Add flag to select tx throttler tablet type

config/tablet/default.yaml

@@ -118,6 +118,7 @@ cacheResultFields: true                  # enable-query-plan-field-caching
 # enable-tx-throttler
 # tx-throttler-config
 # tx-throttler-healthcheck-cells
+# tx-throttler-tablet-types
 # enable_transaction_limit
 # enable_transaction_limit_dry_run
 # transaction_limit_per_user

doc/design-docs/ReplicationLagBasedThrottlingOfTransactions.md

@@ -30,7 +30,13 @@ If this is not specified a [default](https://github.com/vitessio/vitess/tree/mai
 * *tx-throttler-healthcheck-cells*
 
 A comma separated list of datacenter cells. The throttler will only monitor
-the non-RDONLY replicas found in these cells for replication lag.
+the replicas found in these cells for replication lag.
+
+* *tx-throttler-tablet-types*
+
+A comma separated list of tablet types. The throttler will only monitor tablets
+with these types. Only `replica` and/or `rdonly` types are supported. The default
+is `replica`.
 
 # Caveats and Known Issues
 * The throttler keeps trying to explore the maximum rate possible while keeping
@@ -39,4 +45,3 @@ lag limit may occasionally be slightly violated.
 
 * Transactions are considered homogeneous. There is currently no support
 for specifying how `expensive` a transaction is.
-

doc/design-docs/TabletServerParamsAsYAML.md

@@ -146,6 +146,7 @@ sanitizeLogMessages: false               # sanitize_log_messages
 # enable-tx-throttler
 # tx-throttler-config
 # tx-throttler-healthcheck-cells
+# tx-throttler-tablet-types
 # enable_transaction_limit
 # enable_transaction_limit_dry_run
 # transaction_limit_per_user

go/flags/endtoend/vttablet.txt

@@ -345,7 +345,8 @@ Usage of vttablet:
       --twopc_enable                                                     if the flag is on, 2pc is enabled. Other 2pc flags must be supplied.
       --tx-throttler-config string                                       Synonym to -tx_throttler_config (default "target_replication_lag_sec: 2\nmax_replication_lag_sec: 10\ninitial_rate: 100\nmax_increase: 1\nemergency_decrease: 0.5\nmin_duration_between_increases_sec: 40\nmax_duration_between_increases_sec: 62\nmin_duration_between_decreases_sec: 20\nspread_backlog_across_sec: 20\nage_bad_rate_after_sec: 180\nbad_rate_increase: 0.1\nmax_rate_approach_threshold: 0.9\n")
       --tx-throttler-healthcheck-cells strings                           Synonym to -tx_throttler_healthcheck_cells
-      --tx_throttler_config string                                       The configuration of the transaction throttler as a text formatted throttlerdata.Configuration protocol buffer message (default "target_replication_lag_sec: 2\nmax_replication_lag_sec: 10\ninitial_rate: 100\nmax_increase: 1\nemergency_decrease: 0.5\nmin_duration_between_increases_sec: 40\nmax_duration_between_increases_sec: 62\nmin_duration_between_decreases_sec: 20\nspread_backlog_across_sec: 20\nage_bad_rate_after_sec: 180\nbad_rate_increase: 0.1\nmax_rate_approach_threshold: 0.9\n")
+      --tx-throttler-tablet-types strings                                A comma-separated list of tablet types. Only tablets of this type are monitored for replication lag by the transaction throttler. Supported types are replica and/or rdonly. (default replica)
+      --tx_throttler_config string                                       The configuration of the transaction throttler as a text formatted throttlerdata.Configuration protocol buffer message. (default "target_replication_lag_sec: 2\nmax_replication_lag_sec: 10\ninitial_rate: 100\nmax_increase: 1\nemergency_decrease: 0.5\nmin_duration_between_increases_sec: 40\nmax_duration_between_increases_sec: 62\nmin_duration_between_decreases_sec: 20\nspread_backlog_across_sec: 20\nage_bad_rate_after_sec: 180\nbad_rate_increase: 0.1\nmax_rate_approach_threshold: 0.9\n")
       --tx_throttler_healthcheck_cells strings                           A comma-separated list of cells. Only tabletservers running in these cells will be monitored for replication lag by the transaction throttler.
       --unhealthy_threshold duration                                     replication lag after which a replica is considered unhealthy (default 2h0m0s)
       --use_super_read_only                                              Set super_read_only flag when performing planned failover.

