docs(network): traffic classes as built, node-to-node on VLAN 46 (#576 )

NETWORK-traffic-classes.md described table 48 and rules 980/981/990/991, which #857 replaced with one table (47) decided by longest prefix, and listed management as not started. Rewritten to the live state: five classes, table 47 contents owned by bird, node IP and etcd on node0, the etcd leave-and-rejoin procedure, and the mixed-state hazard hit during the node IP move. The bird DaemonSet comment still named table 48. Closes #576 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Merge pull request 'feat(talos): node IP and etcd on VLAN 46 (#576 step 3)' (#922 ) from feat/node-ip-vlan-46 into master
2026-09-23 14:48:12 +02:00 · 2026-09-23 12:47:06 +00:00 · 2026-09-23 14:43:17 +02:00 · 2026-09-23 11:31:32 +00:00 · 2026-09-23 11:31:23 +00:00 · 2026-09-23 13:28:57 +02:00
49 changed files with 506 additions and 283 deletions
@@ -101,7 +101,7 @@ steps:
        bash scripts/ci/make-diag-kubeconfig.sh

  - name: alert-triage-agent
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
        from_secret: claude_code_oauth_token
@@ -350,17 +350,17 @@ steps:
          --destination gitea.renner.uno/gitea_admin/heating-can:$TAG \
          --destination gitea.renner.uno/gitea_admin/heating-can:latest

-  # netbird-proxy-services: the controller that reconciles NetBird reverse-proxy
+  # netbird-service-sync: the controller that reconciles NetBird reverse-proxy
  # services (public Ingress hosts -> TLS passthrough -> Traefik ClusterIP).
  # Runtime artifact like heating-can; same registry so the audit tracks it.
-  - name: build-netbird-proxy-services
+  - name: build-netbird-service-sync
    image: registry.gitlab.com/gitlab-ci-utils/container-images/kaniko:v1.25.18-debug
-    directory: ci/images/netbird-proxy-services
+    directory: ci/images/netbird-service-sync
    depends_on: []
    when:
      - event: deployment
      - event: push
-        path: ["ci/images/netbird-proxy-services/**", ".woodpecker/build-ci-images.yml"]
+        path: ["ci/images/netbird-service-sync/**", ".woodpecker/build-ci-images.yml"]
    environment:
      GITEA_USER: gitea_admin
      GITEA_TOKEN:
@@ -374,11 +374,11 @@ steps:
        /kaniko/executor \
          --context . \
          --dockerfile Containerfile \
-          --destination gitea.renner.uno/gitea_admin/netbird-proxy-services:$TAG \
-          --destination gitea.renner.uno/gitea_admin/netbird-proxy-services:latest
+          --destination gitea.renner.uno/gitea_admin/netbird-service-sync:$TAG \
+          --destination gitea.renner.uno/gitea_admin/netbird-service-sync:latest

  # wiki-index: embedding index over the Notes repo's wiki/, fed from Gitea and
-  # queried by ~/Notes/bin/wiki_rank.py. Runtime artifact like netbird-proxy-services.
+  # queried by ~/Notes/bin/wiki_rank.py. Runtime artifact like netbird-service-sync.
  - name: build-wiki-index
    image: registry.gitlab.com/gitlab-ci-utils/container-images/kaniko:v1.25.18-debug
    directory: ci/images/wiki-index
@@ -14,8 +14,8 @@
 #                 and  ci-claude:latest.  The Containerfile makes this layer
 #                 byte-reproducible, so rebuilding an unchanged version dedups
 #                 in the registry instead of orphaning a fresh ~260 MB copy.
-#   3. bump-pr  — (only when the version actually changed) rewrite the 5
-#                 consumer pins to  ci-claude:<version>  and open a PR. A human
+#   3. bump-pr  — (only when any pin is off the latest version) rewrite every
+#                 consumer pin to  ci-claude:<version>  and open a PR. A human
 #                 reviews and merges it; nothing auto-merges.
 #
 # TAG SCHEME — deliberately version tags (ci-claude:<claude-version>), NOT the dated
@@ -107,31 +107,22 @@ steps:
          [0-9]*.[0-9]*.[0-9]*) : ;;
          *) echo "npm returned an implausible version '$$NEW' — aborting."; exit 1 ;;
        esac
-        # The tag every consumer currently references. They are kept identical,
-        # so this must resolve to EXACTLY ONE non-empty tag; anything else
-        # (no match, or a partial/disagreeing set) aborts rather than risk a
-        # corrupting sed later. Counting with `grep -c .` (non-empty lines)
-        # instead of `wc -l` is deliberate: an empty result yields 0 here but
-        # 1 under wc -l, which would slip past the guard.
-        CUR=$$(grep -hoE 'ci-claude:[0-9A-Za-z._-]+' $$CONSUMERS | sed 's/^ci-claude://' | sort -u)
-        NTAGS=$$(printf '%s' "$$CUR" | grep -c . || true)
-        if [ -z "$$CUR" ] || [ "$$NTAGS" -ne 1 ]; then
-          echo "Expected exactly one ci-claude tag across the consumers, got [$$CUR]."
-          echo "Refusing to guess — fix the pins by hand."; exit 1
-        fi
-        echo "latest=$$NEW  current=$$CUR  event=$$CI_PIPELINE_EVENT"
+        # Distinct tags the consumers currently reference. Divergent pins are
+        # normal (a pipeline bumped on its own); the bump rewrites all of them.
+        CUR=$$(grep -hoE 'ci-claude:[0-9][0-9A-Za-z._-]*' $$CONSUMERS | sed 's/^ci-claude://' | sort -u | tr '\n' ' ' | sed 's/ $$//')
+        [ -n "$$CUR" ] || { echo "No ci-claude pin found in the consumers — aborting."; exit 1; }
+        echo "latest=$$NEW  current=[$$CUR]  event=$$CI_PIPELINE_EVENT"
        printf '%s' "$$NEW" > "$$CI_WORKSPACE/.ci-refresh/version"
        printf '%s' "$$CUR" > "$$CI_WORKSPACE/.ci-refresh/current"
-        # Build when the version changed, or on a push (Containerfile edit must
-        # ship even if the version is unchanged). Bump the pins only when the
-        # version the consumers reference is not the latest.
+        # Build when any pin is off latest, or on a push (Containerfile edit must
+        # ship even if the version is unchanged). Bump when any pin is off latest.
        if [ "$$CUR" != "$$NEW" ] || [ "$$CI_PIPELINE_EVENT" = "push" ]; then
          touch "$$CI_WORKSPACE/.ci-refresh/do-build"; echo "will build $$NEW"
        else
          echo "already on $$NEW — nothing to build"
        fi
        if [ "$$CUR" != "$$NEW" ]; then
-          touch "$$CI_WORKSPACE/.ci-refresh/do-bump"; echo "will open bump PR $$CUR -> $$NEW"
+          touch "$$CI_WORKSPACE/.ci-refresh/do-bump"; echo "will open bump PR [$$CUR] -> $$NEW"
        else
          echo "consumers already pin $$NEW — no PR"
        fi
@@ -175,6 +166,9 @@ steps:
      - |
        [ -f "$$CI_WORKSPACE/.ci-refresh/do-bump" ] || { echo "no version change — no PR"; exit 0; }
        set -eu
+        # Secret values can carry a trailing newline; it breaks the push URL and the auth header.
+        PUSH_TOKEN=$$(printf '%s' "$$PUSH_TOKEN" | tr -d '\r\n')
+        [ -n "$$PUSH_TOKEN" ] || { echo "PUSH_TOKEN is empty — aborting"; exit 1; }
        NEW=$$(cat "$$CI_WORKSPACE/.ci-refresh/version")
        CUR=$$(cat "$$CI_WORKSPACE/.ci-refresh/current")
        # Both were validated non-empty by the decide step; re-assert before a
@@ -190,9 +184,11 @@ steps:
        # is not a maintained list.
        CONSUMERS=$$(grep -rlE 'ci-claude:[0-9]' .woodpecker)
        [ -n "$$CONSUMERS" ] || { echo "No ci-claude consumers found — refusing to continue."; exit 1; }
-        # Exact-string swap of the current tag for the new one, consumers only.
+        # NEW goes into a sed replacement: allow only a plain x.y.z version.
+        printf '%s' "$$NEW" | grep -qE '^[0-9]+\.[0-9]+\.[0-9]+$$' || { echo "refusing non-x.y.z version '$$NEW'"; exit 1; }
+        # Every version-pinned tag becomes NEW, whatever it was; consumers only.
        for f in $$CONSUMERS; do
-          sed -i "s#ci-claude:$$CUR#ci-claude:$$NEW#g" "$$f"
+          sed -i "s#ci-claude:[0-9][0-9A-Za-z._-]*#ci-claude:$$NEW#g" "$$f"
        done
        # Author is claude-bot (the PAT's account); the commit subject makes
        # the automated-bump nature clear.
@@ -17,6 +17,6 @@ when:

 steps:
  - name: model-selection-tests
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    commands:
      - sh scripts/ci/test-model-selection.sh
@@ -26,7 +26,7 @@ steps:
    # fetch-models.sh response mapping. No cluster, no secrets, no network — the
    # ci-claude image just provides sh + jq. Keeps the flag-injection regression
    # test from rotting (it exercises the exact injection shapes).
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    commands:
      - sh scripts/ci/test-model-selection.sh

@@ -62,7 +62,7 @@ steps:
      - bash scripts/ci/make-diag-kubeconfig.sh

  - name: issue-fix-agent
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
        from_secret: claude_code_oauth_token
@@ -65,7 +65,7 @@ when:

 steps:
  - name: pr-review-agent
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
        from_secret: claude_code_oauth_token
@@ -70,7 +70,7 @@ steps:
    # this step is skipped — otherwise we'd fork-bomb PRs.
    when:
      - branch: master
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
        from_secret: claude_code_oauth_token
@@ -30,7 +30,7 @@ when:

 steps:
  - name: update-audit-agent
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.250
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
        from_secret: claude_code_oauth_token
@@ -16,7 +16,7 @@ when:

 steps:
  - name: checkout-notes
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.278
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    pull: true
    environment:
      WIKI_BOT_TOKEN:
@@ -32,7 +32,7 @@ steps:
        echo "raw backlog: $$(ls notes/raw/*.md | grep -v _index.md | wc -l) file(s)"

  - name: distill
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.278
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    pull: true
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
@@ -62,7 +62,7 @@ steps:
          -p

  - name: audit
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.278
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    pull: true
    environment:
      CLAUDE_CODE_OAUTH_TOKEN:
@@ -91,7 +91,7 @@ steps:
          -p

  - name: push-notes
-    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.278
+    image: gitea.renner.uno/gitea_admin/ci-claude:2.1.280
    pull: true
    environment:
      WIKI_BOT_TOKEN:
@@ -1,4 +1,4 @@
-# netbird-proxy-services: reconciles NetBird reverse-proxy services (public
+# netbird-service-sync: reconciles NetBird reverse-proxy services (public
 # domains -> TLS passthrough -> Traefik ClusterIP) from Ingresses carrying the
 # public label. Standard library only, so the image is the interpreter plus one
 # file; runs unprivileged as its own user.
@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-"""netbird-proxy-services: reconcile NetBird reverse-proxy services from Ingresses.
+"""netbird-service-sync: reconcile NetBird reverse-proxy services from Ingresses.

 Source of truth: every Ingress carrying the public label (dns.renner.uno/public).
 For each host on such an Ingress the NetBird management gets one reverse-proxy
@@ -30,11 +30,11 @@ import urllib.error
 import urllib.parse
 import urllib.request

-LOG = logging.getLogger("netbird-proxy-services")
+LOG = logging.getLogger("netbird-service-sync")

 NB = os.environ.get("NB_MANAGEMENT_URL", "https://netbird.renner.uno").rstrip("/")
 BAO_ADDR = os.environ.get("BAO_ADDR", "http://openbao.openbao.svc.cluster.local:8200").rstrip("/")
-BAO_ROLE = os.environ.get("BAO_ROLE", "netbird-proxy-services")
+BAO_ROLE = os.environ.get("BAO_ROLE", "netbird-service-sync")
 BAO_PAT_PATH = os.environ.get("BAO_PAT_PATH", "homelab/data/infrastructure/secret-rotators-bootstrap/netbird")
 BAO_PAT_KEY = os.environ.get("BAO_PAT_KEY", "pat")
 SERVICE_CIDR = os.environ.get("SERVICE_CIDR", "10.96.0.0/12")
@@ -156,7 +156,7 @@ def ensure_network(nb):
        if n["name"] == NETWORK_NAME:
            return n["id"]
    n = nb.post("networks", {"name": NETWORK_NAME,
-                             "description": "managed by netbird-proxy-services; public ingress via TLS passthrough"})
+                             "description": "managed by netbird-service-sync; public ingress via TLS passthrough"})
    LOG.info("created network %s (%s)", NETWORK_NAME, n["id"])
    return n["id"]

@@ -166,7 +166,7 @@ def ensure_resource(nb, net):
        if r.get("address") == SERVICE_CIDR:
            return r["id"]
    r = nb.post(f"networks/{net}/resources", {"name": "cluster service CIDR", "address": SERVICE_CIDR,
-                                              "description": "managed by netbird-proxy-services",
+                                              "description": "managed by netbird-service-sync",
                                              "enabled": True, "groups": []})
    LOG.info("created subnet resource %s (%s)", SERVICE_CIDR, r["id"])
    return r["id"]
@@ -1,117 +1,99 @@
 # Traffic classes on the k8s nodes

-Issue #576. Three of the four classes are separated; management is not.
+Issue #576. Five classes, each on its own VLAN.

 ## Where each class runs

 | class | link | mechanism |
 |---|---|---|
 | Ceph replication | `ceph-san`, VLAN 49 | the whole link is handed to the OSD pod (Multus `host-device`) |
-| world→pod | `ingress0`, VLAN 47 | BGP session sourced there, so the advertised next hop is that address |
-| pod→world | `ingress0`, VLAN 47 | routing rule on the pod source prefix → table 47 |
-| pod↔pod | `pod0`, VLAN 48 | routing rule on pod source **and** destination → table 48 |
-| management | `mgmt0`, VLAN 40 | not started |
+| pod↔pod | `pod0`, VLAN 48 | pod-sourced traffic is routed from table 47; bird puts the peers' pod CIDRs there via `pod0` |
+| world↔pod | `ingress0`, VLAN 47 | BGP session sourced there (next hop of every LoadBalancer VIP); pod egress takes table 47's default |
+| node↔node | `node0`, VLAN 46 | the node InternalIP and etcd live there; so do apiserver↔kubelet, cilium-health and host→remote-pod traffic |
+| world↔node | `mgmt0`, VLAN 40 | API VIP for kubectl, talosctl from outside, node egress (DNS, NTP, image pulls) |

 Hard constraint, set by the owner: inter-node and pod traffic never traverses
-the UDM. Everything above stays on the 10G switch.
+the UDM. Everything but world↔node stays on the 10G switch. VLAN 46 and 48 are
+statically addressed, without a gateway; VLAN 46 is isolated in UniFi.

-## Why policy routing rather than moving the node IP
+## Pod traffic: one table

-Cilium reaches a peer's pods at that peer's Kubernetes InternalIP, in native
-and tunnel mode alike. `--ipv4-node` is overwritten from the Node object, and
-`nodePort.directRoutingDevice` selects the BPF NodePort device, not a next hop.
-Both were established by test (#837, #838), not by reading.
+Two rules, one per family, send every packet with a pod source to table 47
+(`talos/patches/site-homelab.yaml`, priorities 1000/1001, below main). The
+table decides by longest prefix; no rule carves out exceptions.

