feat(knowledge): stage Phase 2.4 K8s deployment manifests (Sprint E B2)
NOT YET APPLIED — push to origin/main is gated on the DNS A record knowledge.iamworkin.lan -> 10.0.56.200 being live. Per memory feedback_pfsense_dns_required_for_acme, applying the Certificate without DNS in place puts cert-manager into ~2h HTTP-01 backoff and needs `kubectl -n knowledge delete order <name>` recovery. Manifests authored: - apps/knowledge/knowledge.yaml — Namespace, PVC (knowledge-vector-store Longhorn 20Gi RWO), Deployment (single replica, Recreate, image localhost/fc-knowledge-web:v202604272200 placeholder, runAsNonRoot 1654, readOnlyRootFilesystem, drop ALL caps, /healthz startupProbe + readinessProbe, tcpSocket livenessProbe), Service (ClusterIP port 80 -> 8080), Certificate (step-ca-acme ClusterIssuer, 90d duration), IngressRoute (knowledge.iamworkin.lan, websecure entrypoint). - apps/knowledge/kustomization.yaml — `kubectl kustomize` preview file (matches fc-distribution shape; ApplicationSet uses dir generator). - apps/knowledge/README.md — deployment order checklist with the DNS preflight, image build/import loop for all 3 RKE2 nodes, push procedure, smoke verification, initial-deploy-state notes (zero editions until *.db files are pushed to the PVC), resource sizing, probe + middleware notes. Companion artifacts (separate repos, separate commits): - FlowerCore.Knowledge@eb91eb4 — Dockerfile.deploy at repo root - FlowerCore.Notes@96cd443 — scripts/deploy-knowledge.sh Apply order (from apps/knowledge/README.md): 1. Add DNS A record knowledge.iamworkin.lan -> 10.0.56.200 via FlowerCore.DNS or pfSense web UI. 2. Run `bash scripts/deploy-knowledge.sh` from FlowerCore.Notes — this builds + imports the image to all 3 RKE2 nodes with FLOWERCORE_DEPLOY_SKIP_ROLLOUT=1 (since the Deployment doesn't exist yet on the cluster). 3. Bump the image tag in this manifest to match the freshly-imported tag, then `git push` from this repo to land on main. ArgoCD picks up within ~3 minutes and creates `infra-knowledge`. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
