reload: "vlogs-oauth-credentials,vault-ca-cert"
+    secret.reloader.stakater.com/reload: "logviewer-oauth-credentials,vault-ca-cert"
 spec:
   replicas: 1
   selector:
     matchLabels:
-      app: vlogs-oauth2
+      app: logviewer-oauth2
   strategy:
     rollingUpdate:
       maxUnavailable: 1
@@ -19,7 +19,7 @@ spec:
   template:
     metadata:
       labels:
-        app: vlogs-oauth2
+        app: logviewer-oauth2
     spec:
       serviceAccountName: default
       automountServiceAccountToken: false
@@ -69,23 +69,23 @@ spec:
               protocol: TCP
           envFrom:
             - configMapRef:
-                name: vlogs-oauth2-env
+                name: logviewer-oauth2-env
                 optional: false
           env:
             - name: OAUTH2_PROXY_CLIENT_ID
               valueFrom:
                 secretKeyRef:
-                  name: vlogs-oauth-credentials
+                  name: logviewer-oauth-credentials
                   key: client_id
             - name: OAUTH2_PROXY_CLIENT_SECRET
               valueFrom:
                 secretKeyRef:
-                  name: vlogs-oauth-credentials
+                  name: logviewer-oauth-credentials
                   key: client_secret
             - name: OAUTH2_PROXY_COOKIE_SECRET
               valueFrom:
                 secretKeyRef:
-                  name: vlogs-oauth-credentials
+                  name: logviewer-oauth-credentials
                   key: cookie_secret
           volumeMounts:
             - name: combined-certs

apps/base/logging/logviewer/service.yaml

@@ -0,0 +1,36 @@
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: logviewer
+  namespace: logging
+spec:
+  internalTrafficPolicy: Cluster
+  ports:
+    - name: http
+      port: 8080
+      protocol: TCP
+      targetPort: http
+  selector:
+    app: logviewer
+  sessionAffinity: None
+  type: ClusterIP
+---
+# Front-door entry Service: the HTTPRoute for logviewer.unkin.net targets this;
+# all traffic enters via oauth2-proxy.
+apiVersion: v1
+kind: Service
+metadata:
+  name: logviewer-oauth2
+  namespace: logging
+spec:
+  internalTrafficPolicy: Cluster
+  ports:
+    - name: http
+      port: 80
+      protocol: TCP
+      targetPort: http
+  selector:
+    app: logviewer-oauth2
+  sessionAffinity: None
+  type: ClusterIP

apps/base/logging/logviewer/vaultstaticsecret.yaml

@@ -0,0 +1,21 @@
+---
+# Authentik OIDC client for logviewer (client_id, client_secret, cookie_secret)
+# seeded at kv/kubernetes/namespace/logging/default/oauth-credentials; the
+# logging/default templated policy already grants read, so no terraform-vault
+# change is needed.
+apiVersion: secrets.hashicorp.com/v1beta1
+kind: VaultStaticSecret
+metadata:
+  name: logviewer-oauth-credentials
+  namespace: logging
+spec:
+  destination:
+    create: true
+    name: logviewer-oauth-credentials
+    overwrite: true
+  hmacSecretData: true
+  mount: kv
+  path: kubernetes/namespace/logging/default/oauth-credentials
+  refreshAfter: 5m
+  type: kv-v2
+  vaultAuthRef: default

apps/base/logging/nats-bootstrap-job.yaml

@@ -0,0 +1,158 @@
+---
+# Declarative JetStream provisioning: the LOGS stream + durable consumers.
+# ArgoCD PostSync hook, idempotent create-or-UPDATE, re-runs each sync.
+#
+# Stream LOGS: file storage, 3 replicas, retention=limits (NOT workqueue) so the
+# transform tier AND the archiver each independently see every message — reading
+# never deletes; only max-age/max-bytes do. S2 compression is on (logs compress
+# well). Replay window = max-age (3d default). Beyond that, the S3 archive is the
+# ONLY long-term source — everything else is gone after 3 days (accepted design).
+#
+# TUNABLE LIMITS LIVE IN A CONFIGMAP (nats-stream-limits): max_age, max_bytes,
+# dupe_window. Change the ConfigMap and re-sync — this Job re-runs and applies
+# the new limits via `nats stream edit` (no manual surgery). The ConfigMap is
+# generated with a content-hash suffix (kustomize), so editing it changes both
+# the ConfigMap name AND this Job's env reference → the PostSync hook Job's spec
+# changes and Argo re-runs it (belt-and-suspenders on top of hooks running each
+# sync; hook-delete-policy=BeforeHookCreation recreates it every time).