-So the InternalIP decides where pod traffic goes — unless the lookup happens in
-a different table. The rules do that, and nothing about the node's identity
-moves. The routing decision runs before Cilium's masquerade hook, so a rule
-matching the pod source sees the pod address, and the masquerade then rewrites
-it to the egress link's.
+bird (`kubernetes/infrastructure/bird/`) owns table 47 and fills it with:

-## Rule order
+- the node's attached networks — `mgmt0`, `ingress0`, `pod0`, `node0` —
+  taken from the interfaces;
+- the peers' pod CIDRs, learned over an iBGP mesh on `pod0`, next hop the
+  peer's `pod0` address;
+- the node's own pod CIDR via `cilium_host`, originated rather than read back
+  from the kernel (reading it back made BIRD depend on the shape of Cilium's
+  route, and KRT rejects the IPv4 one);
+- the egress default via the UDM on `ingress0`, from the node annotations
+  `net.renner.uno/egress-gw-v4/v6`.

-Lower priority runs first.
+`node0` has to be in the table: a pod reaching a node address — the
+`kubernetes` ClusterIP, which socket LB rewrites to the apiserver's advertise
+address, or a hostNetwork pod — would otherwise take the default to the UDM.

-| prio | match | table |
-|---|---|---|
-| 980 | src pod, dst pod | 48 |
-| 981 | src pod, dst pod (v4) | 48 |
-| 990 | dst `fd00::/8` | main |
-| 991 | dst `10.0.0.0/8` | main |
-| 1000 | src pod | 47 |
-| 1001 | src pod (v4) | 47 |
+The routing decision runs before Cilium's masquerade hook, so the rule sees the
+pod address and the masquerade then rewrites it to the egress link's.

-980/981 match source **and** destination. Source alone would also catch the
-node's own traffic to a pod, which would then leave with a `pod0` source — an
-address `CiliumNode.spec.addresses` does not list, so Cilium classifies it as
-`world` rather than `host`.
-
-990/991 are Cilium's `ipv4/ipv6NativeRoutingCIDR`, and they are what keeps
-cluster-internal traffic out of table 47. Without them a pod's packet to a
-service matches the source rule, finds only a default route, and leaves for the
-UDM — measured, in-cluster DNS stopped resolving.
-
-Traffic to a pod on the same node falls through on its own: table 48 holds only
-the peers' CIDRs, so the lookup misses and 990/991 send it to main.
-
-## Where each table comes from
-
- **47** — two default routes declared per node in `talos/patches/k8s-*.yaml`.
-  The IPv6 gateway is the UDM's configured address on the segment, not the
-  RA-learned link-local; `accept_ra_defrtr` stays `0` on `ingress0` so the RA
-  default does not land in main.
- **48** — filled by the `bird` DaemonSet: an iBGP mesh over `pod0`, one route
-  per peer pod CIDR with the peer's `pod0` address as next hop.
-
-## The bird mesh
-
-`kubernetes/infrastructure/bird/`. Each node advertises only its own pod CIDR
-and installs what it learns into table 48. Cilium's routes in main are
-untouched, which is why the two coexist.
+Node-originated traffic does not match the rules and uses main, where Cilium
+routes each peer's pod CIDR via the peer's InternalIP — on `node0`.

 No address is written into this repo twice. Each node publishes its `pod0`
-addresses as `nodeAnnotations`, beside the `VLANConfig` that assigns them; the
-pod CIDRs come from the Node objects. An init container renders both from the
-API with `kubectl -o go-template-file`, excludes the node itself with a field
-selector, and ends with `bird -p` so BIRD's own parser gates startup.
+addresses as `nodeAnnotations` beside the `VLANConfig` that assigns them; the
+pod CIDRs come from the Node objects. bird's init container renders both from
+the API and ends with `bird -p`, so BIRD's own parser gates startup.

-The node's own CIDR is **originated**, not read back from the kernel. Reading
-it back made BIRD depend on the shape Cilium installs, and the IPv4 route is
-rejected outright: its next hop sits inside the prefix it describes, which KRT
-refuses with `strange next-hop`. The IPv6 route is a device route and passed,
-so the failure was silent on one family only.
+## Node↔node
+
+`kubelet.nodeIP.validSubnets` and `cluster.etcd.advertisedSubnets` point at
+VLAN 46. Talos picks one InternalIP per family, IPv4 first — the service CIDR
+is IPv4, and the apiserver's advertise address must match it.
+
+The API VIP stays on `mgmt0` for kubectl. No in-cluster client uses it:
+kubelet, scheduler, controller-manager and Cilium go through KubePrism
+(`localhost:7445`), which balances over every address of every control-plane
+member; pods use the ClusterIP.
+
+Talos never updates an existing etcd member's peer URLs
+(siderolabs/talos#12720). Changing `advertisedSubnets` means, per member:
+stage the config, `talosctl etcd forfeit-leadership`, `talosctl etcd leave`,
+reboot. `leave` removes only `/var/lib/etcd`; the node rejoins as a learner
+and is promoted. EPHEMERAL does not need a wipe.
+
+## Moving node IPs: the mixed state
+
+While some nodes have their InternalIP on the new network and some on the old,
+host→remote-pod traffic leaves with a node address that is not an InternalIP:
+the kernel picks the source from the route to the peer's InternalIP. Cilium
+classifies that as `world`, and every ingress policy that admits only
+`host`/`remote-node` (rook-ceph, monitoring, …) drops it. Kernel Ceph clients
+lose the OSDs on the other side; RBD mounts hang.
+
+Bridge it for the duration with a temporary `CiliumClusterwideNetworkPolicy`
+admitting the node subnets of both networks, both families, with
+`enableDefaultDeny: false` so it restricts nothing else, and delete it once
+every node has moved. Keep all OSD nodes schedulable while one is drained:
+draining two takes Ceph below its data availability.

 ## Failure behaviour

-Benign at every layer. An empty or absent table 48 makes 980/981 fall through
-to main, and pod traffic returns to VLAN 40 — unseparated, not broken. The same
-holds if the DaemonSet cannot start, which is the case during a cold rebuild
-until Gitea serves its image.
-
-## Management, the remaining class
-
-Not started. The levers differ per service and none of them is the node IP:
-
-| service | can leave the node IP? | lever |
-|---|---|---|
-| etcd | yes | `cluster.etcd.advertisedSubnets`; needs a reboot per node |
-| kube-apiserver endpoint | yes | it is a VIP — `Layer2VIPConfig.link` |
-| Talos API | n/a | `apid` answers on every node address; a firewall question |
-| kubelet | no | it *is* the node IP |
-
-Two approaches were rejected by the owner and are not to be proposed again:
-moving the node IP and etcd to `pod0` (#838), and relabelling VLAN 40 as the
-cluster VLAN while moving etcd and the API VIP to a new management VLAN.
-Replacing the CNI (#839, Antrea) was shelved.
-
-Note that the reasoning which produced that list also concluded pod↔pod was
-blocked, which policy routing disproved. Treat it as a record of decisions, not
-as an exhaustive analysis.
+If bird stops, table 47 empties, the rules find nothing and fall through to
+main: pod traffic returns to the node networks, unseparated rather than
+misrouted. The same holds during a cold rebuild until Gitea serves bird's
+image. The default route lives in table 47, not in the Talos config, so a
+default cannot outlive the specific routes and send pod traffic to the UDM.