227
apps/knowledge/knowledge.yaml
Normal file
227
apps/knowledge/knowledge.yaml
Normal file
@@ -0,0 +1,227 @@
|
||||
# FlowerCore.Knowledge.Web — fleet vector indexing & RAG hub.
|
||||
#
|
||||
# Phase 2.4 of the Knowledge service plan. REST + MCP service that scans
|
||||
# *.db files under /data/vector-stores and exposes:
|
||||
# - REST: /api/v1/editions, /api/v1/corpus/search, /healthz
|
||||
# - MCP: list_editions, describe_edition, corpus_search
|
||||
# - Static OpenAPI/Scalar via UseFlowerCoreApi
|
||||
#
|
||||
# Architecture:
|
||||
# Plan: FlowerCore.Notes/docs/ai-agents/flowercore-knowledge-service-plan.md
|
||||
# Sprint: FlowerCore.Notes/docs/ai-station/sprint-e-xxl-plan.md (Track B)
|
||||
# Repo: D:\git\FlowerCore\FlowerCore.Knowledge\
|
||||
# Shared: FlowerCore.Common -> FlowerCore.Shared.Indexing (chunkers, vector
|
||||
# stores, edition profiles, ICorpusSearchService facade)
|
||||
#
|
||||
# Deployment order (see apps/knowledge/README.md and the bluejay-infra/README.md
|
||||
# top-level checklist):
|
||||
# 1. FlowerCore.DNS public A record knowledge.iamworkin.lan -> 10.0.56.200
|
||||
# MUST exist BEFORE the Certificate is created, or cert-manager HTTP-01
|
||||
# backs off ~2h. Memory: feedback_pfsense_dns_required_for_acme.
|
||||
# 2. Build + import the image to ALL RKE2 nodes (server + both agents) since
|
||||
# the Pod uses a Longhorn PVC and may schedule anywhere.
|
||||
# Memory: feedback_rke2_localhost_imagepullpolicy.
|
||||
# 3. Bump the image tag in this file, git push.
|
||||
# 4. ArgoCD ApplicationSet picks up within ~3 minutes and creates
|
||||
# infra-knowledge.
|
||||
#
|
||||
# Initial-deploy state:
|
||||
# The Longhorn PVC is empty on first deploy. Knowledge.Web's filesystem
|
||||
# catalog will report zero editions until vector-store *.db files are
|
||||
# pushed into /data/vector-stores. Initial population is a follow-up step
|
||||
# (Phase 2.5+, Blazor admin UI's "Rebuild" button); for the first deploy
|
||||
# the goal is just to prove the pod boots, /healthz returns 200, and the
|
||||
# Traefik IngressRoute serves the Scalar UI.
|
||||
---
|
||||
apiVersion: v1
|
||||
kind: Namespace
|
||||
metadata:
|
||||
name: knowledge
|
||||
labels:
|
||||
app.kubernetes.io/part-of: bluejay-infra
|
||||
---
|
||||
apiVersion: v1
|
||||
kind: PersistentVolumeClaim
|
||||
metadata:
|
||||
name: knowledge-vector-store
|
||||
namespace: knowledge
|
||||
spec:
|
||||
accessModes:
|
||||
- ReadWriteOnce
|
||||
storageClassName: longhorn
|
||||
resources:
|
||||
requests:
|
||||
storage: 20Gi
|
||||
---
|
||||
apiVersion: apps/v1
|
||||
kind: Deployment
|
||||
metadata:
|
||||
name: knowledge-web
|
||||
namespace: knowledge
|
||||
labels:
|
||||
app: knowledge-web
|
||||
app.kubernetes.io/name: knowledge-web
|
||||
app.kubernetes.io/part-of: bluejay-infra
|
||||
spec:
|
||||
replicas: 1
|
||||
revisionHistoryLimit: 3
|
||||
# RWO Longhorn PVC blocks rolling updates (multi-attach error). Recreate
|
||||
# is the canonical pattern (memory: feedback_rwo_pvc_blocks_rolling).
|
||||
strategy:
|
||||
type: Recreate
|
||||
selector:
|
||||
matchLabels:
|
||||
app: knowledge-web
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
app: knowledge-web
|
||||
app.kubernetes.io/name: knowledge-web
|
||||
app.kubernetes.io/part-of: bluejay-infra
|
||||
annotations:
|
||||
prometheus.io/scrape: "true"
|
||||
prometheus.io/port: "8080"
|
||||
prometheus.io/path: "/metrics"
|
||||
spec:
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
fsGroup: 1654
|
||||
fsGroupChangePolicy: OnRootMismatch
|
||||
containers:
|
||||
- name: web
|
||||
# Placeholder tag — bump to the image you built + imported to ALL
|
||||
# RKE2 nodes via scripts/deploy-knowledge.sh before applying.
|
||||
image: localhost/fc-knowledge-web:v202604272200
|
||||
imagePullPolicy: Never
|
||||
ports:
|
||||
- containerPort: 8080
|
||||
name: http
|
||||
env:
|
||||
- name: ASPNETCORE_URLS
|
||||
value: "http://+:8080"
|
||||
- name: ASPNETCORE_ENVIRONMENT
|
||||
value: "Production"
|
||||
- name: DOTNET_SYSTEM_GLOBALIZATION_INVARIANT
|
||||
value: "false"
|
||||
# Vector-store directory + embedding model + edition profile dir.
|
||||
# Profile JSON is baked into the image at /app/editions via the
|
||||
# csproj Content-link from FlowerCore.Common/editions/.
|
||||
- name: Knowledge__VectorStoresDirectory
|
||||
value: "/data/vector-stores"
|
||||
- name: Knowledge__EmbeddingModel
|
||||
value: "nomic-embed-text"
|
||||
- name: Knowledge__DefaultLimit
|
||||
value: "5"
|
||||
- name: Knowledge__MaxLimit
|
||||
value: "50"
|
||||
- name: FlowerCore__Editions__ProfileDirectory
|
||||
value: "/app/editions"