+#
+# Consumers (independent offsets = true fan-out):
+#   transform  -> whole log stream, feeds the ClickHouse transform tier
+#   archiver   -> configurable security-relevant subset, feeds the S3 archiver.
+#                 Default filter is Vault audit (logs.k8s.vault.>); ADD subjects
+#                 by editing ARCHIVE_SUBJECTS (space-separated -> repeated
+#                 --filter). Exact default set is an open decision for Ben.
+#
+# Runbook (replay):
+#   (a) reprocess from JetStream (within max-age, 3d): scale the transform tier
+#       to 0, then `nats consumer rm LOGS transform` and re-run this Job
+#       (recreates at DeliverAll), or `nats consumer edit`/`--replay` from a
+#       start seq/time.
+#   (b) long-horizon (beyond JetStream): re-ingest S3 archive objects back
+#       through the transform tier (vector aws_s3 source or a one-shot Job).
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: nats-bootstrap
+  namespace: logging
+  annotations:
+    argocd.argoproj.io/hook: PostSync
+    argocd.argoproj.io/hook-delete-policy: BeforeHookCreation
+  labels:
+    app.kubernetes.io/name: nats-bootstrap
+    app.kubernetes.io/component: bootstrap
+spec:
+  backoffLimit: 20
+  activeDeadlineSeconds: 1800
+  ttlSecondsAfterFinished: 3600
+  template:
+    metadata:
+      labels:
+        app.kubernetes.io/name: nats-bootstrap
+        vector.dev/exclude: "true"
+    spec:
+      restartPolicy: OnFailure
+      securityContext:
+        runAsNonRoot: true
+        runAsUser: 1000
+        runAsGroup: 1000
+      containers:
+        - name: nats-bootstrap
+          image: docker.io/natsio/nats-box:0.18.0
+          # nats CLI stats the working directory when loading its response
+          # schemas; under readOnlyRootFilesystem + runAsUser 1000 the image's
+          # default WORKDIR is not accessible ("stat .: permission denied"), so
+          # run from the writable /tmp emptyDir.
+          workingDir: /tmp
+          securityContext:
+            allowPrivilegeEscalation: false
+            readOnlyRootFilesystem: true
+            capabilities:
+              drop:
+                - ALL
+          env:
+            - name: HOME
+              value: /tmp
+            - name: NATS_URL
+              value: "nats://nats.logging.svc.cluster.local:4222"
+            - name: NATS_ADMIN_PASSWORD
+              valueFrom:
+                secretKeyRef:
+                  name: nats-auth
+                  key: admin_password
+            # Tunable stream limits — sourced from the ConfigMap.
+            - name: MAX_AGE
+              valueFrom:
+                configMapKeyRef:
+                  name: nats-stream-limits
+                  key: max_age
+            - name: MAX_BYTES
+              valueFrom:
+                configMapKeyRef:
+                  name: nats-stream-limits
+                  key: max_bytes
+            - name: DUPE_WINDOW
+              valueFrom:
+                configMapKeyRef:
+                  name: nats-stream-limits
+                  key: dupe_window
+            # Space-separated subject filters for the archiver consumer.
+            - name: ARCHIVE_SUBJECTS
+              value: "logs.k8s.vault.>"
+          command:
+            - /bin/sh
+            - -ec
+            - |
+              export NATS_USER=log-admin NATS_PASSWORD="$NATS_ADMIN_PASSWORD"
+              echo "Waiting for NATS + JetStream ..."
+              until nats --server "$NATS_URL" account info >/dev/null 2>&1; do
+                echo "  not ready, retry in 5s"; sleep 5
+              done
+
+              echo "Ensuring stream LOGS (max_age=$MAX_AGE max_bytes=$MAX_BYTES dupe=$DUPE_WINDOW) ..."
+              # Create if absent; otherwise converge the mutable limits from the
+              # ConfigMap. (storage/retention/replicas are immutable, set only on
+              # create.)
+              nats stream add LOGS \
+                --subjects='logs.>' --storage=file --replicas=3 \
+                --retention=limits --discard=old --compression=s2 \
+                --max-age="$MAX_AGE" --max-bytes="$MAX_BYTES" \
+                --max-msgs=-1 --max-msgs-per-subject=-1 --max-msg-size=-1 \
+                --max-consumers=-1 --dupe-window="$DUPE_WINDOW" --defaults 2>/dev/null \
+              && echo "  created" \
+              || nats stream edit -f LOGS \
+                --subjects='logs.>' --discard=old --compression=s2 \
+                --max-age="$MAX_AGE" --max-bytes="$MAX_BYTES" \
+                --max-msgs=-1 --max-msgs-per-subject=-1 --max-msg-size=-1 \
+                --max-consumers=-1 --dupe-window="$DUPE_WINDOW"
+
+              echo "Ensuring consumer transform ..."
