unkin-agent
c83a886e74
The cluster preloads no statement-statistics library, so there is no per-query cost attribution in postgres and slow query paths have to be inferred from application-side metrics instead of read straight out of the database. - preload `pg_stat_statements` - set `pg_stat_statements.max` and `.track`, which is what makes CNPG manage the extension and create it in every database Requires a postgres restart. Stacked on `benvin/authentik-cnpg-resources`. Reviewed-on: #475 Co-authored-by: unkin-agent <unkin-agent@unkin.net> Co-committed-by: unkin-agent <unkin-agent@unkin.net>
138 lines
4.2 KiB
YAML
138 lines
4.2 KiB
YAML
---
|
|
apiVersion: postgresql.cnpg.io/v1
|
|
kind: Cluster
|
|
metadata:
|
|
name: postgres
|
|
namespace: authentik
|
|
spec:
|
|
affinity:
|
|
podAntiAffinityType: preferred
|
|
backup:
|
|
# 30-day retention (DEFAULT — adjust per cluster if needed). Enforced by CNPG
|
|
# against the object store on each successful base backup.
|
|
retentionPolicy: 30d
|
|
barmanObjectStore:
|
|
# Dedicated per-cluster Ceph RGW bucket (cephrgw-operator provisions it).
|
|
destinationPath: s3://cnpg-authentik
|
|
endpointURL: https://s3.ceph.unkin.net
|
|
# radosgw serves a Vault-PKI cert; trust the internal CA (reflected into
|
|
# every namespace as the vault-ca-cert Secret).
|
|
endpointCA:
|
|
name: vault-ca-cert
|
|
key: ca.crt
|
|
# Keys minted by the ObjectStoreUser in cnpg_backup.yaml; never hardcoded.
|
|
s3Credentials:
|
|
accessKeyId:
|
|
name: cnpg-authentik-backup-s3
|
|
key: AWS_ACCESS_KEY_ID
|
|
secretAccessKey:
|
|
name: cnpg-authentik-backup-s3
|
|
key: AWS_SECRET_ACCESS_KEY
|
|
# Path prefix within the bucket; keep stable across restores (see docs).
|
|
serverName: authentik
|
|
data:
|
|
compression: bzip2
|
|
jobs: 2
|
|
wal:
|
|
compression: zstd
|
|
maxParallel: 2
|
|
bootstrap:
|
|
initdb:
|
|
database: authentik
|
|
encoding: UTF8
|
|
localeCType: C
|
|
localeCollate: C
|
|
owner: authentik
|
|
secret:
|
|
name: postgres-credentials
|
|
enablePDB: true
|
|
enableSuperuserAccess: false
|
|
failoverDelay: 0
|
|
imageName: ghcr.io/cloudnative-pg/postgresql:18.1-system-trixie
|
|
instances: 3
|
|
logLevel: info
|
|
maxSyncReplicas: 0
|
|
minSyncReplicas: 0
|
|
monitoring:
|
|
customQueriesConfigMap:
|
|
- key: queries
|
|
name: cnpg-default-monitoring
|
|
disableDefaultQueries: false
|
|
enablePodMonitor: false
|
|
postgresql:
|
|
parameters:
|
|
archive_mode: "on"
|
|
archive_timeout: 5min
|
|
dynamic_shared_memory_type: posix
|
|
effective_cache_size: 1536MB
|
|
full_page_writes: "on"
|
|
# Replicas report their oldest xmin to the primary, so multi-second reads on
|
|
# a hot standby stop exhausting max_standby_streaming_delay and being
|
|
# cancelled. Retained-dead-tuple cost is negligible on a ~155MB database.
|
|
hot_standby_feedback: "on"
|
|
log_destination: csvlog
|
|
log_directory: /controller/log
|
|
log_filename: postgres
|
|
log_rotation_age: "0"
|
|
log_rotation_size: "0"
|
|
log_truncate_on_rotation: "false"
|
|
logging_collector: "on"
|
|
max_connections: "200"
|
|
max_parallel_workers: "16"
|
|
max_replication_slots: "16"
|
|
max_worker_processes: "16"
|
|
# A pg_stat_statements.* parameter is what makes CNPG treat the extension as
|
|
# managed and run CREATE EXTENSION in every database; preloading alone does
|
|
# not create it.
|
|
pg_stat_statements.max: "10000"
|
|
pg_stat_statements.track: top
|
|
shared_buffers: 512MB
|
|
shared_memory_type: mmap
|
|
ssl_max_protocol_version: TLSv1.3
|
|
ssl_min_protocol_version: TLSv1.3
|
|
wal_keep_size: 256MB
|
|
wal_level: logical
|
|
wal_log_hints: "on"
|
|
wal_receiver_timeout: 5s
|
|
wal_sender_timeout: 5s
|
|
# CNPG merges this with the libraries it manages itself.
|
|
shared_preload_libraries:
|
|
- pg_stat_statements
|
|
syncReplicaElectionConstraint:
|
|
enabled: false
|
|
primaryUpdateMethod: restart
|
|
primaryUpdateStrategy: unsupervised
|
|
probes:
|
|
liveness:
|
|
isolationCheck:
|
|
connectionTimeout: 1000
|
|
enabled: true
|
|
requestTimeout: 1000
|
|
replicationSlots:
|
|
highAvailability:
|
|
enabled: true
|
|
slotPrefix: _cnpg_
|
|
synchronizeReplicas:
|
|
enabled: true
|
|
updateInterval: 30
|
|
resources:
|
|
limits:
|
|
# 500m is a 50ms CFS quota per 100ms period, exhausted by bursts even at
|
|
# ~0.01 cores average, so every query pays throttle latency.
|
|
cpu: "2"
|
|
# 512Mi OOMKilled replicas under load (shared_buffers 128MB +
|
|
# max_connections 200 leave no headroom) — see incident 2026-07-28.
|
|
# shared_buffers 512MB needs the same headroom multiple, hence 2Gi.
|
|
memory: 2Gi
|
|
requests:
|
|
cpu: 500m
|
|
memory: 1Gi
|
|
smartShutdownTimeout: 180
|
|
startDelay: 3600
|
|
stopDelay: 1800
|
|
storage:
|
|
resizeInUseVolumes: true
|
|
size: 20Gi
|
|
storageClass: cephrbd-fast-delete
|
|
switchoverDelay: 3600
|