go/vt/vttablet/tabletserver/tabletenv/config.go

@@ -31,8 +31,10 @@ import (
 	"vitess.io/vitess/go/vt/dbconfigs"
 	"vitess.io/vitess/go/vt/log"
 	querypb "vitess.io/vitess/go/vt/proto/query"
+	topodatapb "vitess.io/vitess/go/vt/proto/topodata"
 	"vitess.io/vitess/go/vt/servenv"
 	"vitess.io/vitess/go/vt/throttler"
+	"vitess.io/vitess/go/vt/topo/topoproto"
 )
 
 // These constants represent values for various config parameters.
@@ -138,8 +140,9 @@ func registerTabletEnvFlags(fs *pflag.FlagSet) {
 	fs.StringVar(&currentConfig.TwoPCCoordinatorAddress, "twopc_coordinator_address", defaultConfig.TwoPCCoordinatorAddress, "address of the (VTGate) process(es) that will be used to notify of abandoned transactions.")
 	SecondsVar(fs, &currentConfig.TwoPCAbandonAge, "twopc_abandon_age", defaultConfig.TwoPCAbandonAge, "time in seconds. Any unresolved transaction older than this time will be sent to the coordinator to be resolved.")
 	flagutil.DualFormatBoolVar(fs, &currentConfig.EnableTxThrottler, "enable_tx_throttler", defaultConfig.EnableTxThrottler, "If true replication-lag-based throttling on transactions will be enabled.")
-	flagutil.DualFormatStringVar(fs, &currentConfig.TxThrottlerConfig, "tx_throttler_config", defaultConfig.TxThrottlerConfig, "The configuration of the transaction throttler as a text formatted throttlerdata.Configuration protocol buffer message")
+	flagutil.DualFormatStringVar(fs, &currentConfig.TxThrottlerConfig, "tx_throttler_config", defaultConfig.TxThrottlerConfig, "The configuration of the transaction throttler as a text formatted throttlerdata.Configuration protocol buffer message.")
 	flagutil.DualFormatStringListVar(fs, &currentConfig.TxThrottlerHealthCheckCells, "tx_throttler_healthcheck_cells", defaultConfig.TxThrottlerHealthCheckCells, "A comma-separated list of cells. Only tabletservers running in these cells will be monitored for replication lag by the transaction throttler.")
+	fs.Var((*topoproto.TabletTypeListFlag)(&currentConfig.TxThrottlerTabletTypes), "tx-throttler-tablet-types", "A comma-separated list of tablet types. Only tablets of this type are monitored for replication lag by the transaction throttler. Supported types are replica and/or rdonly.")
 
 	fs.BoolVar(&enableHotRowProtection, "enable_hot_row_protection", false, "If true, incoming transactions for the same row (range) will be queued and cannot consume all txpool slots.")
 	fs.BoolVar(&enableHotRowProtectionDryRun, "enable_hot_row_protection_dry_run", false, "If true, hot row protection is not enforced but logs if transactions would have been queued.")
@@ -310,9 +313,10 @@ type TabletConfig struct {
 	TwoPCCoordinatorAddress string  `json:"-"`
 	TwoPCAbandonAge         Seconds `json:"-"`
 
-	EnableTxThrottler           bool     `json:"-"`
-	TxThrottlerConfig           string   `json:"-"`
-	TxThrottlerHealthCheckCells []string `json:"-"`
+	EnableTxThrottler           bool                    `json:"-"`
+	TxThrottlerConfig           string                  `json:"-"`
+	TxThrottlerHealthCheckCells []string                `json:"-"`
+	TxThrottlerTabletTypes      []topodatapb.TabletType `json:"-"`
 
 	EnableLagThrottler bool `json:"-"`
 	EnableTableGC      bool `json:"-"` // can be turned off programmatically by tests
@@ -561,6 +565,7 @@ var defaultConfig = TabletConfig{
 	EnableTxThrottler:           false,
 	TxThrottlerConfig:           defaultTxThrottlerConfig(),
 	TxThrottlerHealthCheckCells: []string{},
+	TxThrottlerTabletTypes:      []topodatapb.TabletType{topodatapb.TabletType_REPLICA},
 
