Minimal prod delivery, authored directly on main (= prod's live revision d1a6628)
rather than by merging work/prod-standup-campaign. That branch carries the DNS apex
inversion, which would have set prod's BASE_DOMAIN to home.keithjackson.dev; prod
runs external-dns with policy: sync and txtOwnerId: prod, so it would have begun
rewriting apex records dev owns while the non-k3s prod still serves 443 there.
Operator sequencing is "dev migrates first". Nothing under clusters/ is touched
here at all, which is what makes this safe to land now.
Also deliberately EXCLUDED, because prod's live cluster-vars has no CNPG_BACKUP_PATH
key and StrictPostBuildSubstitutions would fail the whole tier: the R2 rework of
platform/velero/ and the CNPG destinationPath changes. Those land as
feat/r2-backup-prod together with the endpoint and bucket knobs.
WHAT THIS FIXES
1. Nothing was watching CNPG. All three prod databases have run since bootstrap with
ContinuousArchiving=False and firstRecoverabilityPoint=null, and no alert existed
anywhere. Because archive_mode=on and the archive command fails, Postgres refuses
to recycle WAL: keycloak-db holds every segment ever written (257 MiB on a 5 GiB
volume, ~6.8 MiB/h) — roughly 29 days to a disk-full PANIC on the cluster's SSO
database. Found by hand, 38h late. CNPGWALArchiveBacklog would have fired within
the hour, and evaluates to 3 firing series on prod today.
Rules key on cnpg_collector_* rather than cnpg_pg_stat_archiver_*, because
nextcloud-db's exporter cannot log in to its own database (permission denied for
database "nextcloud", both clusters) so the pg_stat_archiver family does not exist
for it — the obvious metric would have been blind to one database in three.
2. No PVC outside namespace=monitoring had any fill alert. New rules are partitioned
by namespace so they cannot double-fire with MonitoringVolumeFillingUp, pinned to
job="kubelet" (the apiserver copy of the same series reports namespace=default),
and guarded by inodes > 0 — block-mode PVCs such as unifi-rootdisk report
available=0 forever and would otherwise have fired critical permanently.
3. VPNTunnelDown could never clear on a cluster with no gateway node; it has been
firing critical on prod since 2026-09-05. Now guarded so prod self-arms the moment
a node carries the egress-gateway label, with dev's behaviour unchanged today.
4. The Keycloak realm import fails OPEN: an unresolved placeholder is stored
LITERALLY and the import still reports success, then sits at Done=True and never
re-runs. On 2026-09-06 all 12 prod OIDC clients held the literal
$(env:..._CLIENT_SECRET) string and every SSO login failed, while every signal was
green. The CronJob is the level-triggered correction; two alerts make its silence
observable. Repaired live already — this stops it recurring.
Also: the R-NET-04 conformance scan descended into nested agent worktrees under
.claude/, counting 4 egress policies where 1 is authored.
Co-Authored-By: Claude Opus 5 <[email protected]>
OPERATOR DECISION 2026-09-06: "use the VIP as intended — a virtual address to reach
the cluster and provide fault tolerant HA. Without a change it will not work in HA."
THE DEFECT. Cilium's kube-proxy replacement dialled a SINGLE NODE. Re-confirmed live
before changing anything: all three prod cilium-agent pods carried
KUBERNETES_SERVICE_HOST=192.168.2.10 — k3s-prod-1's own address — while the kube-vip
VIP 192.168.2.9 was Running on all three nodes with a held lease. Lose prod-1 and the
CNI's control-plane path is dead on an "HA" cluster. clusters/prod's own comment said
the value must be the VIP and NOT .10, then called it "Not yet decided" — I treated
that stale comment as an open question when the evidence already settled it.
SPLIT, NOT REPOINT, because k3s_api_host had two consumers with opposite needs:
k3s_api_host unchanged — the FOUNDER's IP: the cold-bootstrap join
target and a tls-san. infra/tofu/outputs.tf GENERATES this
name, so renaming would ripple into tofu and the runbooks.
k3s_apiserver_endpoint NEW = {{ kube_vip_address }} — the HA address; what Cilium
consumes. Derived, not restated, so an environment overrides
ONE line and the VIP, the SAN and the CNI endpoint move
together. Prod needed no new override.