 ## Sources

-Cilium v1.20.1 `pkg/datapath/linux/node.go`, `pkg/node/types/node.go`,
-`pkg/node/sync/local_node_sync.go`; BIRD 2.17 user manual (kernel protocol,
-`learn all`, `strict bind`, `direct`); Talos v1.13 `RoutingRuleConfig` and
-`machine.nodeAnnotations`. PRs #831, #833, #835, #836, #837, #842, #852, #854,
-#855; closed unmerged: #830, #838, #839.
+Cilium v1.20 `pkg/datapath/linux/node.go`; BIRD 2.17 user manual (kernel
+protocol, `strict bind`, `direct`); Talos v1.13 `RoutingRuleConfig`,
+`machine.nodeAnnotations`, configuration patching (lists are appended,
+`$patch: delete`), etcd `LeaveCluster`; KubePrism docs;
+siderolabs/talos#12720. PRs #831, #833, #835, #836, #837, #842, #852, #854,
+#855, #857, #858, #918, #919, #922; closed unmerged: #830, #838, #839.
@@ -2,7 +2,7 @@ apiVersion: networking.k8s.io/v1
 kind: Ingress
 metadata:
  name: gitea
-  namespace: apps
+  namespace: gitea
  labels:
    app.kubernetes.io/name: gitea
  annotations:
@@ -2,7 +2,7 @@ apiVersion: traefik.io/v1alpha1
 kind: IngressRouteTCP
 metadata:
  name: gitea-ssh
-  namespace: apps
+  namespace: gitea
  labels:
    app.kubernetes.io/name: gitea
 spec:
@@ -1,28 +1,8 @@
-apiVersion: cilium.io/v2
-kind: CiliumNetworkPolicy
-metadata:
-  name: gitea-allow-metrics-scrape
-  namespace: apps
-spec:
-  # Prometheus (monitoring ns) scrapes gitea's /metrics on :3000. The apps
-  # default-deny-ingress only allows same-ns / home-assistant / host, so the
-  # scrape is dropped (context deadline exceeded). This became necessary only
-  # after the LXC->in-cluster cutover — the old static target was external, so
-  # no in-cluster policy applied. Scoped to the gitea pod + :3000.
-  endpointSelector:
-    matchLabels:
-      app.kubernetes.io/name: gitea
-  ingress:
-    - fromEndpoints:
-        - matchLabels:
-            "k8s:io.kubernetes.pod.namespace": monitoring
-      toPorts:
-        - ports:
-            - port: "3000"
-              protocol: TCP
---
+# Prometheus scrape and Traefik are admitted by the namespace baseline in
+# infrastructure/network-policies/gitea-ns.yaml.
+#
 # Secret rotation: the flux-git rotator (openbao ns) reaches gitea to mint a new
-# flux PAT. It uses the in-cluster Service (gitea.apps.svc:3000) — the public
+# flux PAT. It uses the in-cluster Service (gitea.gitea.svc:3000) — the public
 # gitea.renner.uno name hairpins off the ingress LB from a pod (same failure as
 # pve.renner.uno). Scoped to the gitea pod :3000 and the rotator label; the
 # egress half is rotator-gitea-flux-egress in network-policies/openbao-ns.yaml.
@@ -30,7 +10,7 @@ apiVersion: cilium.io/v2
 kind: CiliumNetworkPolicy
 metadata:
  name: gitea-allow-flux-rotator
-  namespace: apps
+  namespace: gitea
 spec:
  endpointSelector:
    matchLabels:
@@ -54,7 +34,7 @@ apiVersion: cilium.io/v2
 kind: CiliumNetworkPolicy
 metadata:
  name: gitea-allow-review-bot-rotator
-  namespace: apps
+  namespace: gitea
 spec:
  endpointSelector:
    matchLabels:
@@ -13,7 +13,7 @@ apiVersion: policy/v1
 kind: PodDisruptionBudget
 metadata:
  name: gitea
-  namespace: apps
+  namespace: gitea
  labels:
    app.kubernetes.io/name: gitea
 spec:
@@ -2,7 +2,7 @@ apiVersion: v1
 kind: Service
 metadata:
  name: gitea
-  namespace: apps
+  namespace: gitea
  labels:
    app.kubernetes.io/name: gitea
 spec:
@@ -2,7 +2,7 @@ apiVersion: apps/v1
 kind: StatefulSet
 metadata:
  name: gitea
-  namespace: apps
+  namespace: gitea
  labels:
    app.kubernetes.io/name: gitea
 spec:
@@ -152,7 +152,7 @@ spec:
          # NOTE: volumeClaimTemplates is immutable on an existing StatefulSet,
          # so bumping this value cannot be `kubectl apply`-ed in place — Flux
          # will reject the whole object. Landing this size change requires a
-          # one-time `kubectl delete statefulset gitea -n apps --cascade=orphan`
+          # one-time `kubectl delete statefulset gitea -n gitea --cascade=orphan`
          # (pod + PVC keep running) so Flux re-creates the STS from this
          # template and re-adopts them. After the cleanup_packages GC reclaims
          # the ~10G of orphaned blobs, real usage is ~9G, so 40Gi is headroom.
@@ -3,8 +3,8 @@ kind: Kustomization
 resources:
  - gatus/
  - external-dns/
-  - gitea/
  - ical-proxy/
+  # gitea/ has its own `gitea` Kustomization (clusters/homelab/gitea.yaml, #878).
  # woodpecker/ moved to its own dedicated `woodpecker` Kustomization
  # (clusters/homelab/woodpecker.yaml) so its wp-pg CNPG cluster's barman
  # dependency + Cluster health-gating don't contaminate this shared apps KS.
@@ -91,12 +91,11 @@ spec:
        # Additional scrape targets not covered by ServiceMonitors.
        additionalScrapeConfigs:
          # Gitea exposes /metrics anonymously on its main HTTP port. The
-          # `gitea_*` metrics power the gitea-overview dashboard. Gitea now runs
-          # in-cluster (StatefulSet in apps ns; LXC 114 retired), so scrape the
-          # ClusterIP Service directly instead of the dead LXC hostname.
+          # `gitea_*` metrics power the gitea-overview dashboard. Scrapes the
+          # ClusterIP Service of the StatefulSet in the gitea namespace.
          - job_name: gitea
            static_configs:
-              - targets: ["gitea.apps.svc.cluster.local:3000"]
+              - targets: ["gitea.gitea.svc.cluster.local:3000"]
            metrics_path: /metrics