+              nats consumer add LOGS transform \
+                --pull --filter='logs.>' --deliver=all --ack=explicit \
+                --max-deliver=-1 --replay=instant --defaults 2>/dev/null \
+              || echo "  transform already exists"
+
+              echo "Ensuring consumer archiver (filters: $ARCHIVE_SUBJECTS) ..."
+              filter_args=""
+              for s in $ARCHIVE_SUBJECTS; do filter_args="$filter_args --filter=$s"; done
+              # shellcheck disable=SC2086
+              nats consumer add LOGS archiver \
+                --pull $filter_args --deliver=all --ack=explicit \
+                --max-deliver=-1 --replay=instant --defaults 2>/dev/null \
+              || echo "  archiver already exists"
+
+              echo "Done."
+              nats stream info LOGS
+          resources:
+            requests:
+              cpu: 50m
+              memory: 64Mi
+            limits:
+              cpu: 500m
+              memory: 256Mi
+          volumeMounts:
+            - name: tmp
+              mountPath: /tmp
+      volumes:
+        - name: tmp
+          emptyDir: {}

apps/base/logging/serviceaccount_logarchiver.yaml

@@ -0,0 +1,9 @@
+---
+# Dedicated SA for logarchiver's Vault k8s-auth login (role logging_logarchiver,
+# terraform-vault). Only used to fetch the logarchive public key.
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: logarchiver
+  namespace: logging
+automountServiceAccountToken: false

apps/base/logging/vaultstaticsecret.yaml

@@ -0,0 +1,73 @@
+---
+# ClickHouse credentials for the `vector` user.
+#
+# Seed the Vault KV entry once (values are NOT stored in git), e.g.:
+#   PW=$(openssl rand -base64 24)
+#   HASH=$(printf '%s' "$PW" | sha256sum | cut -d' ' -f1)
+#   vault kv put kv/kubernetes/namespace/logging/default/clickhouse-credentials \
+#     username=vector password="$PW" password_sha256_hex="$HASH"
+#
+# The `logging/default` ServiceAccount reads this path via the templated
+# `policies/kv/kubernetes/default.yaml` policy (k8s auth role `default`), so no
+# terraform-vault change is required — only the value above must be written.
+apiVersion: secrets.hashicorp.com/v1beta1
+kind: VaultStaticSecret
+metadata:
+  name: clickhouse-credentials
+  namespace: logging
+spec:
+  destination:
+    create: true
+    name: clickhouse-credentials
+    overwrite: true
+  hmacSecretData: true
+  mount: kv
+  path: kubernetes/namespace/logging/default/clickhouse-credentials
+  refreshAfter: 5m
+  type: kv-v2
+  vaultAuthRef: default
+---
+# ClickHouse credentials for the read-only `logreader` user (CLI tools +
+# logviewer UI). Seeded the same way as clickhouse-credentials above:
+#   PW=$(openssl rand -hex 24)
+#   HASH=$(printf '%s' "$PW" | sha256sum | cut -d' ' -f1)
+#   vault kv put kv/kubernetes/namespace/logging/default/clickhouse-logreader \
+#     username=logreader password="$PW" password_sha256_hex="$HASH"
+apiVersion: secrets.hashicorp.com/v1beta1
+kind: VaultStaticSecret
+metadata:
+  name: clickhouse-logreader
+  namespace: logging
+spec:
+  destination:
+    create: true
+    name: clickhouse-logreader
+    overwrite: true
+  hmacSecretData: true
+  mount: kv
+  path: kubernetes/namespace/logging/default/clickhouse-logreader
+  refreshAfter: 5m
+  type: kv-v2
+  vaultAuthRef: default
+---
+# NATS JetStream auth. Distinct passwords for the producer (edge), consumer
+# (transform tier + archiver) and admin (bootstrap Job) users. Seed once:
+#   for k in admin producer consumer; do declare P_$k=$(openssl rand -base64 24); done
+#   vault kv put kv/kubernetes/namespace/logging/default/nats-auth \
+#     admin_password="$P_admin" producer_password="$P_producer" consumer_password="$P_consumer"
+apiVersion: secrets.hashicorp.com/v1beta1
+kind: VaultStaticSecret
+metadata:
+  name: nats-auth
+  namespace: logging
+spec:
+  destination:
+    create: true
+    name: nats-auth
+    overwrite: true
+  hmacSecretData: true
+  mount: kv
+  path: kubernetes/namespace/logging/default/nats-auth
+  refreshAfter: 5m
+  type: kv-v2
+  vaultAuthRef: default

apps/base/logging/vector/agent.yaml

@@ -0,0 +1,43 @@
+---
+# Vector EDGE agent pipeline (DaemonSet) — thin publisher, single source of
+# truth. Mounted via existingConfigMaps (NOT the chart's customConfig, whose
+# Helm `tpl` pass collides with Vector's own {{ }} / ${ } syntax). Tails all pod
+# logs, attaches only routing tokens, publishes to JetStream. No parsing.