THE SECOND HAZARD, WHICH MY BRIEF DID NOT NAME and the agent found: pointing Cilium at
the VIP breaks a COLD bootstrap for the same reason the join would — site.yml ran
node-baseline -> k3s -> cilium, so Cilium would install against an address nothing
announced yet. That is the 2026-09-01 deadlock shape. Fixed by SEQUENCE, not value:
site.yml now imports kube-vip.yml BETWEEN k3s and cilium. It cannot go earlier
(kube-vip is a k3s auto-deploy manifest, so it needs a cluster) and it works there
because the DaemonSet is hostNetwork with KUBERNETES_SERVICE_HOST=127.0.0.1 — needing
neither CNI nor kube-proxy. roles/kube_vip gained a read-only `wait_for` on the VIP so
a cold run stops THERE with a named cause instead of surfacing as unreachable agents.
Also fixed a pre-existing bug that only mattered once this play joined site.yml: the
role's read-only interface probe ran without `check_mode: false`, so under --check it
was skipped and the assert fired with a misleading "set kube_vip_interface explicitly".
VERIFIED BY ME, not just reported:
renders dev K3S_API_HOST=192.168.1.49 prod=192.168.2.9 (each environment's VIP)
cilium template consumes k3s_apiserver_endpoint, not k3s_api_host
cold join still `--server https://{{ k3s_api_host }}:6443` — the founder
order site.yml: node-baseline -> k3s -> kube-vip -> cilium -> verify -> flux
gate full CI gate PASS
Agent-side, additionally: in-memory render of cilium-values.yaml.j2 per host showing
k8sServiceHost=the VIP with the join still on the founder; k3s_tls_sans renders
['.10','.9','127.0.0.1','.11','.12'] so NO cert reissue and NO k3s restart is involved
— .9 already answers under full CA verification; and --check --diff of kube-vip.yml
against prod reported 6 ok / 0 changed, i.e. the role edits change nothing about the
kube-vip prod already runs.
ha-control-plane-as-code 92 -> 112 assertions, turning two escalations into machine
checks: the Cilium template must use the endpoint var; the endpoint must derive from
kube_vip_address and k3s_api_host must not; per environment K3S_API_HOST must equal
that environment's VIP and be none of its node IPs (the [parity] contract, now
machine-checked rather than a comment); site.yml must order k3s -> kube-vip -> cilium.
Every new assertion mutation-tested and bites. One honest negative recorded: reverting
only the BASE default is not caught, because both overlays patch the key — the check
reads effective per-environment values, so the base default is unguarded.
APPLY ORDER, and this is the operator's step — nothing was applied:
PROD FIRST, and it is the low-risk direction (prod is the broken one). Push and let
Flux reconcile platform/cilium: the value change rolls the cilium-agent DaemonSet
ONE NODE AT A TIME (maxUnavailable: 1 — the knob that bounded the 2026-09-01
incident), it will not proceed past an unready pod, existing datapath survives an
agent restart, and .9 is already live before the change lands.
DEV SEPARATELY, operator present. Same mechanics but live dev agents dial
192.168.1.50, and dev is mid-DNS-migration on dev-live. Do not fold it into that.
ANSIBLE: nothing to run for either live cluster; the Ansible half matters only for a
cold build. Remove the now-redundant standalone kube-vip.yml step from any runbook.
NEW NAMED GAP, unwritten: the L2 test cannot assert the running agents' live
KUBERNETES_SERVICE_HOST, which is the drift the L0 check cannot see. Recorded in that
test's gap list.
Refs R-PLAT-01, R-NET-01, R-ENV-02.
Co-Authored-By: Claude Opus 5 <[email protected]>
pia-credentials and pia-custom-ovpn were gluetun's PIA creds; gluetun is retired and the node
tunnel uses host-side Ansible vault creds (docs/07), so these cluster Secrets are unused. Remove
them from the kustomization + delete the encrypted files (recoverable from git history). Flux
will prune the Secrets from the vpn-egress namespace.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
The policy is wired into the kustomization, reconciled by Flux, and conformance-green; gluetun
is retired. Update the header banner + bottom checklist to reflect that. Comment-only — the
CiliumEgressGatewayPolicy spec is unchanged.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
The in-cluster Gluetun gateway is retired; VPN egress is now a node-level host
OpenVPN split-tunnel (Ansible role vpn_egress_gateway) steered by a Cilium
Egress Gateway policy with a host nftables kill-switch (ADR-0015). Credentials
moved host-side to the Ansible vault. Update docs/comments only:
- platform/vpn-egress/README.md: rewrite for ADR-0015 (live + conformance-green);
drop the resolved "option 1 not viable with Gluetun" finding to a brief history note.
- platform/kustomization.yaml: Phase 3 comment -> node tunnel + Cilium steering.
- platform/kyverno/.../require-egress-class-label.yaml: ADR-0013 -> ADR-0015 steering.