 	EnableLagThrottler: false, // Feature flag; to switch to 'true' at some stage in the future
 

go/vt/vttablet/tabletserver/txthrottler/tx_throttler.go

@@ -109,6 +109,15 @@ func tryCreateTxThrottler(config *tabletenv.TabletConfig, topoServer *topo.Serve
 		return newTxThrottler(&txThrottlerConfig{enabled: false})
 	}
 
+	for _, tabletType := range config.TxThrottlerTabletTypes {
+		switch tabletType {
+		case topodatapb.TabletType_REPLICA, topodatapb.TabletType_RDONLY:
+			continue
+		default:
+			return nil, fmt.Errorf("%q tablet type is not supported", tabletType)
+		}
+	}
+
 	var throttlerConfig throttlerdatapb.Configuration
 	if err := prototext.Unmarshal([]byte(config.TxThrottlerConfig), &throttlerConfig); err != nil {
 		return nil, err
@@ -122,6 +131,7 @@ func tryCreateTxThrottler(config *tabletenv.TabletConfig, topoServer *topo.Serve
 	return newTxThrottler(&txThrottlerConfig{
 		enabled:          true,
 		topoServer:       topoServer,
+		tabletConfig:     config,
 		throttlerConfig:  &throttlerConfig,
 		healthCheckCells: healthCheckCells,
 	})
@@ -136,6 +146,7 @@ type txThrottlerConfig struct {
 	enabled bool
 
 	topoServer      *topo.Server
+	tabletConfig    *tabletenv.TabletConfig
 	throttlerConfig *throttlerdatapb.Configuration
 	// healthCheckCells stores the cell names in which running vttablets will be monitored for
 	// replication lag.
@@ -166,6 +177,8 @@ type TopologyWatcherInterface interface {
 
 // txThrottlerState holds the state of an open TxThrottler object.
 type txThrottlerState struct {
+	tabletConfig *tabletenv.TabletConfig
+
 	// throttleMu serializes calls to throttler.Throttler.Throttle(threadId).
 	// That method is required to be called in serial for each threadId.
 	throttleMu      sync.Mutex
@@ -283,7 +296,8 @@ func newTxThrottlerState(config *txThrottlerConfig, keyspace, shard, cell string
 		return nil, err
 	}
 	result := &txThrottlerState{
-		throttler: t,
+		tabletConfig: config.tabletConfig,
+		throttler:    t,
 	}
 	createTxThrottlerHealthCheck(config, result, cell)
 
@@ -351,14 +365,12 @@ func (ts *txThrottlerState) deallocateResources() {
 
 // StatsUpdate updates the health of a tablet with the given healthcheck.
 func (ts *txThrottlerState) StatsUpdate(tabletStats *discovery.TabletHealth) {
-	// Ignore PRIMARY and RDONLY stats.
-	// We currently do not monitor RDONLY tablets for replication lag. RDONLY tablets are not
-	// candidates for becoming primary during failover, and it's acceptable to serve somewhat
-	// stale date from these.
-	// TODO(erez): If this becomes necessary, we can add a configuration option that would
-	// determine whether we consider RDONLY tablets here, as well.
-	if tabletStats.Target.TabletType != topodatapb.TabletType_REPLICA {
-		return
+	// Monitor tablets for replication lag if they have a tablet
+	// type specified by the --tx_throttler_tablet_types flag.
+	for _, expectedTabletType := range ts.tabletConfig.TxThrottlerTabletTypes {
+		if tabletStats.Target.TabletType == expectedTabletType {
+			ts.throttler.RecordReplicationLag(time.Now(), tabletStats)
+			return
+		}
 	}
-	ts.throttler.RecordReplicationLag(time.Now(), tabletStats)
 }

go/vt/vttablet/tabletserver/txthrottler/tx_throttler_test.go

@@ -115,6 +115,7 @@ func TestEnabledThrottler(t *testing.T) {
 	config := tabletenv.NewDefaultConfig()
 	config.EnableTxThrottler = true
 	config.TxThrottlerHealthCheckCells = []string{"cell1", "cell2"}
+	config.TxThrottlerTabletTypes = []topodatapb.TabletType{topodatapb.TabletType_REPLICA}
 
 	throttler, err := tryCreateTxThrottler(config, ts)
 	if err != nil {