Référence API ClickHouse

Cette référence détaille l’utilisation de ClickHouse sur Hikube, que ce soit en configuration simple ou distribuée avec shards et réplicas.

---

Structure de Base

Ressource Clickhouse

Exemple de configuration YAML

apiVersion: apps.cozystack.io/v1alpha1
kind: ClickHouse
metadata:
  name: clickhouse-name
  namespace: default
spec:

Paramètres

Paramètres Communs

Paramètre	Type	Description	Défaut	Requis
replicas	int	Number of ClickHouse replicas	2	Oui
shards	int	Number of ClickHouse shards	1	Oui
resources	object	Explicit CPU and memory configuration for each replica. Si vide, resourcesPreset est appliqué	{}	Non
resources.cpu	quantity	CPU available to each replica	null	Non
resources.memory	quantity	Memory (RAM) available to each replica	null	Non
resourcesPreset	string	Default sizing preset (nano, micro, small, medium, large, xlarge, 2xlarge)	"small"	Oui
size	quantity	Persistent Volume Claim size, available for application data	10Gi	Oui
storageClass	string	StorageClass used to store the data	""	Non

Exemple de configuration YAML

clickhouse.yaml

replicas: 2
shards: 1
resources:
  cpu: 4000m
  memory: 4Gi
resourcesPreset: small
size: 20Gi
storageClass: replicated

Paramètres d'application spécifique

Paramètre	Type	Description	Défaut	Requis
logStorageSize	quantity	Size of Persistent Volume for logs	2Gi	Non
logTTL	int	TTL (expiration time) for query_log and query_thread_log	15	Non
users	map[string]object	Users configuration	{...}	Non
users[name].password	string	Password for the user	null	Oui
users[name].readonly	bool	User is readonly, default is false	null	Non

Exemple de configuration YAML

clickhouse.yaml

logStorageSize: 5Gi
logTTL: 30
users:
  analyst:
    password: analyst123
    readonly: true
  admin:
    password: adminStrongPwd

Paramètres de backup

Paramètre	Type	Description	Défaut	Requis
backup	object	Backup configuration	{}	Non
backup.enabled	bool	Enable regular backups	false	Non
backup.s3Region	string	AWS S3 region where backups are stored	us-east-1	Non
backup.s3Bucket	string	S3 bucket used for storing backups	s3.example.org/clickhouse-backups	Non
backup.schedule	string	Cron schedule for automated backups	"0 2 * * *"	Non
backup.cleanupStrategy	string	Retention strategy for cleaning up old backups	"--keep-last=3 --keep-daily=3 --keep-within-weekly=1m"	Non
backup.s3AccessKey	string	Access key for S3, used for authentication	<your-access-key>	Oui
backup.s3SecretKey	string	Secret key for S3, used for authentication	<your-secret-key>	Oui
backup.resticPassword	string	Password for Restic backup encryption	<password>	Oui

Exemple de configuration YAML

clickhouse.yaml

backup:
  enabled: true
  s3Region: eu-central-1
  s3Bucket: backups.hikube.clickhouse
  schedule: "0 3 * * *"
  cleanupStrategy: "--keep-last=5 --keep-daily=7 --keep-weekly=4"
  s3AccessKey: "<your-access-key>"
  s3SecretKey: "<your-secret-key>"
  resticPassword: "<password>"

Paramètres de Clickhouse keeper

Paramètre	Type	Description	Défaut	Requis
clickhouseKeeper	object	ClickHouse Keeper configuration	{}	Non
clickhouseKeeper.enabled	bool	Deploy ClickHouse Keeper for cluster coordination	true	Oui
clickhouseKeeper.size	quantity	Persistent Volume Claim size, available for application data	1Gi	Oui
clickhouseKeeper.resourcesPreset	string	Default sizing preset (nano, micro, small, medium, large, xlarge, 2xlarge)	micro	Oui
clickhouseKeeper.replicas	int	Number of Keeper replicas	3	Oui

Exemple de configuration YAML

clickhouse.yaml

clickhouseKeeper:
  enabled: true
  replicas: 3
  resourcesPreset: medium
  size: 5Gi  

resources et resourcesPreset

Le champ resources permet de définir explicitement la configuration CPU et mémoire de chaque réplique ClickHouse.
Si ce champ est laissé vide, la valeur du paramètre resourcesPreset est utilisée.

Exemple de configuration YAML

clickhouse.yaml

resources:
  cpu: 4000m
  memory: 4Gi

⚠️ Attention : si resources est défini, la valeur de resourcesPreset est ignorée.

Preset name	CPU	Mémoire
nano	250m	128Mi
micro	500m	256Mi
small	1	512Mi
medium	1	1Gi
large	2	2Gi
xlarge	4	4Gi
2xlarge	8	8Gi