- schemas/servicedefinition.schema.json: egress=vpn description -> node tunnel + Cilium.
- clusters/base/cluster-vars.yaml: reframe PIA_REGION / CLUSTER_POD_CIDR WHY comments
to ADR-0015 (Cilium excludedCIDRs); both keys retained (still referenced/documented).
- platform/vpn-egress/namespace.yaml: no privileged gateway pod under ADR-0015.
- pia-credentials.sops.yaml / pia-custom-ovpn.sops.yaml: comment headers only note
they are now UNUSED (creds host-side), retained pending cleanup; payloads untouched.
No runtime manifest behavior, alert exprs, kustomization resources, or SOPS payloads
changed. No spec-repo edits.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Prepared on a feature branch for operator review — NOT merged/pushed (retiring the running
fallback needs explicit approval). The node-level host OpenVPN tunnel steered by Cilium Egress
Gateway is validated end-to-end and gated by green conformance (egress-class + vpn-killswitch).
A Gluetun pod's tun0 was never steerable by Cilium egress gateway (the reason ADR-0015 moved the
tunnel to the node), so gluetun serves no steering role. The now-unused pia-* SOPS secrets are
retained pending a separate cleanup (creds are host-side in the Ansible vault).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
The CiliumEgressGatewayPolicy authored in 38fcfad was inadvertently removed from main when
feature/vpn-egress-gateway-role (which carried a git-rm of the duplicated file) was merged.
The kustomization now references it, so restore the file. Content identical to 38fcfad.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Host OpenVPN split-tunnel on k3s-dev-3 validated 2026-08-24: tun0 up, VPN egress
158.173.241.36, node default unchanged (WAN), scoped nft kill-switch active. Wire the
CiliumEgressGatewayPolicy into the runtime kustomization so egress-class=vpn pods steer
to the node tun0.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Add platform/vpn-egress/egress-steering.yaml — a CiliumEgressGatewayPolicy
that steers homelab.keithjackson.dev/egress-class=vpn pods out a gateway
node's host tun0 split-tunnel interface, per ADR-0015 (node-level tunnel +
Cilium Egress Gateway). Honors all four ADR-0015 invariants: only vpn-class
pods are selected (default/internet + node stay on the local/WAN path),
RFC1918 + cluster ranges are excludedCIDRs so only public internet is
tunneled and tun0 is never a default route, and the host nftables kill-switch
(Ansible) fails vpn-class CLOSED while internet-class fails OPEN.
Interface contract shared verbatim with the parallel Ansible host-tunnel role:
node label homelab.keithjackson.dev/egress-gateway=true, interface tun0,
client label homelab.keithjackson.dev/egress-class=vpn.
Cilium Egress Gateway prerequisites confirmed enabled (egressGateway.enabled,
bpf.masquerade, kubeProxyReplacement all true, [parity]) — no HelmRelease change.
Intentionally NOT added to platform/vpn-egress/kustomization.yaml: the datapath
stays out of the runtime until the live tunnel exists and the egress-class /
vpn-killswitch conformance tests pass (README rule; DoD). Gluetun is retained
as the validated fallback until then. Serves R-NET-05/06/07.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Author the internet/vpn egress-class fixture pods for the egress-class + vpn-killswitch
conformance tests. Record the datapath finding: ADR-0013's preferred steering (Cilium
egress gateway -> gluetun tun0) is NOT viable because gluetun must run hostNetwork to
put tun0 on the node, and gluetun then redirects the NODE's default route through the
VPN (breaks k3s/etcd). Escalated: remaining viable path is option 2 (gateway-pod
routing). No steering datapath committed until the mechanism is chosen + validated.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Once the tunnel redirects the pod default route via tun0, the pod's response to the
kubelet's httpGet liveness probe egresses the tunnel and never returns → probe times
out → pod restarts ~every 2min despite the tunnel being up (PIA egress IP confirmed).