+data_dir: /vector-data-dir
+
+api:
+  enabled: false
+
+sources:
+  kubernetes_logs:
+    type: kubernetes_logs
+
+transforms:
+  # Routing metadata only: NATS-subject-safe namespace + container tokens.
+  keymeta:
+    type: remap
+    inputs:
+      - kubernetes_logs
+    source: |
+      ns = to_string(.kubernetes.pod_namespace || "unknown") ?? "unknown"
+      .ns_token = replace(ns, r'[^a-zA-Z0-9_-]', "_")
+      cont = to_string(.kubernetes.container_name || "unknown") ?? "unknown"
+      .cont_token = replace(cont, r'[^a-zA-Z0-9_-]', "_")
+
+sinks:
+  to_jetstream:
+    type: nats
+    inputs:
+      - keymeta
+    url: nats://nats.logging.svc.cluster.local:4222
+    connection_name: vector-agent
+    subject: "logs.k8s.{{ ns_token }}.{{ cont_token }}"
+    jetstream:
+      enabled: true
+    auth:
+      strategy: user_password
+      user_password:
+        user: log-producer
+        password: ${NATS_PRODUCER_PASSWORD}
+    encoding:
+      codec: json

apps/base/logging/vector/aggregator-tests.yaml

@@ -0,0 +1,668 @@
+---
+# `vector test` unit tests for the aggregator transforms. Merged with
+# aggregator.yaml in CI (.woodpecker/vector-test.yaml). This is the pattern the
+# per-app parsing follow-ups extend: add a test per new transform here.
+tests:
+  - name: subject_routes_k8s_vs_vm
+    inputs:
+      - insert_at: route
+        type: log
+        log_fields:
+          subject: "logs.k8s.shop.web"
+          message: "routed"
+    outputs:
+      - extract_from: route.k8s
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.message, "routed")
+
+  - name: k8s_log_is_normalised
+    inputs:
+      - insert_at: k8s_shape
+        type: log
+        log_fields:
+          message: "hello from pod"
+          stream: "stdout"
+          timestamp: "2026-07-27T00:00:00Z"
+          kubernetes.pod_name: "web-abc"
+          kubernetes.pod_namespace: "shop"
+          kubernetes.container_name: "web"
+          kubernetes.pod_node_name: "node-1"
+    outputs:
+      - extract_from: k8s_shape
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "shop")
+              assert_eq!(.pod, "web-abc")
+              assert_eq!(.container, "web")
+              assert_eq!(.host, "node-1")
+              assert_eq!(.stream, "stdout")
+              assert_eq!(.message, "hello from pod")
+
+  - name: vm_log_is_normalised
+    inputs:
+      - insert_at: vm_shape
+        type: log
+        log_fields:
+          message: "sshd started"
+          host: "vm-db-1"
+          severity: "info"
+          role: "database"
+    outputs:
+      - extract_from: vm_shape
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.host, "vm-db-1")
+              assert_eq!(.severity, "info")
+              assert_eq!(.message, "sshd started")
+              assert_eq!(.labels.role, "database")
+
+  # --- catch-all preservation: an unclaimed k8s event still flows app_route ->
+  # generic route -> k8s_shape (proves the two-stage chain keeps the fallback) ---
+  - name: unclaimed_k8s_falls_through_to_generic
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.shop.web"
+          message: "plain app log"
+    outputs:
+      - extract_from: route.k8s
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.message, "plain app log")
+
+  # --- Tier-1: Authentik SSO (k8s, LIVE NOW) ---
+  - name: authentik_routes_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.authentik.server"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.authentik
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: authentik_parse_extracts_event
+    inputs:
+      - insert_at: authentik_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.authentik.server"
+          stream: "stdout"
+          kubernetes.pod_namespace: "authentik"
+          kubernetes.container_name: "server"
+          kubernetes.pod_node_name: "node-2"
+          message: '{"event":"login","action":"login","user":"alice","client_ip":"203.0.113.9","result":"success","level":"info","logger":"authentik.events","timestamp":"2026-07-27T00:00:00Z"}'
+    outputs:
+      - extract_from: authentik_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "authentik")
+              assert_eq!(.container, "server")
+              assert_eq!(.severity, "info")
+              assert_eq!(.message, "login")
+              assert_eq!(.labels.app, "authentik")
+              assert_eq!(.fields.event, "login")
+              assert_eq!(.fields.action, "login")
+              assert_eq!(.fields.user, "alice")
