jakman85andClaude Opus 4.8
7b8777c1e1
Seerr v3.4.1 (ghcr.io/seerr-team/seerr) runs as the non-root user node:node (UID/GID 1000), unlike the old root-running sctx/overseerr image. Its root-owned Longhorn RWO config PVC at /app/config was not writable by UID 1000, so the pod CrashLoopBackOff'd with: Error: EACCES: permission denied, mkdir '/app/config/logs/' Add a podSecurityContext with fsGroup=1000 (plus runAsUser/runAsGroup/runAsNonRoot) to the seerr ServiceDefinition — the same schema 1.3.0 mechanism semaphore uses. Longhorn's CSIDriver fsGroupPolicy: ReadWriteOnceWithFSType applies fsGroup to this RWO filesystem volume, group-owning the PVC to GID 1000 so the process can write. Re-rendered apps/wave-2/seerr.yaml from the catalog source. Co-Authored-By: Claude Opus 4.8 <[email protected]>
79 lines
5.0 KiB
YAML
79 lines
5.0 KiB
YAML
# ServiceDefinition: seerr
|
|
# Source of truth: spec docs/13-service-catalog.md row 5; docs/15-identity-and-access.md;
|
|
# docs/03-networking-isolation-and-egress.md (explicit allow rules).
|
|
# public | egress internet | config on Longhorn | auth: native (Plex accounts) |
|
|
# seerr -> sonarr:8989, seerr -> radarr:7878, seerr -> plex:32400.
|
|
#
|
|
# Seerr is the media request UI (Overseerr's maintained successor): users request titles, it
|
|
# hands them to the arr stack and reflects library state from Plex. It is publicly exposed (edge port-forward + public
|
|
# DNS, docs/09) and authenticates with Plex's own account system — a DOCUMENTED SSO exception
|
|
# (docs/13/15; R-IDP-02): NOT Keycloak, so no oidc/forward-auth and no auth.groups. The three
|
|
# east-west hops below are EXPLICIT allows on top of default-deny (docs/03; R-NET-01/02);
|
|
# everything else (except DNS) is denied. Validates against schemas/servicedefinition.schema.json.
|
|
#
|
|
# Image: Seerr (ghcr.io/seerr-team/seerr) — the maintained successor into which the original
|
|
# Overseerr (sctx/overseerr, now unmaintained) and Jellyseerr (fallenbagel/jellyseerr) merged;
|
|
# the Plex-native request UI on port 5055 (docs/13 row 5 names "seerr" generically). Pinned to a
|
|
# specific stable tag (v3.4.1) per the arr-stack convention (sonarr/radarr/qbittorrent/gitea all
|
|
# pin explicit tags; R-UPD-01); update.range is the advisory bound the ImagePolicy will enforce
|
|
# once the Flux image controllers are installed. Config lives under /app/config (upstream default).
|
|
#
|
|
# No `bulk` volume: seerr keeps only its own config PVC (settings + request/cache DB) on the
|
|
# longhorn tier. It stores no media and needs no bulk mount, so OQ-2 (unresolved bulk backing)
|
|
# does not apply here.
|
|
name: seerr
|
|
kind: container
|
|
source:
|
|
image: ghcr.io/seerr-team/seerr:v3.4.1 # Seerr: maintained Overseerr/Jellyseerr successor, Plex-native request UI (docs/13 row 5); served on 5055
|
|
# The ghcr.io/seerr-team/seerr image runs as the non-root user `node:node` (UID 1000 / GID 1000 —
|
|
# the standard node base-image user; confirmed from the image config's User field). On boot the app
|
|
# creates /app/config/logs and writes its settings/DB under the mounted /app/config PVC. A fresh
|
|
# Longhorn PVC is root:root, so UID 1000 cannot mkdir there -> the pod CrashLoopBackOffs with
|
|
# "EACCES: permission denied, mkdir '/app/config/logs/'". (The old sctx/overseerr image ran as root
|
|
# and never hit this.) fsGroup=1000 makes the kubelet group-own the mounted PVC to GID 1000 and add
|
|
# it as a supplemental group so the process can write — Longhorn's CSIDriver
|
|
# fsGroupPolicy: ReadWriteOnceWithFSType applies fsGroup to this RWO filesystem volume. runAsUser/
|
|
# runAsGroup pin the image's own identity, runAsNonRoot asserts it (same pattern as semaphore; R-STATE-01/02).
|
|
podSecurityContext:
|
|
runAsUser: 1000
|
|
runAsGroup: 1000
|
|
runAsNonRoot: true
|
|
fsGroup: 1000
|
|
exposure: public # edge port-forward + public DNS (docs/09; R-DNS/R-NET)
|
|
egress: internet # normal WAN egress (docs/03; R-NET-05) — TMDB/Plex.tv metadata over the internet
|
|
storage:
|
|
- name: config
|
|
size: 5Gi
|
|
class: longhorn # settings + request/cache DB: replicated app state on the 3-replica tier (docs/04; R-STATE-01/02)
|
|
mountPath: /app/config # Seerr/Overseerr/Jellyseerr keep config + DB under /app/config (docs.seerr.dev); /data/<name> default would not persist it
|
|
dependencies:
|
|
- to: sonarr # seerr -> sonarr: submit TV requests (docs/03 allow rules; docs/13 graph; R-NET-02)
|
|
port: 8989
|
|
protocol: tcp
|
|
- to: radarr # seerr -> radarr: submit movie requests (docs/03 allow rules; docs/13 graph; R-NET-02)
|
|
port: 7878
|
|
protocol: tcp
|
|
- to: plex # seerr -> plex: read library / watch state (docs/03 allow rules; docs/13 graph; R-NET-02)
|
|
port: 32400
|
|
protocol: tcp
|
|
ingressRoutes:
|
|
- host: seerr.home.keithjackson.dev
|
|
port: 5055 # Overseerr web UI + API; wildcard TLS terminates at the shared Gateway (docs/08/09)
|
|
auth:
|
|
mode: native # Plex account auth — documented exception, NOT Keycloak (docs/13/15; R-IDP-02)
|
|
update:
|
|
channel: stable # R-UPD-01 (never pre-release)
|
|
range: ">=3.4.0" # acceptable versions (advisory; image-automation off during onboarding) — Seerr 3.x line
|
|
cadence: "0 4 * * 0" # weekly stable check/apply
|
|
autoRollbackOnFailedHealth: true # R-UPD-03
|
|
observability:
|
|
metrics: true
|
|
slis:
|
|
- up # R-OBS-01: service-down alert
|
|
backup:
|
|
# seerr's config is on the longhorn (replicated) tier, so the L0 Kyverno backup policy
|
|
# (servicedefinition-backup; R-STATE-03) forbids `reacquirable` here — it must reference an
|
|
# off-cluster schedule. The request/settings DB is cheap to rebuild but still lives on the
|
|
# irreplaceable tier, so it is backed up off-cluster like the other arr-stack config volumes.
|
|
policy: daily-offcluster # off-cluster backup of the longhorn config volume (docs/04; R-STATE-03)
|