    # Grafana
@@ -1946,6 +1946,7 @@ if phase_runs wait-apps; then
    # apps waits on infrastructure-ceph-cluster (Ceph mon/mgr/OSD bring-up
    # is genuinely 5–10 min on a fresh OSD-wiped cluster).
    wait_kustomization apps                "apps"                                                90
+    wait_kustomization gitea               "Gitea (incl. data-gitea-0 PVC)"                      90
    wait_kustomization influxdb            "InfluxDB (HR install + token Jobs)"                  600
    wait_kustomization woodpecker          "Woodpecker CI (incl. wp-pg CNPG backend)"            600
    wait_kustomization authentik           "Authentik (incl. ak-pg CNPG backend)"                600
@@ -0,0 +1,20 @@
+apiVersion: kustomize.toolkit.fluxcd.io/v1
+kind: Kustomization
+metadata:
+  name: gitea
+  namespace: flux-system
+spec:
+  # Gitea in its own `gitea` namespace (#878), formerly under the shared `apps` KS.
+  interval: 10m
+  retryInterval: 2m
+  timeout: 15m
+  sourceRef:
+    kind: GitRepository
+    name: flux-system
+  path: ./kubernetes/apps/gitea
+  prune: true
+  wait: true
+  dependsOn:
+    - name: infrastructure
+    - name: infrastructure-crds
+    - name: infrastructure-ceph-cluster
@@ -23,7 +23,7 @@ protocol device { scan time 10; }
 # down. The usual advice is not to export these to the kernel because it
 # maintains them itself — true of main, not of a table the kernel never fills.
 protocol direct attached {
-  interface "mgmt0", "ingress0", "pod0";
+  interface "mgmt0", "ingress0", "pod0", "node0";
  ipv4 { table master4; };
  ipv6 { table master6; };
 }
@@ -48,8 +48,8 @@ spec:
        tier: node
        app: bird
    spec:
-      # Carries pod east-west once the routing rules land. Without it table 48
-      # stays empty and the rules fall through to main, i.e. back to VLAN 40.
+      # Owns table 47, where all pod traffic is routed. Without it the table
+      # stays empty and the rules fall through to main, unseparated.
      priorityClassName: system-node-critical
      # Writes host routes and peers on host addresses.
      hostNetwork: true
@@ -118,14 +118,21 @@ spec:
      # traffic out to the VIP and back (which fails with L2 announcements).
      lbExternalClusterIP: true
    cni:
-      # Multus (kubernetes/infrastructure/multus) runs alongside Cilium and
-      # writes 00-multus.conf to /etc/cni/net.d so kubelet picks multus
-      # first; multus then delegates to Cilium's 05-cilium.conflist via
-      # clusterNetwork. The default `exclusive: true` makes cilium-cni's
-      # install step rename anything alphabetically before its own config
-      # to *.cilium_bak on every reconcile, which kept nuking multus.
-      # Setting it false lets multus's config coexist.
+      # Multus is the only writer of /etc/cni/net.d (#832): Cilium's conflist
+      # goes to /etc/cni/net.d/multus/, which the container runtime does not
+      # read, and Multus delegates to it via clusterNetwork. Without this a
+      # pod created after a reboot, before Multus rewrote 00-multus.conf, came
+      # up with Cilium alone. See kubernetes/infrastructure/multus.
+      confPath: /etc/cni/net.d/multus
+      # The default `exclusive: true` renames other configs in the conf dir to
+      # *.cilium_bak on every reconcile, which kept nuking multus.
      exclusive: false
+    # One agent at a time: the chart default of 2 restarts two of three
+    # nodes' datapath at once.
+    updateStrategy:
+      type: RollingUpdate
+      rollingUpdate:
+        maxUnavailable: 1
    # Talos-specific settings
    securityContext:
      capabilities:
@@ -19,7 +19,7 @@ resources:
  - cloudnative-pg/
  - mariadb-operator/
  - netbird-ingress/
-  - netbird-proxy-services/
+  - netbird-service-sync/
  - coredns-config/
  - k8s-gateway/
  - external-dns-unifi/
@@ -111,10 +111,18 @@ metadata:
  name: multus
  namespace: kube-system
 ---
+# Multus is the only writer of /etc/cni/net.d (#832). Cilium writes its conflist
+# to /etc/cni/net.d/multus/, a subdirectory the container runtime does not read
+# (libcni skips directories); Multus picks it up there via multusAutoconfigDir
+# and writes 00-multus.conf. Until that file exists the runtime has no network
+# config at all, so no pod can be created with Cilium alone after a reboot.
+#
+# The "-subdir" names date from the node-by-node rollout; renaming would roll
+# Multus again for nothing.
 kind: ConfigMap
 apiVersion: v1
 metadata:
-  name: multus-daemon-config
+  name: multus-daemon-config-subdir
  namespace: kube-system
  labels:
    tier: node
@@ -127,7 +135,7 @@ data:
        "logLevel": "verbose",
        "logToStderr": true,
        "cniConfigDir": "/host/etc/cni/net.d",
-        "multusAutoconfigDir": "/host/etc/cni/net.d",
+        "multusAutoconfigDir": "/host/etc/cni/net.d/multus",
        "multusConfigFile": "auto",
        "socketDir": "/host/run/multus/"
    }
@@ -135,16 +143,16 @@ data:
 apiVersion: apps/v1
 kind: DaemonSet
 metadata:
-  name: kube-multus-ds
+  name: kube-multus-ds-subdir
  namespace: kube-system
  labels:
    tier: node
    app: multus
-    name: multus
+    name: multus-subdir
 spec:
  selector:
    matchLabels:
-      name: multus
+      name: multus-subdir
  updateStrategy:
    type: RollingUpdate
  template:
@@ -152,7 +160,7 @@ spec:
      labels:
        tier: node
        app: multus
-        name: multus
+        name: multus-subdir
    spec:
      # Upstream ships this DaemonSet with no priority class, so it runs at 0
      # and can preempt nothing. Any event that makes the pod need a fresh
@@ -224,6 +232,10 @@ spec:
          volumeMounts:
            - name: cni
              mountPath: /host/etc/cni/net.d
+            # Exists before Multus starts, so it waits for Cilium's conflist
+            # instead of exiting on a missing directory.
+            - name: cni-cilium-conf
+              mountPath: /host/etc/cni/net.d/multus
            # multus-daemon expects that cnibin path must be identical between pod and container host.
            # e.g. if the cni bin is in '/opt/cni/bin' on the container host side, then it should be mount to '/opt/cni/bin' in multus-daemon,
            # not to any other directory, like '/opt/bin' or '/usr/bin'.
@@ -279,6 +291,10 @@ spec:
        - name: cni
          hostPath:
            path: /etc/cni/net.d
+        - name: cni-cilium-conf
+          hostPath:
+            path: /etc/cni/net.d/multus
+            type: DirectoryOrCreate
        - name: cnibin
          hostPath:
            path: /opt/cni/bin
@@ -287,7 +303,7 @@ spec:
            path: /
        - name: multus-daemon-config
          configMap:
-            name: multus-daemon-config
+            name: multus-daemon-config-subdir
            items:
              - key: daemon-config.json
                path: daemon-config.json
@@ -39,9 +39,11 @@ spec:
        name: whereabouts-reconciler
    spec:
      serviceAccountName: whereabouts
-      # Every node here is a control plane node.
+      # Every node here is a control plane node. Keyed: a keyless Exists would
+      # also tolerate node.kubernetes.io/unschedulable and ignore cordons.
      tolerations:
-        - operator: Exists
+        - key: node-role.kubernetes.io/control-plane
+          operator: Exists
          effect: NoSchedule
      containers:
        - name: whereabouts-reconciler
@@ -179,6 +179,22 @@ kind: Namespace
 metadata:
  name: woodpecker
 ---
+# Gitea, moving out of the shared `apps` namespace (issue #878). Declared here
+# like `woodpecker` above, so it exists before infrastructure-crds applies its
+# CiliumNetworkPolicies. No pod-security labels: the cluster default applies.
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: gitea
+---
+# netbird-service-sync, moving out of `openbao` (issue #876). Declared here like
+# `woodpecker` above, so it exists before infrastructure-crds applies its
+# CiliumNetworkPolicies. No pod-security labels: the cluster default applies.
+apiVersion: v1
+kind: Namespace
+metadata:
+  name: netbird-service-sync