+              assert_eq!(.fields.client_ip, "203.0.113.9")
+              assert_eq!(.fields.result, "success")
+
+  # --- Tier-1: Traefik ingress (k8s, JSON access logs) ---
+  - name: traefik_routes_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.traefik-system.traefik"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.traefik
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: traefik_parse_extracts_access_fields
+    inputs:
+      - insert_at: traefik_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.traefik-system.traefik"
+          kubernetes.pod_namespace: "traefik-system"
+          kubernetes.container_name: "traefik"
+          kubernetes.pod_node_name: "node-3"
+          message: '{"RouterName":"web@kubernetes","ServiceName":"shop-svc@kubernetes","RequestMethod":"GET","RequestPath":"/api","RequestHost":"shop.example.net","RequestProtocol":"HTTP/1.1","DownstreamStatus":200,"Duration":5000000,"ClientHost":"203.0.113.5","StartUTC":"2026-07-27T00:00:00Z"}'
+    outputs:
+      - extract_from: traefik_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "traefik-system")
+              assert_eq!(.labels.app, "traefik")
+              assert_eq!(.message, "GET /api 200")
+              assert_eq!(.fields.route, "web@kubernetes")
+              assert_eq!(.fields.service, "shop-svc@kubernetes")
+              assert_eq!(.fields.method, "GET")
+              assert_eq!(.fields.path, "/api")
+              assert_eq!(.fields.host, "shop.example.net")
+              assert_eq!(.fields.status, "200")
+              assert_eq!(.fields.duration_ms, "5")
+              assert_eq!(.fields.client_ip, "203.0.113.5")
+
+  # --- Tier-1: Vault/OpenBao file audit (VM, awaiting VM vector) ---
+  - name: vault_routes_by_file
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.vault1_syd1"
+          file: "/var/log/vault_audit.log"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.vault
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: vault_parse_extracts_request
+    inputs:
+      - insert_at: vault_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.vault1_syd1"
+          host: "vault1"
+          file: "/var/log/vault_audit.log"
+          message: '{"time":"2026-07-27T00:00:00Z","type":"response","auth":{"display_name":"token"},"request":{"operation":"read","path":"secret/data/app","remote_address":"10.0.0.9"},"error":""}'
+    outputs:
+      - extract_from: vault_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.host, "vault1")
+              assert_eq!(.labels.app, "vault")
+              assert_eq!(.message, "read secret/data/app")
+              assert_eq!(.fields.type, "response")
+              assert_eq!(.fields.display_name, "token")
+              assert_eq!(.fields.operation, "read")
+              assert_eq!(.fields.path, "secret/data/app")
+              assert_eq!(.fields.remote_address, "10.0.0.9")
+
+  # --- Tier-1: nginx access (VM, awaiting VM vector) ---
+  - name: nginx_access_routes_by_file
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.web1_syd1"
+          file: "/var/log/nginx/shop_access.log"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.nginx_access
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: nginx_access_parse_extracts_combined
+    inputs:
+      - insert_at: nginx_access_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.web1_syd1"
+          host: "web1"
+          file: "/var/log/nginx/shop_access.log"
+          message: '192.0.2.10 - - [27/Jul/2026:00:00:00 +0000] "GET /index.html HTTP/1.1" 200 1024 "https://ref.example/" "Mozilla/5.0" 0.012'
+    outputs:
+      - extract_from: nginx_access_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.stream, "access")
+              assert_eq!(.labels.log_type, "access")
+              assert_eq!(.fields.client_ip, "192.0.2.10")
+              assert_eq!(.fields.method, "GET")
+              assert_eq!(.fields.path, "/index.html")
+              assert_eq!(.fields.status, "200")
+              assert_eq!(.fields.bytes, "1024")
+              assert_eq!(.fields.referer, "https://ref.example/")
+              assert_eq!(.fields.user_agent, "Mozilla/5.0")
+              assert_eq!(.fields.request_time, "0.012")
+
+  # --- Tier-1: nginx error (VM, awaiting VM vector) ---
+  - name: nginx_error_routes_by_file
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.web1_syd1"
+          file: "/var/log/nginx/shop_error.log"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.nginx_error
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: nginx_error_parse_extracts_fields
