J'ai donc suivi le guide (https://dev.mysql.com/doc/mysql-operator/en/mysql-operator-installation-kubectl.html) pour installer le cluster MySql sur le cluster Kubernetes Ubuntu Server 22.04.
J'ai exécuté avec succès la commande suivante pour importer le fichier de déploiement yaml requis
kubectl apply -f https://raw.githubusercontent.com/mysql/mysql-operator/trunk/deploy/deploy-crds.yaml kubectl apply -f https://raw.githubusercontent.com/mysql/mysql-operator/trunk/deploy/deploy-operator.yaml
Maintenant, lorsque j'exécute la commande de déploiement, le pod continue de redémarrer.
kubectl get deployment mysql-operator --namespace mysql-operator
Voici le résultat du journal (j'ai eu l'erreur)
kubectl logs mysql-operator-755b988dd9-jn7f6 --namespace=mysql-operator exec /usr/bin/mysqlsh: exec format error
Voici la description :
kubectl describe pod mysql-operator-755b988dd9-jn7f6 --namespace=mysql-operator Name: mysql-operator-755b988dd9-jn7f6 Namespace: mysql-operator Priority: 0 Service Account: mysql-operator-sa Node: mnetk8s-node-a1/10.30.0.11 Start Time: Wed, 07 Jun 2023 22:52:40 +0100 Labels: name=mysql-operator pod-template-hash=755b988dd9 Annotations: <none> Status: Running IP: 10.244.3.6 IPs: IP: 10.244.3.6 Controlled By: ReplicaSet/mysql-operator-755b988dd9 Containers: mysql-operator: Container ID: docker://785da7db660ae71ad09e1d6ecf7672f68fbbfbf1e03b64b4619bde188c7a250f Image: container-registry.oracle.com/mysql/community-operator:8.0.33-2.0.10 Image ID: docker-pullable://container-registry.oracle.com/mysql/community-operator@sha256:9c411c679b416b7d4bce4e14f65b311105429d25c3183f864fdd9baec8d5647f Port: <none> Host Port: <none> Args: mysqlsh --log-level=@INFO --pym mysqloperator operator State: Waiting Reason: CrashLoopBackOff Last State: Terminated Reason: Error Exit Code: 1 Started: Wed, 07 Jun 2023 23:34:29 +0100 Finished: Wed, 07 Jun 2023 23:34:29 +0100 Ready: False Restart Count: 13 Readiness: exec [cat /tmp/mysql-operator-ready] delay=1s timeout=1s period=3s #success=1 #failure=3 Environment: MYSQLSH_USER_CONFIG_HOME: /mysqlsh MYSQLSH_CREDENTIAL_STORE_SAVE_PASSWORDS: never Mounts: /mysqlsh from mysqlsh-home (rw) /tmp from tmpdir (rw) /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-jdn2d (ro) Conditions: Type Status Initialized True Ready False ContainersReady False PodScheduled True Volumes: mysqlsh-home: Type: EmptyDir (a temporary directory that shares a pod's lifetime) Medium: SizeLimit: <unset> tmpdir: Type: EmptyDir (a temporary directory that shares a pod's lifetime) Medium: SizeLimit: <unset> kube-api-access-jdn2d: Type: Projected (a volume that contains injected data from multiple sources) TokenExpirationSeconds: 3607 ConfigMapName: kube-root-ca.crt ConfigMapOptional: <nil> DownwardAPI: true QoS Class: BestEffort Node-Selectors: <none> Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s node.kubernetes.io/unreachable:NoExecute op=Exists for 300s Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 46m default-scheduler Successfully assigned mysql-operator/mysql-operator-755b988dd9-jn7f6 to mnetk8s-node-a1 Normal Pulled 45m (x5 over 46m) kubelet Container image "container-registry.oracle.com/mysql/community-operator:8.0.33-2.0.10" already present on machine Normal Created 45m (x5 over 46m) kubelet Created container mysql-operator Normal Started 45m (x5 over 46m) kubelet Started container mysql-operator Warning BackOff 97s (x221 over 46m) kubelet Back-off restarting failed container mysql-operator in pod mysql-operator-755b988dd9-jn7f6_mysql-operator(9e7f8c17-4fdd-4f98-a074-d2112019b28e)
Toute aide ou idée serait géniale !
Vous pouvez y jeter un œil.
Un outil open source appelé kubeblocks pour déployer et gérer MySQL sur Kubernetes.
Je pense que cela peut résoudre votre problème.
exec format error
Signifie généralement que l'image que vous utilisez n'est pas adaptée à l'architecture sur laquelle vous déployez. Autrement dit, vous utilisez une image ARM sur un nœud AMD64 et vice versa.Étant donné que toutes les images MySQL sur Oracle Container Registry sont des images AMD64, je suppose que votre cluster fonctionne sur une architecture ARM, telle que les instances Ampere.
Vous devrez exécuter sur une instance arch x64, ou trouver une image pour ARM, ou accéder à la source et reconstruire l'image pour ARM.