+---
 # Local LLM inference (kubernetes/apps/llama-server). Declared here, like
 # `woodpecker` above, so the `infrastructure` KS creates it before
 # `infrastructure-crds` applies the namespace's CiliumNetworkPolicies.
@@ -1,28 +1,28 @@
 # Reconciles NetBird reverse-proxy services from Ingresses that carry
 # dns.renner.uno/public: one TLS-passthrough service per host on :443,
 # targeting Traefik's ClusterIP through a subnet resource over the service
-# CIDR (ci/images/netbird-proxy-services/controller.py).
+# CIDR (ci/images/netbird-service-sync/controller.py).
 #
-# openbao namespace, like the rotators: its default-deny already allows egress
-# to OpenBao, DNS, the apiserver and `world` (netbird.renner.uno), and the
-# NetBird PAT is read from OpenBao on every pass through the
-# netbird-proxy-services role — policies/netbird-proxy-services.hcl, applied
-# once by hand like the rotator roles. Until that role exists the pod logs an
-# auth error each minute and does nothing.
+# Own namespace; its egress (DNS, apiserver, OpenBao, NetBird API) is in
+# network-policies/netbird-service-sync-ns.yaml. The NetBird PAT is read from
+# OpenBao on every pass through the netbird-service-sync role —
+# policies/netbird-service-sync.hcl, applied once by hand like the rotator
+# roles. Until that role exists the pod logs an auth error each minute and does
+# nothing.
 apiVersion: v1
 kind: ServiceAccount
 metadata:
-  name: netbird-proxy-services
-  namespace: openbao
+  name: netbird-service-sync
+  namespace: netbird-service-sync
  labels:
-    app.kubernetes.io/name: netbird-proxy-services
+    app.kubernetes.io/name: netbird-service-sync
 ---
 # Read-only, cluster-wide: public Ingresses live in many namespaces, the
 # Traefik Service in ingress/.
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRole
 metadata:
-  name: netbird-proxy-services-read
+  name: netbird-service-sync-read
 rules:
  - apiGroups: ["networking.k8s.io"]
    resources: ["ingresses"]
@@ -34,23 +34,23 @@ rules:
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRoleBinding
 metadata:
-  name: netbird-proxy-services-read
+  name: netbird-service-sync-read
 roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
-  name: netbird-proxy-services-read
+  name: netbird-service-sync-read
 subjects:
  - kind: ServiceAccount
-    name: netbird-proxy-services
-    namespace: openbao
+    name: netbird-service-sync
+    namespace: netbird-service-sync
 ---
 apiVersion: apps/v1
 kind: Deployment
 metadata:
-  name: netbird-proxy-services
-  namespace: openbao
+  name: netbird-service-sync
+  namespace: netbird-service-sync
  labels:
-    app.kubernetes.io/name: netbird-proxy-services
+    app.kubernetes.io/name: netbird-service-sync
 spec:
  replicas: 1
  # One writer against the management API at a time.
@@ -58,13 +58,13 @@ spec:
    type: Recreate
  selector:
    matchLabels:
-      app.kubernetes.io/name: netbird-proxy-services
+      app.kubernetes.io/name: netbird-service-sync
  template:
    metadata:
      labels:
-        app.kubernetes.io/name: netbird-proxy-services
+        app.kubernetes.io/name: netbird-service-sync
    spec:
-      serviceAccountName: netbird-proxy-services
+      serviceAccountName: netbird-service-sync
      securityContext:
        runAsNonRoot: true
        # Numeric, or the kubelet refuses the image's named user with
@@ -80,7 +80,7 @@ spec:
          # image directory (branch and master alike) and the registry keeps
          # only the newest manifest per tag, so the digest must come from
          # the MASTER build of the code change.
-          image: gitea.renner.uno/gitea_admin/netbird-proxy-services:2026-09-07@sha256:caffa308d802c6fcc6aea919ac94657c08b993651baa6b8d18de395bcc40523e
+          image: gitea.renner.uno/gitea_admin/netbird-service-sync:2026-09-23@sha256:64f6057c508640e1dbc50457dc26e9b2de058a111fd46ca096021d7444b5bbff
          securityContext:
            allowPrivilegeEscalation: false
            readOnlyRootFilesystem: true
@@ -92,7 +92,7 @@ spec:
            - name: BAO_ADDR
              value: http://openbao.openbao.svc.cluster.local:8200
            - name: BAO_ROLE
-              value: netbird-proxy-services
+              value: netbird-service-sync
            # Must match cluster.network.serviceSubnets in talos/patches/cluster.yaml.
            - name: SERVICE_CIDR
              value: "10.96.0.0/12"
@@ -167,7 +167,7 @@ spec:
            "k8s:io.kubernetes.pod.namespace": ingress
            app.kubernetes.io/name: traefik
        - matchLabels:
-            "k8s:io.kubernetes.pod.namespace": apps
+            "k8s:io.kubernetes.pod.namespace": gitea
            app.kubernetes.io/name: gitea
      toPorts:
        - ports:
@@ -45,16 +45,6 @@ spec:
        - ports:
            - port: "1883"
              protocol: TCP
-    # gitea delivers the notes repo's push webhook to wiki-index (ai ns) on :8080.
-    # wiki-index's own ingress policy scopes this to the gitea pod.
-    - toEndpoints:
-        - matchLabels:
-            "k8s:io.kubernetes.pod.namespace": ai
-            app.kubernetes.io/name: wiki-index
-      toPorts:
-        - ports:
-            - port: "8080"
-              protocol: TCP
    - toEntities:
        - host
        - remote-node
@@ -0,0 +1,104 @@
+---
+# Baseline policy for the `gitea` namespace (issue #878).
+#
+# endpointSelector: {} — every rule applies to every pod in the namespace.
+# Workload-scoped grants go in their own policy with their own selector.
+apiVersion: cilium.io/v2
+kind: CiliumNetworkPolicy
+metadata:
+  name: default-deny
+  namespace: gitea
+spec:
+  endpointSelector: {}
+  ingress:
+    - fromEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": gitea
+    # Prometheus scrape of /metrics.
+    - fromEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": monitoring
+    # Kubelet probes and hostNetwork siblings.
+    - fromEntities:
+        - host
+        - remote-node
+    # ICMP from any cluster pod; required once an L4 policy selects the pod.
+    # See woodpecker-ns.yaml for the rationale. INGRESS ONLY.
+    - fromEndpoints:
+        - matchExpressions:
+            - key: io.kubernetes.pod.namespace
+              operator: Exists
+      icmps:
+        - fields:
+            - { type: 0,   family: IPv4 }
+            - { type: 3,   family: IPv4 }
+            - { type: 8,   family: IPv4 }
+            - { type: 11,  family: IPv4 }
+            - { type: 12,  family: IPv4 }
+            - { type: 1,   family: IPv6 }
+            - { type: 2,   family: IPv6 }
+            - { type: 3,   family: IPv6 }
+            - { type: 4,   family: IPv6 }
+            - { type: 128, family: IPv6 }
+            - { type: 129, family: IPv6 }
+            - { type: 133, family: IPv6 }
+            - { type: 134, family: IPv6 }
+            - { type: 135, family: IPv6 }
+            - { type: 136, family: IPv6 }
+            - { type: 143, family: IPv6 }
+  egress:
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": gitea
+    # Webhooks to hosts served by Traefik (ci.renner.uno).
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": ingress
+            app.kubernetes.io/name: traefik
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": kube-system
+            k8s-app: kube-dns
+      toPorts:
+        - ports:
+            - port: "53"
+              protocol: UDP
+            - port: "53"
+              protocol: TCP
+    # Push webhook of the notes repo to wiki-index.
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": ai
+            app.kubernetes.io/name: wiki-index
+      toPorts:
+        - ports:
+            - port: "8080"
+              protocol: TCP
+    # Push mirror to GitHub and other external targets.
+    - toEntities:
+        - host
+        - remote-node
+        - world
+---
+# Traefik serves git.renner.uno / gitea.renner.uno (:3000) and the gitea-ssh
+# entrypoint (:2222). Lives beside the baseline so both land in one reconcile.
+apiVersion: cilium.io/v2
+kind: CiliumNetworkPolicy
+metadata:
+  name: gitea-allow-traefik
+  namespace: gitea
+spec:
+  endpointSelector:
+    matchLabels:
+      app.kubernetes.io/name: gitea
+  ingress:
+    - fromEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": ingress
+            app.kubernetes.io/name: traefik
+      toPorts:
+        - ports:
+            - port: "3000"
+              protocol: TCP
+            - port: "2222"
+              protocol: TCP
@@ -22,6 +22,8 @@ resources:
  - external-secrets-ns.yaml
  - woodpecker-ci-ns.yaml
  - woodpecker-ns.yaml
+  - gitea-ns.yaml
+  - netbird-service-sync-ns.yaml
  - trmnl-ns.yaml
  - wordpress-ns.yaml
  - ai-ns.yaml
@@ -0,0 +1,43 @@
+---
+# Baseline policy for the `netbird-service-sync` namespace (issue #876).
+# The controller serves nothing; it only calls out: the Kubernetes API
+# (Ingress/Service reads), OpenBao (NetBird PAT) and the NetBird management API.
+apiVersion: cilium.io/v2
+kind: CiliumNetworkPolicy
+metadata:
+  name: default-deny
+  namespace: netbird-service-sync
+spec:
+  endpointSelector: {}
+  ingress:
+    - fromEntities:
+        - host
+        - remote-node
+  egress:
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": kube-system
+            k8s-app: kube-dns
+      toPorts:
+        - ports:
+            - port: "53"
+              protocol: UDP
+            - port: "53"
+              protocol: TCP
+    - toEntities:
+        - kube-apiserver
+    - toEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": openbao
+            app.kubernetes.io/name: openbao
+      toPorts:
+        - ports:
+            - port: "8200"
+              protocol: TCP
+    # NetBird management API (netbird.renner.uno).
+    - toEntities:
+        - world
+      toPorts:
+        - ports:
+            - port: "443"
+              protocol: TCP
@@ -41,6 +41,16 @@ spec:
        - ports:
            - port: "8200"
              protocol: TCP
+    # netbird-service-sync logs in via auth/kubernetes/role/netbird-service-sync
+    # and reads the NetBird PAT on every pass.
+    - fromEndpoints:
+        - matchLabels:
+            "k8s:io.kubernetes.pod.namespace": netbird-service-sync
+            app.kubernetes.io/name: netbird-service-sync
+      toPorts:
+        - ports:
+            - port: "8200"
+              protocol: TCP
    # (Secret-rotation CronJobs run in THIS namespace since 2026-07-18, so their
    # OpenBao access is covered by the intra-namespace ingress rule above — no
    # per-app rule needed.)
@@ -127,7 +137,7 @@ spec:
 # Egress complement to gitea-allow-flux-rotator (apps/gitea/networkpolicy.yaml).
 # The flux-git rotator mints a PAT against the in-cluster gitea Service; the
 # openbao default-deny only permits TCP egress to openbao/kube-dns/apiserver/
-# world, and gitea (apps ns) is in-cluster, so it needs an explicit allow.
+# world, and gitea (gitea ns) is in-cluster, so it needs an explicit allow.
 # Scoped to the flux-git rotator label and gitea :3000.
 apiVersion: cilium.io/v2
 kind: CiliumNetworkPolicy
@@ -142,7 +152,7 @@ spec:
  egress:
    - toEndpoints:
        - matchLabels:
-            "k8s:io.kubernetes.pod.namespace": apps
+            "k8s:io.kubernetes.pod.namespace": gitea
            app.kubernetes.io/name: gitea
      toPorts:
        - ports:
@@ -150,10 +160,10 @@ spec:
              protocol: TCP
 ---
 # Egress for the review-bot rotator. Its CronJob has two steps: the init step
-# mints review-bot's PAT against gitea (apps :3000), the main step upserts the
+# mints review-bot's PAT against gitea (gitea :3000), the main step upserts the
 # Woodpecker org secret against the woodpecker server (woodpecker :8000). Both run
 # under the same pod labels, so one egress policy covers both hops. Complements
-# gitea-allow-review-bot-rotator (apps) + woodpecker-allow-rotators (woodpecker).
+# gitea-allow-review-bot-rotator (gitea) + woodpecker-allow-rotators (woodpecker).
 apiVersion: cilium.io/v2
 kind: CiliumNetworkPolicy
 metadata:
@@ -167,7 +177,7 @@ spec:
  egress:
    - toEndpoints:
        - matchLabels:
-            "k8s:io.kubernetes.pod.namespace": apps
+            "k8s:io.kubernetes.pod.namespace": gitea
            app.kubernetes.io/name: gitea
      toPorts:
        - ports:
@@ -1,23 +0,0 @@
-# OpenBao policy: netbird-proxy-services
-#
-# The reverse-proxy-services controller (Deployment netbird-proxy-services in
-# the openbao namespace) reads the NetBird PAT on every reconcile pass and
-# calls the management API with it. Read-only on that one path; it shares the
-# bootstrap credential with rotator-netbird but not that role's PATCH rights.
-#
-# NOT applied by Flux. Apply ONCE with an admin token:
-#
-#   bao policy write netbird-proxy-services kubernetes/infrastructure/openbao/policies/netbird-proxy-services.hcl
-#   bao write auth/kubernetes/role/netbird-proxy-services \
-#     bound_service_account_names=netbird-proxy-services \
-#     bound_service_account_namespaces=openbao \
-#     policies=netbird-proxy-services ttl=15m
-#
-# The PAT itself is the one rotator-netbird already uses
-# (homelab/infrastructure/secret-rotators-bootstrap/netbird, key `pat`).
-#
-# Do NOT set token_no_default_policy=true (kv-v2 preflight needs the default policy).
-
-path "homelab/data/infrastructure/secret-rotators-bootstrap/netbird" {
-  capabilities = ["read"]
-}
@@ -0,0 +1,24 @@
+# OpenBao policy: netbird-service-sync
+#
+# The controller that syncs public Ingresses into NetBird reverse-proxy services
+# (Deployment netbird-service-sync in the netbird-service-sync namespace) reads
+# the NetBird PAT on every reconcile pass and calls the management API with it.
+# Read-only on that one path; it shares the bootstrap credential with
+# rotator-netbird but not that role's PATCH rights.
+#
+# NOT applied by Flux. Apply ONCE with an admin token:
+#
+#   bao policy write netbird-service-sync kubernetes/infrastructure/openbao/policies/netbird-service-sync.hcl
+#   bao write auth/kubernetes/role/netbird-service-sync \
+#     bound_service_account_names=netbird-service-sync \
+#     bound_service_account_namespaces=netbird-service-sync \
+#     policies=netbird-service-sync ttl=15m
+#
+# The PAT itself is the one rotator-netbird already uses
+# (homelab/infrastructure/secret-rotators-bootstrap/netbird, key `pat`).
+#
+# Do NOT set token_no_default_policy=true (kv-v2 preflight needs the default policy).
+
+path "homelab/data/infrastructure/secret-rotators-bootstrap/netbird" {
+  capabilities = ["read"]
+}
@@ -6,7 +6,7 @@
 # newest. A token that fails verification is deleted and the run aborts without
 # storing, so Flux is never stranded.
 #
-# gitea is reached at the in-cluster Service gitea.apps.svc:3000 — the public
+# gitea is reached at the in-cluster Service gitea.gitea.svc:3000 — the public
 # name hairpins off the ingress LB from a pod. Needs BOTH netpol halves
 # (gitea-allow-flux-rotator + rotator-gitea-flux-egress).
 #
@@ -76,7 +76,7 @@ spec:
                - name: BAO_PATH
                  value: homelab/infrastructure/flux-git
                - name: GITEA_URL
-                  value: http://gitea.apps.svc.cluster.local:3000
+                  value: http://gitea.gitea.svc.cluster.local:3000
                - name: GITEA_VERIFY_REPO
                  value: k.renner/homelab
                - name: GITEA_SCOPES
@@ -24,7 +24,7 @@
 # synced. Quarterly cadence + backoffLimit:1 makes a two-cycle failure unlikely,
 # but a failed sync should be fixed before the next run rather than left.
 #
-# gitea is reached in-cluster at gitea.apps.svc:3000, woodpecker at
+# gitea is reached in-cluster at gitea.gitea.svc:3000, woodpecker at
 # woodpecker.woodpecker.svc:8000 (it moved out of `apps` in the namespace
 # migration — see kubernetes/apps/woodpecker/MIGRATION-namespace.md). Needs
 # the netpol halves: gitea-allow-review-bot-rotator + woodpecker-allow-rotators
@@ -99,7 +99,7 @@ spec:
                - name: BAO_PATH
                  value: homelab/infrastructure/review-bot
                - name: GITEA_URL
-                  value: http://gitea.apps.svc.cluster.local:3000