+    inputs:
+      - insert_at: nginx_error_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.web1_syd1"
+          host: "web1"
+          file: "/var/log/nginx/shop_error.log"
+          message: '2026/07/27 00:00:00 [error] 1234#0: *5 open() "/var/www/x" failed (2: No such file or directory), client: 192.0.2.20, server: shop, request: "GET / HTTP/1.1", host: "shop"'
+    outputs:
+      - extract_from: nginx_error_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.stream, "error")
+              assert_eq!(.severity, "error")
+              assert_eq!(.labels.log_type, "error")
+              assert_eq!(.fields.level, "error")
+              assert_eq!(.fields.pid, "1234")
+              assert_eq!(.fields.cid, "5")
+              assert_eq!(.fields.client_ip, "192.0.2.20")
+
+  # --- Tier-1: HAProxy httplog (VM journald, awaiting VM vector) ---
+  - name: haproxy_routes_by_identifier
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.halb1_syd1"
+          SYSLOG_IDENTIFIER: "haproxy"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.haproxy
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: haproxy_parse_extracts_timers
+    inputs:
+      - insert_at: haproxy_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.halb1_syd1"
+          host: "halb1"
+          SYSLOG_IDENTIFIER: "haproxy"
+          message: '192.0.2.30:54321 [27/Jul/2026:00:00:00.123] fe_http be_app/app1 10/0/1/2/13 200 512 - - ---- 5/4/3/2/0 0/0 "GET /health HTTP/1.1"'
+    outputs:
+      - extract_from: haproxy_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.labels.app, "haproxy")
+              assert_eq!(.fields.client_ip, "192.0.2.30")
+              assert_eq!(.fields.frontend, "fe_http")
+              assert_eq!(.fields.backend, "be_app")
+              assert_eq!(.fields.server, "app1")
+              assert_eq!(.fields.tq, "10")
+              assert_eq!(.fields.tw, "0")
+              assert_eq!(.fields.tc, "1")
+              assert_eq!(.fields.tr, "2")
+              assert_eq!(.fields.tt, "13")
+              assert_eq!(.fields.termination_state, "----")
+              assert_eq!(.fields.retries, "0")
+              assert_eq!(.fields.status, "200")
+              assert_eq!(.fields.bytes, "512")
+
+  # --- Tier-1: glauth LDAP (VM, awaiting VM vector) ---
+  - name: glauth_routes_by_identifier
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.ldap1_syd1"
+          SYSLOG_IDENTIFIER: "glauth"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.glauth
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: glauth_parse_extracts_bind
+    inputs:
+      - insert_at: glauth_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.ldap1_syd1"
+          host: "ldap1"
+          SYSLOG_IDENTIFIER: "glauth"
+          message: '{"level":"info","msg":"Bind success as user","bindDN":"cn=admin,dc=example,dc=com","src":"192.0.2.40:1234","time":"2026-07-27T00:00:00Z"}'
+    outputs:
+      - extract_from: glauth_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.host, "ldap1")
+              assert_eq!(.severity, "info")
+              assert_eq!(.labels.app, "glauth")
+              assert_eq!(.fields.bindDN, "cn=admin,dc=example,dc=com")
+              assert_eq!(.fields.remote, "192.0.2.40:1234")
+              assert_eq!(.fields.success, "true")
+
+  # ================= Tier-2 (stacks on #318) =================
+
+  # --- BIND query logs (k8s bind-* + VM named) ---
+  - name: bind_routes_k8s_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.bind-internal.named"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.bind_query
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: bind_routes_vm_by_identifier
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.dns1_syd1"
+          SYSLOG_IDENTIFIER: "named"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.bind_query
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: bind_parse_extracts_query
+    inputs:
+      - insert_at: bind_query_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.bind-internal.named"
+          kubernetes.pod_namespace: "bind-internal"
+          kubernetes.container_name: "named"
+          kubernetes.pod_node_name: "node-4"
+          message: '02-Aug-2026 00:00:00.123 client @0x7f 192.0.2.1#40426 (www.example.com): view internal: query: www.example.com IN A +E(0)K (198.18.200.7)'
+    outputs:
+      - extract_from: bind_query_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "bind-internal")