Switch to a local wget exec probe (stays on lo).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Operator provides a new .ovpn with an IP remote (gluetun's custom parser rejects
hostnames), so drop the resolve-remote initContainer/VPN_ENDPOINT_IP attempts and
mount the .ovpn directly (key pia.ovpn). Bump image to latest v3.41.3.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
gluetun's custom parser rejects the .ovpn's 'remote sweden.privacy.network' (needs
an IP). sweden.privacy.network LB pool = 158.173.241.{22,34,55,112,239}; pin .22
(verified to handshake) via VPN_ENDPOINT_IP/PORT while the .ovpn supplies cipher/CA.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Operator's kubectl create secret keyed the file by its name
(sweden-aes-128-cbc-udp-dns.ovpn); mount item now references that key -> pia.ovpn.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
PIA-provider server selection is flaky here (random dead IPs); the operator's .ovpn
uses sweden.privacy.network (load-balanced, reliable). Switch to custom mode
(VPN_SERVICE_PROVIDER=custom, OPENVPN_CUSTOM_CONFIG=/config/pia.ovpn), mount the
pia-custom-ovpn Secret, keep creds + kill-switch + tun-mtu 1300. Drops PIA-provider
server list/updater and native PF (Phase-6 concern). Pending the operator-provisioned
pia-custom-ovpn Secret (not merged until it exists).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
gluetun silently ignored PIA_ENCRYPTION_PRESET (renamed) and defaulted to strong
(RSA-4096) the whole time. The correct v3.x var is
PRIVATE_INTERNET_ACCESS_OPENVPN_ENCRYPTION_PRESET=normal → RSA-2048/AES-128-CBC/1198,
the smaller handshake that fits the 1450 pod MTU.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
RSA-4096 strong handshake wouldn't hold over the pod's 1450 MTU (Cilium VXLAN);
MTU tuning is exhausted (no --tls-mtu in OpenVPN 2.6, PIA rejects --fragment).
Operator downloaded the lower-encryption PIA config (port 1198, RSA-2048,
AES-128-CBC, SHA1) = gluetun's 'normal' preset. Smaller handshake fits 1450.
Dropped the strong-handshake band-aid flags; kept OPENVPN_MTU 1300 for the data channel.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
RSA-4096 strong-preset handshake is marginal over the 1450 pod MTU (Cilium VXLAN).
--mtu-disc no clears DF so oversized control packets fragment through instead of
being silently dropped. Last MTU-path lever (OpenVPN 2.6 has no --tls-mtu).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
OpenVPN 2.6.16 rejects --tls-mtu (not a real option) and PIA doesn't support
--fragment, so control-channel packets can't be shrunk. Keep --hand-window 120 to
give the RSA-4096 handshake more time over a lossy/marginal path, plus tun-mtu 1300.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
RSA-4096 strong-preset handshake is marginal over the pod's 1450 MTU (Cilium VXLAN).
Lower tun-mtu to 1300 and add --tls-mtu 1200 (smaller control packets) + --hand-window
120 (more time to complete the handshake over a lossy path). Empirical test of the
MTU-tuning path (operator's choice over Cilium native routing).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Operator's working config is sweden.privacy.network:1197 with RSA-4096, AES-256-CBC,
SHA256 = PIA's 'strong' preset. gluetun defaults PIA to 'normal' (AES-128-GCM), a
cipher mismatch against the 1197 (strong) endpoint, so the handshake never completed
(silent after UDP link). PIA_ENCRYPTION_PRESET=strong matches it and selects 1197.
Reverts the 8080 port red herring.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
UPDATER_PERIOD only refreshes AFTER the tunnel is ready (chicken-and-egg: can't
connect with stale servers), and gluetun's bundled PIA Sweden list is retired
46.246.x (current is 158.173.x). Add an `update-servers` initContainer running
`gluetun update -providers "private internet access"` into a shared /gluetun
emptyDir the main container reads, so it dials current servers. UPDATER_PERIOD
raised to 720h (docs-compliant) for ongoing in-pod refresh.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
CORRECTION: the connection failures were gluetun dialing RETIRED PIA IPs (46.246.x
= dead → timeouts, which I mis-read as a network block). A fresh PIA .ovpn connects
from the same host, and PIA's current SE Stockholm servers (158.173.x) are reachable.
Cilium was independently cleared (default-allow, host-fw off, Hubble FORWARDED).
Fix: UPDATER_PERIOD=24h so gluetun fetches PIA's current servers; keep
OPENVPN_MTU/MSSFIX=1350 for the real Cilium-VXLAN 1450 pod MTU (data channel);
PIA default UDP port. Re-enable vpn-egress. Full correction in DECISIONS-LOG.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Decisive diagnosis: OpenVPN's outbound init packets are 14 bytes with zero
replies (so MTU was never the blocker), pod UDP egress is healthy (NTP replies),
but TCP 501/443 to PIA IPs time out while Google/Cloudflare are OPEN — across many
PIA servers. This 192.168.2.x WiFi segment blackholes PIA's IP ranges (and
intercepts DNS). Not code-fixable. Reverted the port/MTU red-herring env to PIA
defaults; disabled vpn-egress pending an operator network fix (move to the
unfiltered 192.168.1.x LAN, or unblock PIA on the router). Full trace in
DECISIONS-LOG.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Actual root cause (operator-directed layered MTU check): Cilium runs VXLAN
(routing-mode tunnel) so pods get MTU 1450 (1500-50 encap) while node, hypervisor,
and the Google Fiber WAN are all 1500. OpenVPN's ~1500-byte handshake packets
exceed 1450 and are dropped (PMTUD broken). The ports were a red herring.
