sanitycheck

Documentation

Index

• func FixClusterSplit(ctx context.Context, admin redis.AdminInterface, config *config.Redis, ...) (bool, error)
• func FixFailedNodes(ctx context.Context, admin redis.AdminInterface, cluster *rapi.RedisCluster, ...) (bool, error)
• func FixTerminatingPods(cluster *rapi.RedisCluster, podControl pod.RedisClusterControlInterface, ...) (bool, error)
• func FixUntrustedNodes(ctx context.Context, admin redis.AdminInterface, ...) (bool, error)
• func RunSanityChecks(ctx context.Context, admin redis.AdminInterface, config *config.Redis, ...) (actionDone bool, err error)

Functions

func FixClusterSplit

func FixClusterSplit(ctx context.Context, admin redis.AdminInterface, config *config.Redis, infos *redis.ClusterInfos, dryRun bool) (bool, error)

FixClusterSplit use to detect and fix a cluster split

func FixFailedNodes

func FixFailedNodes(ctx context.Context, admin redis.AdminInterface, cluster *rapi.RedisCluster, infos *redis.ClusterInfos, dryRun bool) (bool, error)

FixFailedNodes fix failed nodes: in some cases (cluster without enough primary after crash or scale down), some nodes may still know about fail nodes

func FixTerminatingPods

func FixTerminatingPods(cluster *rapi.RedisCluster, podControl pod.RedisClusterControlInterface, maxDuration time.Duration, dryRun bool) (bool, error)

FixTerminatingPods used for the deletion of pods blocked in terminating status.

func FixUntrustedNodes

func FixUntrustedNodes(ctx context.Context, admin redis.AdminInterface, podControl pod.RedisClusterControlInterface, cluster *rapi.RedisCluster, infos *redis.ClusterInfos, dryRun bool) (bool, error)

FixUntrustedNodes used to remove nodes that are not trusted by other nodes. It is useful when a node is removed from the cluster (with the FORGET command) but tries to rejoin the cluster.

func RunSanityChecks

func RunSanityChecks(ctx context.Context, admin redis.AdminInterface, config *config.Redis, podControl pod.RedisClusterControlInterface, cluster *rapi.RedisCluster, infos *redis.ClusterInfos, dryRun bool) (actionDone bool, err error)

RunSanityChecks function used to run all the sanity check on the current cluster Return actionDone = true if a modification has been made on the cluster