|
||||
# Embed via BLUEJAY-WS GPU (R9700, 32GB VRAM). Pi5 Ollama is
|
||||
# ~4-5x slower; use the workstation while we have it.
|
||||
# Memory: feedback_pi5_nomic_embed_slow.
|
||||
- name: FlowerCore__Ollama__BaseUrl
|
||||
value: "http://10.0.56.20:11434"
|
||||
resources:
|
||||
requests:
|
||||
cpu: 100m
|
||||
memory: 256Mi
|
||||
limits:
|
||||
cpu: 1000m
|
||||
memory: 1Gi
|
||||
# /healthz is mapped by HealthController (controller-based route).
|
||||
# tcpSocket liveness is the defensive fallback in case middleware
|
||||
# later gates /healthz behind auth (memory:
|
||||
# feedback_k8s_probes_behind_auth_middleware).
|
||||
startupProbe:
|
||||
httpGet:
|
||||
path: /healthz
|
||||
port: 8080
|
||||
initialDelaySeconds: 5
|
||||
periodSeconds: 5
|
||||
failureThreshold: 30
|
||||
readinessProbe:
|
||||
httpGet:
|
||||
path: /healthz
|
||||
port: 8080
|
||||
periodSeconds: 10
|
||||
failureThreshold: 3
|
||||
livenessProbe:
|
||||
tcpSocket:
|
||||
port: 8080
|
||||
initialDelaySeconds: 30
|
||||
periodSeconds: 30
|
||||
failureThreshold: 3
|
||||
securityContext:
|
||||
runAsNonRoot: true
|
||||
runAsUser: 1654
|
||||
runAsGroup: 1654
|
||||
allowPrivilegeEscalation: false
|
||||
readOnlyRootFilesystem: true
|
||||
capabilities:
|
||||
drop:
|
||||
- ALL
|
||||
volumeMounts:
|
||||
- name: vector-store
|
||||
mountPath: /data/vector-stores
|
||||
- name: tmp
|
||||
mountPath: /tmp
|
||||
- name: logs
|
||||
mountPath: /app/logs
|
||||
volumes:
|
||||
- name: vector-store
|
||||
persistentVolumeClaim:
|
||||
claimName: knowledge-vector-store
|
||||
- name: tmp
|
||||
emptyDir: {}
|
||||
- name: logs
|
||||
emptyDir: {}
|
||||
---
|
||||
apiVersion: v1
|
||||
kind: Service
|
||||
metadata:
|
||||
name: knowledge-web
|
||||
namespace: knowledge
|
||||
labels:
|
||||
app: knowledge-web
|
||||
app.kubernetes.io/name: knowledge-web
|
||||
app.kubernetes.io/part-of: bluejay-infra
|
||||
spec:
|
||||
type: ClusterIP
|
||||
selector:
|
||||
app: knowledge-web
|
||||
ports:
|
||||
- name: http
|
||||
port: 80
|
||||
targetPort: 8080
|
||||
---
|
||||
apiVersion: cert-manager.io/v1
|
||||
kind: Certificate
|
||||
metadata:
|
||||
name: knowledge-tls
|
||||
namespace: knowledge
|
||||
spec:
|
||||
secretName: knowledge-tls
|
||||
issuerRef:
|
||||
name: step-ca-acme
|
||||
kind: ClusterIssuer
|
||||
dnsNames:
|
||||
- knowledge.iamworkin.lan
|
||||
duration: 2160h # 90d
|
||||
renewBefore: 720h # 30d
|
||||
---
|
||||
apiVersion: traefik.io/v1alpha1
|
||||
kind: IngressRoute
|
||||
metadata:
|
||||
name: knowledge
|
||||
namespace: knowledge
|
||||
spec:
|
||||
entryPoints:
|
||||
- websecure
|
||||
routes:
|
||||
- match: Host(`knowledge.iamworkin.lan`)
|
||||
kind: Rule
|
||||
services:
|
||||
- name: knowledge-web
|
||||
port: 80
|
||||
tls:
|
||||
secretName: knowledge-tls
|
||||
Reference in New Issue
Block a user