GetOrchestrator slowness when PM Sender data is not cached (livepee…

…r#2849)

cmd/livepeer/starter/starter.go

@@ -55,9 +55,9 @@ var (
 	txCostMultiplier = 100
 
 	// The interval at which to clean up cached max float values for PM senders and balances per stream
-	cleanupInterval = 1 * time.Minute
+	cleanupInterval = 10 * time.Minute
 	// The time to live for cached max float values for PM senders (else they will be cleaned up) in seconds
-	smTTL = 60 // 1 minute
+	smTTL = 172800 // 2 days
 )
 
 const (

pm/sendermonitor.go

@@ -8,6 +8,7 @@ import (
 	"time"
 
 	ethcommon "github.com/ethereum/go-ethereum/common"
+	"github.com/ethereum/go-ethereum/common/hexutil"
 	"github.com/ethereum/go-ethereum/core/types"
 	"github.com/ethereum/go-ethereum/event"
 	"github.com/golang/glog"
@@ -272,7 +273,6 @@ func (sm *LocalSenderMonitor) cache(addr ethcommon.Address) {
 	queue.Start()
 	done := make(chan struct{})
 	go sm.startTicketQueueConsumerLoop(queue, done)
-
 	sm.senders[addr] = &remoteSender{
 		pendingAmount: big.NewInt(0),
 		queue:         queue,
@@ -347,6 +347,7 @@ func (sm *LocalSenderMonitor) cleanup() {
 			v.subScope.Close() // close the maxfloat subscriptions
 			delete(sm.senders, k)
 			sm.smgr.Clear(k)
+			glog.V(6).Infof("sender cleared from cache addr: %v last access: %v ttl: %vs", hexutil.Encode(k.Bytes()), v.lastAccess, sm.cfg.TTL)
 		}
 	}
 }