+                  value: http://gitea.gitea.svc.cluster.local:3000
                # Private repo review-bot collaborates on: the pull check is a
                # real access verification (git-upload-pack on a PRIVATE repo
                # needs a token with access), not the "any valid token" weakness
@@ -225,15 +225,8 @@ metadata:
  labels:
    app.kubernetes.io/part-of: homelab-backup
 spec:
-  # Daily at 07:15 local (next free stagger slot after home-assistant-daily
-  # at 06:45). Covers every stateful workload in the shared `apps`
-  # namespace — currently Gitea (data-gitea-0, the Flux GitOps source of
-  # truth) and Woodpecker (data-woodpecker-server-0), plus gatus-data.
-  # Added 2026-07-14 (issue #186): the apps namespace had no Velero
-  # schedule at all, so the in-cluster Gitea/Woodpecker PVCs migrated from
-  # the legacy LXCs (see MIGRATION-gitea-woodpecker.md) were unprotected.
-  # No CNPG cluster runs in this namespace (both Gitea and Woodpecker use
-  # SQLite), so no labelSelector exclusion is needed.
+  # Daily at 07:15 local. Every PVC in the shared `apps` namespace (gatus-data).
+  # No CNPG cluster runs here, so no labelSelector exclusion is needed.
  schedule: "15 7 * * *"
  useOwnerReferencesInBackup: false
  template:
@@ -247,6 +240,26 @@ spec:
 ---
 apiVersion: velero.io/v1
 kind: Schedule
+metadata:
+  name: gitea-daily
+  namespace: velero
+  labels:
+    app.kubernetes.io/part-of: homelab-backup
+spec:
+  # Daily at 07:45 local. data-gitea-0 (SQLite, repos, LFS, packages).
+  schedule: "45 7 * * *"
+  useOwnerReferencesInBackup: false
+  template:
+    includedNamespaces:
+      - gitea
+    includedResources:
+      - persistentvolumeclaims
+    storageLocation: default
+    ttl: 720h0m0s
+    snapshotMoveData: true
+---
+apiVersion: velero.io/v1
+kind: Schedule
 metadata:
  name: trmnl-daily
  namespace: velero
@@ -96,7 +96,7 @@ spec:
        # that traffic is always a pod IP — collapsing all external clients
        # to one source. The NetBird reverse-proxy services are TLS
        # passthrough and carry the WAN client address in a PROXY protocol
-        # v2 header (options.proxy_protocol, set by netbird-proxy-services);
+        # v2 header (options.proxy_protocol, set by netbird-service-sync);
        # X-Forwarded-For stays trusted for HTTP-mode services. Both are
        # accepted from the pod CIDRs only; connections without a header
        # (direct LAN/mesh) keep their RemoteAddr.
@@ -58,6 +58,12 @@ machine:
      imageGCHighThresholdPercent: 70
      imageGCLowThresholdPercent: 60
 cluster:
+  # etcd peers on node0, VLAN 46 (#576). Pinned: Talos never updates an
+  # existing member's peer URLs, so changing this subnet means leave + rejoin
+  # per member (talosctl etcd leave, reboot).
+  etcd:
+    advertisedSubnets:
+      - 10.2.46.0/24
  # controlPlane.endpoint is NOT here: `talosctl gen config` takes it as its
  # second argument, and bootstrap.sh already passes https://k8s.renner.uno:6443
  # there. Setting it again in this patch restated the same value in a second
@@ -17,6 +17,8 @@ machine:
    # pod0 is a closed L2 fabric between the three nodes: statically
    # addressed, no gateway, nothing to learn from an RA.
    net.ipv6.conf.pod0.accept_ra: "0"
+    # node0 is closed the same way.
+    net.ipv6.conf.node0.accept_ra: "0"
    # ingress0 keeps RAs, at the default accept_ra of 2: an IPv6 default
    # router is a link-local address learned only from the RA, and with
    # net.ipv6.conf.all.forwarding=1 — which Cilium sets — Linux ignores RAs
@@ -122,8 +124,8 @@ kind: LinkConfig
 name: uplink
 up: true
 ---
-# Management, VLAN 40. Carries the node IP, etcd, the API and Ceph's public
-# network.
+# Management, VLAN 40: world↔node. The API VIP, admin access to the nodes, and
+# node egress (DNS, NTP, image pulls). Node↔node traffic is on node0 (#576).
 apiVersion: v1alpha1
 kind: VLANConfig
 name: mgmt0
@@ -177,6 +179,19 @@ addresses:
  - address: 10.2.48.11/24
  - address: fd02:48::11/64
 ---
+# Node network, VLAN 46: traffic among the three nodes themselves (#576) — the
+# node IP, etcd, apiserver↔kubelet. Closed like pod0: statically addressed, no
+# gateway, never routed via the UDM.
+apiVersion: v1alpha1
+kind: VLANConfig
+name: node0
+vlanID: 46
+parent: uplink
+up: true
+addresses:
+  - address: fd02:46::11/64
+  - address: 10.2.46.11/24
+---
 # Ingress network, VLAN 47. World-to-pod: the BGP session, the next hop for
 # every advertised LoadBalancer VIP, and the masquerade source for pod egress.
 # Publicly routed, which is what separates it from pod0.
@@ -16,6 +16,7 @@ machine:
    # ingress0 keeps RAs: it learns its IPv6 router and its address from them,
    # and only the default route is held back — see k8s-1.yaml.
    net.ipv6.conf.pod0.accept_ra: "0"
+    net.ipv6.conf.node0.accept_ra: "0"
    net.ipv6.conf.ingress0.accept_ra_defrtr: "0"
    # Answer ARP only on the link that owns the address — see k8s-1.yaml (#793).
    net.ipv4.conf.all.arp_ignore: "1"
@@ -82,7 +83,7 @@ kind: LinkConfig
 name: uplink
 up: true
 ---
-# Management, VLAN 40. Node IP, etcd, the API and Ceph's public network.
+# Management, VLAN 40: world↔node — see k8s-1.yaml.
 apiVersion: v1alpha1
 kind: VLANConfig
 name: mgmt0
@@ -122,6 +123,17 @@ addresses:
  - address: 10.2.48.12/24
  - address: fd02:48::12/64
 ---
+# Node network, VLAN 46 — see k8s-1.yaml.
+apiVersion: v1alpha1
+kind: VLANConfig
+name: node0
+vlanID: 46
+parent: uplink
+up: true
+addresses:
+  - address: fd02:46::12/64
+  - address: 10.2.46.12/24
+---
 # Ingress network, VLAN 47. World-to-pod, publicly routed, statically
 # addressed — see k8s-1.yaml for why it is not leased.
 apiVersion: v1alpha1
@@ -12,6 +12,7 @@ machine:
    # ingress0 keeps RAs: it learns its IPv6 router and its address from them,
    # and only the default route is held back — see k8s-1.yaml.
    net.ipv6.conf.pod0.accept_ra: "0"
+    net.ipv6.conf.node0.accept_ra: "0"
    net.ipv6.conf.ingress0.accept_ra_defrtr: "0"
    # Answer ARP only on the link that owns the address — see k8s-1.yaml (#793).
    net.ipv4.conf.all.arp_ignore: "1"
@@ -109,6 +110,17 @@ addresses:
  - address: 10.2.48.13/24
  - address: fd02:48::13/64
 ---
+# Node network, VLAN 46 — see k8s-1.yaml.
+apiVersion: v1alpha1
+kind: VLANConfig
+name: node0
+vlanID: 46
+parent: uplink
+up: true
+addresses:
+  - address: fd02:46::13/64
+  - address: 10.2.46.13/24
+---
 # Ingress network, VLAN 47. World-to-pod, publicly routed, statically
 # addressed — see k8s-1.yaml for why it is not leased.
 apiVersion: v1alpha1
@@ -7,19 +7,15 @@
 # part of the config that could not be true anywhere but here.
 machine:
  kubelet:
-    # Constrain kubelet's auto-picked node IP to the mgmt LAN. Without this,
-    # kubelet picks the first non-loopback IP, which could land on the v30iot
-    # NIC (ens19, 10.2.30.0/24). Control-plane traffic and hostNetwork pod IPs
-    # must stay on mgmt LAN.
-    # fd40::/64 is the mgmt LAN ULA, and it is listed so the Node object carries
-    # an IPv6 InternalIP at all. Without one, Cilium falls back to detecting the
-    # node's IPv6 from mgmt0 and picks the SLAAC address from the dynamic
-    # provider prefix, which it then probes until the agent restarts (#789).
-    # The GUA is not a control-plane address: BGP peers on fd40::/64 too.
+    # The node IP lives on node0, VLAN 46: the closed node↔node network (#576).
+    # Both families are listed so the Node object carries an IPv6 InternalIP at
+    # all; without one Cilium detects the node's IPv6 itself and picks a SLAAC
+    # address from the dynamic provider prefix (#789). Talos picks one address
+    # per family, IPv4 first, whatever the order here.
    nodeIP:
      validSubnets:
-        - 10.2.40.0/24
-        - fd40::/64
+        - fd02:46::/64
+        - 10.2.46.0/24
  # The UDM on the ingress segment. Published here rather than as a route
  # because bird owns table 47 and originates the default from it; this is the
  # one place the address is stated.