Documentation ¶
Overview ¶
Package buffer provides a buffer for PRIMARY traffic during failovers.
Instead of returning an error to the application (when the vttablet primary becomes unavailable), the buffer will automatically retry buffered requests after the end of the failover was detected.
Buffering (stalling) requests will increase the number of requests in flight within vtgate and at upstream layers. Therefore, it is important to limit the size of the buffer and the buffering duration (window) per request. See the file flags.go for the available configuration and its defaults.
Index ¶
Constants ¶
This section is empty.
Variables ¶
var (
ShardMissingError = vterrors.New(vtrpcpb.Code_UNAVAILABLE, "destination shard is missing after a resharding operation")
)
Functions ¶
func CausedByFailover ¶ added in v0.11.0
CausedByFailover returns true if "err" was supposedly caused by a failover. To simplify things, we've merged the detection for different MySQL flavors in one function. Supported flavors: MariaDB, MySQL
func SetBufferingModeInTestingEnv ¶ added in v0.13.0
func SetBufferingModeInTestingEnv(enabled bool)
SetBufferingModeInTestingEnv should only be used from testing code to change the flag (enable_buffer) default value
Types ¶
type Buffer ¶
type Buffer struct {
// contains filtered or unexported fields
}
Buffer is used to track ongoing PRIMARY tablet failovers and buffer requests while the PRIMARY tablet is unavailable. Once the new PRIMARY starts accepting requests, buffering stops and requests queued so far will be automatically retried.
There should be exactly one instance of this buffer. For each failover, an instance of "ShardBuffer" will be created.
func (*Buffer) HandleKeyspaceEvent ¶ added in v0.12.0
func (b *Buffer) HandleKeyspaceEvent(ksevent *discovery.KeyspaceEvent)
func (*Buffer) ProcessPrimaryHealth ¶ added in v0.12.0
func (b *Buffer) ProcessPrimaryHealth(th *discovery.TabletHealth)
ProcessPrimaryHealth notifies the buffer to record a new primary and end any failover buffering that may be in progress
func (*Buffer) Shutdown ¶
func (b *Buffer) Shutdown()
Shutdown blocks until all pending ShardBuffer objects are shut down. In particular, it guarantees that all launched Go routines are stopped after it returns.
func (*Buffer) WaitForFailoverEnd ¶
func (b *Buffer) WaitForFailoverEnd(ctx context.Context, keyspace, shard string, err error) (RetryDoneFunc, error)
WaitForFailoverEnd blocks until a pending buffering due to a failover for keyspace/shard is over. If there is no ongoing failover, "err" is checked. If it's caused by a failover, buffering may be started. It returns an error if buffering failed (e.g. buffer full). If it does not return an error, it may return a RetryDoneFunc which must be called after the request was retried.
type Config ¶ added in v0.12.0
type Config struct { Enabled bool DryRun bool Window time.Duration Size int MaxFailoverDuration time.Duration MinTimeBetweenFailovers time.Duration DrainConcurrency int // keyspaces has the same purpose as "shards" but applies to a whole keyspace. Keyspaces map[string]bool // shards is a set of keyspace/shard entries to which buffering is limited. // If empty (and *enabled==true), buffering is enabled for all shards. Shards map[string]bool // contains filtered or unexported fields }
func NewConfigFromFlags ¶ added in v0.12.0
func NewConfigFromFlags() *Config
func NewDefaultConfig ¶ added in v0.12.0
func NewDefaultConfig() *Config
type RetryDoneFunc ¶
type RetryDoneFunc context.CancelFunc
RetryDoneFunc will be returned for each buffered request and must be called after the buffered request was retried. Without this signal, the buffer would not know how many buffered requests are currently retried.