Fix: OPENVPN_MTU=1350 + OPENVPN_MSSFIX=1350 so encrypted packets (~1402) fit under
1450. Keeps OpenVPN + VXLAN unchanged. Re-enable vpn-egress. Full layered trace
(pod/node/host) in DECISIONS-LOG.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Port 53 also got no PIA replies: this network transparently intercepts outbound
:53, so dig works but OpenVPN-over-53 is swallowed. An NTP/123 probe from the host
returned fine, proving arbitrary non-53 UDP works bidirectionally (so it's
port-specific: 1197 conflict + 53 DNS-intercept, not a UDP block). Move to 8080
(community-verified for PIA, non-DNS, non-1197).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Wiring the port via "${PIA_OPENVPN_PORT}" made Flux/kustomize emit an unquoted
53 (int) after substitution, which k8s rejects for an env value ("expected
string, got 53"). Same gotcha as the Longhorn numberOfReplicas fix. Hardcode the
quoted string; cluster-var removed with a NOTE explaining why.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Operator identified the block: the home network NATs its own OpenVPN server on
WAN:1197, colliding with the cluster's outbound PIA-on-1197 flow (port-specific,
not a general UDP block — UDP/53 worked in tcpdump). PIA accepts OpenVPN on
53/1194/1197/1198/8080/9201; chose 53 (empirically proven to traverse this net).
- cluster-vars: add PIA_OPENVPN_PORT="53"
- gluetun: OPENVPN_ENDPOINT_PORT=${PIA_OPENVPN_PORT}
- re-enable vpn-egress; DECISIONS-LOG resolution noted
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Gluetun cannot do PIA over WireGuard (upstream unsupported; live "provider name
is not valid for Wireguard"). Operator approved OpenVPN-UDP and ADR-0013 was
amended in the spec. This aligns the impl:
- platform/kustomization.yaml: re-enable `- vpn-egress` (Phase 3, OpenVPN-UDP)
- gluetun-gateway.yaml: clean stale WireGuard comments (header, image note,
VPN_TYPE/creds comments); config already VPN_TYPE=openvpn
- README.md: PIA-over-OpenVPN, status=enabled, steering iface wg0 -> tun0
- DECISIONS-LOG.md: disk/drive incident (k3s -> NVMe; failing Seagate) + Phase-3
OpenVPN resolution
Stale WireGuard comments inside the operator's encrypted pia-credentials.sops.yaml
header left untouched (editing breaks the SOPS MAC; cosmetic).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
The deployed v3.40.0 rejected PIA+WireGuard ('private internet access is not valid for
Wireguard'). Verified v3.41.0 accepts the PIA+WireGuard provider/protocol (passes config
validation; ADR-0013's model holds — v3.40.0 was just too old). Bump the pin.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
gluetun-gateway.yaml referenced ${PIA_REGION}/${CLUSTER_POD_CIDR} but they were never defined
in cluster-vars, and the README didn't say where to set them. Add both to
clusters/base/cluster-vars.yaml (ConfigMap data:, string values, Flux postBuild substitution).
CLUSTER_POD_CIDR is 10.0.0.0/8 (Cilium cluster-pool — pods get 10.x, confirmed live), NOT the
k3s 10.42.0.0/16 I'd assumed. README now names the file + format + per-env override + PF-region note.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Per ADR-0013: platform/vpn-egress/ = a single shared Gluetun gateway (PIA over WireGuard,
built-in kill-switch FIREWALL=on, native port forwarding), namespace (non-app, NET_ADMIN +
/dev/net/tun), control-API Service, and a pre-wired (commented) pia-credentials SOPS line.
Defined the PIA secret location (platform/vpn-egress, keys username/password → OPENVPN_USER/
OPENVPN_PASSWORD) and updated the SOPS template accordingly.
NOT wired into platform/kustomization.yaml (Phase 3, author-first) — deploy needs PIA creds +
validation of the egress-steering mechanism (Cilium Egress Gateway → the Gluetun node's wg0;
documented in README with options + a fail-closed rationale, deliberately NOT committed as an
untested CiliumEgressGatewayPolicy). WireGuard chosen (UDP, protocol != TCP); PIA supports it on
all regions and the secret stays username/password.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
