Documentation
¶
Index ¶
- type Cluster
- func (c *Cluster) AddMetastoreWithAutoJoin(ctx context.Context) error
- func (c *Cluster) CompactionJobsFinished(ctx context.Context) (float64, error)
- func (c *Cluster) GetMetastoreRaftNodeClient() (raftnodepb.RaftNodeServiceClient, error)
- func (c *Cluster) Prepare(ctx context.Context) (err error)
- func (c *Cluster) PushClient() pushv1connect.PusherServiceClient
- func (c *Cluster) QueryClient() querierv1connect.QuerierServiceClient
- func (c *Cluster) Start(ctx context.Context) (err error)
- func (c *Cluster) Stop() func(context.Context) error
- func (c *Cluster) Wait()
- type ClusterOption
- type Component
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type Cluster ¶
type Cluster struct {
Components []*Component
// contains filtered or unexported fields
}
func NewMicroServiceCluster ¶
func NewMicroServiceCluster(opts ...ClusterOption) *Cluster
func (*Cluster) AddMetastoreWithAutoJoin ¶
func (*Cluster) CompactionJobsFinished ¶
func (*Cluster) GetMetastoreRaftNodeClient ¶
func (c *Cluster) GetMetastoreRaftNodeClient() (raftnodepb.RaftNodeServiceClient, error)
func (*Cluster) PushClient ¶
func (c *Cluster) PushClient() pushv1connect.PusherServiceClient
func (*Cluster) QueryClient ¶
func (c *Cluster) QueryClient() querierv1connect.QuerierServiceClient
type ClusterOption ¶
type ClusterOption func(c *Cluster)
func WithSymbolizer ¶
func WithSymbolizer(debuginfodURL string) ClusterOption
func WithV1 ¶
func WithV1() ClusterOption
func WithV2 ¶
func WithV2() ClusterOption
func WithV2Federated ¶ added in v2.1.0
func WithV2Federated() ClusterOption
WithV2Federated configures a V2 cluster suitable for exercising the federated (multi-tenant) query path. It mirrors the V2 cluster but without a compaction-worker: L0 segment blocks therefore remain Format1 (per-tenant dataset_index pseudo-datasets) for the duration of the test, which is what the multi-Format1 resolution path in the query backend handles.
With two segment-writers and adaptive placement defaults, profiles from each tenant are spread across both writers, so each flushed L0 segment is a multi-tenant block and naturally exercises the multi-Format1 resolution path.
Click to show internal directories.
Click to hide internal directories.