+              assert_eq!(.labels.app, "bind")
+              assert_eq!(.message, "query www.example.com A")
+              assert_eq!(.fields.client_ip, "192.0.2.1")
+              assert_eq!(.fields.qname, "www.example.com")
+              assert_eq!(.fields.qclass, "IN")
+              assert_eq!(.fields.qtype, "A")
+              assert_eq!(.fields.view, "internal")
+
+  # --- Rancher audit (k8s, cattle-system sidecar) ---
+  - name: rancher_routes_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.cattle-system.rancher-audit-log"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.rancher_audit
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: rancher_parse_extracts_audit
+    inputs:
+      - insert_at: rancher_audit_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.cattle-system.rancher-audit-log"
+          kubernetes.pod_namespace: "cattle-system"
+          kubernetes.container_name: "rancher-audit-log"
+          kubernetes.pod_node_name: "node-5"
+          message: '{"auditID":"abc-123","requestURI":"/v3/tokens","user":{"name":"u-alice","group":["admins"]},"method":"GET","remoteAddr":"10.42.0.9:1234","responseCode":200,"requestTimestamp":"2026-08-01T00:00:00Z"}'
+    outputs:
+      - extract_from: rancher_audit_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "cattle-system")
+              assert_eq!(.labels.app, "rancher")
+              assert_eq!(.labels.log_type, "audit")
+              assert_eq!(.message, "GET /v3/tokens 200")
+              assert_eq!(.fields.user, "u-alice")
+              assert_eq!(.fields.verb, "GET")
+              assert_eq!(.fields.uri, "/v3/tokens")
+              assert_eq!(.fields.status, "200")
+
+  # --- CNPG Postgres (ONE transform, all clusters) ---
+  - name: cnpg_routes_by_postgres_container
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.litellm.postgres"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.cnpg_pg
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  # mutual exclusivity: an app-namespace CNPG pod (authentik) is claimed by
+  # cnpg_pg, NOT the authentik app route (which now carves out .postgres).
+  - name: cnpg_authentik_postgres_routes_to_cnpg
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.authentik.postgres"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.cnpg_pg
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: cnpg_parse_extracts_record
+    inputs:
+      - insert_at: cnpg_pg_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.litellm.postgres"
+          kubernetes.pod_namespace: "litellm"
+          kubernetes.container_name: "postgres"
+          kubernetes.pod_node_name: "node-6"
+          kubernetes.pod_labels."cnpg.io/cluster": "litellm-postgres"
+          message: '{"level":"info","ts":"2026-08-01T00:00:00Z","logger":"postgres","msg":"record","record":{"user_name":"litellm","database_name":"litellm","error_severity":"LOG","message":"duration: 12.345 ms  statement: SELECT 1","query":""}}'
+    outputs:
+      - extract_from: cnpg_pg_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "litellm")
+              assert_eq!(.severity, "LOG")
+              assert_eq!(.labels.app, "cnpg")
+              assert_eq!(.labels.cluster, "litellm-postgres")
+              assert_eq!(.fields.error_severity, "LOG")
+              assert_eq!(.fields.duration_ms, "12.345")
+              assert_eq!(.fields.user, "litellm")
+              assert_eq!(.fields.database, "litellm")
+
+  # --- Gitea router/access (k8s + VM) ---
+  - name: gitea_routes_k8s_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.gitea.gitea"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.gitea
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: gitea_parse_router_line
+    inputs:
+      - insert_at: gitea_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.gitea.gitea"
+          kubernetes.pod_namespace: "gitea"
+          kubernetes.container_name: "gitea"
+          kubernetes.pod_node_name: "node-7"
+          message: '2026/08/01 00:00:00 .../router.go:100:func() [I] router: completed GET /user/login for 10.0.0.1:0, 200 OK in 12.3ms @ web/base.go:1'
+    outputs:
+      - extract_from: gitea_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "gitea")
+              assert_eq!(.labels.app, "gitea")
+              assert_eq!(.message, "GET /user/login 200")
+              assert_eq!(.fields.method, "GET")
+              assert_eq!(.fields.path, "/user/login")
+              assert_eq!(.fields.status, "200")
+              assert_eq!(.fields.latency, "12.3ms")
+  - name: gitea_parse_access_line
+    inputs:
+      - insert_at: gitea_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.gitea.gitea"
+          kubernetes.pod_namespace: "gitea"
+          kubernetes.container_name: "gitea"
+          message: '10.0.0.5 - alice [01/Aug/2026:00:00:00 +0000] "POST /repo/foo HTTP/1.1" 201 512 "-" "git/2.0"'
+    outputs:
+      - extract_from: gitea_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.fields.method, "POST")
+              assert_eq!(.fields.path, "/repo/foo")
+              assert_eq!(.fields.status, "201")
+              assert_eq!(.fields.user, "alice")
+              assert_eq!(.fields.client_ip, "10.0.0.5")
+
+  # --- PuppetServer / PuppetDB (k8s stdout) ---
+  - name: puppet_routes_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.puppet.puppetserver"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.puppet
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: puppet_parse_logback_line
+    inputs:
+      - insert_at: puppet_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.puppet.puppetserver"
+          kubernetes.pod_namespace: "puppet"
+          kubernetes.container_name: "puppetserver"
+          kubernetes.pod_node_name: "node-8"
+          message: '2026-08-01 00:00:00,123 INFO  [qtp123-45] [puppetserver] Compiled catalog for web01.unkin.net in environment production in 1.23 seconds'
+    outputs:
+      - extract_from: puppet_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "puppet")
+              assert_eq!(.severity, "INFO")
+              assert_eq!(.labels.app, "puppet")
+              assert_eq!(.fields.level, "INFO")
+              assert_eq!(.fields.logger, "puppetserver")
+              assert_eq!(.fields.node, "web01.unkin.net")
+
+  # --- LiteLLM request logs (k8s JSON) ---
+  - name: litellm_routes_by_subject
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.k8s.litellm.litellm"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.litellm
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: litellm_parse_extracts_request
+    inputs:
+      - insert_at: litellm_parse
+        type: log
+        log_fields:
+          subject: "logs.k8s.litellm.litellm"
+          kubernetes.pod_namespace: "litellm"
+          kubernetes.container_name: "litellm"
+          kubernetes.pod_node_name: "node-9"
+          message: '{"message":"Request completed","level":"info","model":"gpt-4o","total_tokens":1234,"response_time":0.532,"api_key":"sk-abc","status":"success","timestamp":"2026-08-01T00:00:00Z"}'
+    outputs:
+      - extract_from: litellm_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "k8s")
+              assert_eq!(.namespace, "litellm")
+              assert_eq!(.severity, "info")
+              assert_eq!(.message, "Request completed")
+              assert_eq!(.labels.app, "litellm")
+              assert_eq!(.fields.model, "gpt-4o")
+              assert_eq!(.fields.tokens, "1234")
+              assert_eq!(.fields.latency, "0.532")
+              assert_eq!(.fields.key, "sk-abc")
+              assert_eq!(.fields.status, "success")
+
+  # --- Postfix maillog (VM, per-line best-effort) ---
+  - name: postfix_routes_by_identifier
+    inputs:
+      - insert_at: app_route
+        type: log
+        log_fields:
+          subject: "logs.vm.mail1_syd1"
+          SYSLOG_IDENTIFIER: "postfix/qmgr"
+          message: "routed"
+    outputs:
+      - extract_from: app_route.postfix
+        conditions:
+          - type: vrl
+            source: 'assert_eq!(.message, "routed")'
+  - name: postfix_parse_extracts_line
+    inputs:
+      - insert_at: postfix_parse
+        type: log
+        log_fields:
+          subject: "logs.vm.mail1_syd1"
+          host: "mail1"
+          SYSLOG_IDENTIFIER: "postfix/smtp"
+          message: 'ABC123DEF: to=<rcpt@example.com>, relay=mx.example.com[1.2.3.4]:25, delay=1.2, delays=0.1/0/0.5/0.6, dsn=2.0.0, status=sent (250 OK)'
+    outputs:
+      - extract_from: postfix_parse
+        conditions:
+          - type: vrl
+            source: |
+              assert_eq!(.source, "vm")
+              assert_eq!(.host, "mail1")
+              assert_eq!(.labels.app, "postfix")
+              assert_eq!(.fields.qid, "ABC123DEF")
+              assert_eq!(.fields.to, "rcpt@example.com")
+              assert_eq!(.fields.relay, "mx.example.com[1.2.3.4]:25")
+              assert_eq!(.fields.delay, "1.2")
+              assert_eq!(.fields.status, "sent")
+              assert_eq!(.fields.program, "postfix/smtp")