fix: cache successful volume lookups instead of failed ones (#7698)

The condition was inverted - it was caching lookups with errors
instead of successful lookups. This caused every replicated write
to make a gRPC call to master for volume location lookup, resulting
in ~1 second latency for writeToReplicas.

The bug particularly affected TTL volumes because:
- More unique volumes are created (separate pools per TTL)
- Volumes expire and get recreated frequently
- Each new volume requires a fresh lookup (cache miss)
- Higher volume churn = more cache misses = more master lookups

With this fix, successful lookups are cached for 10 minutes,
reducing replication latency from ~1s to ~10ms for cached volumes.

1 files changed, 4 insertions, 3 deletions

diff --git a/weed/operation/lookup.go b/weed/operation/lookup.go
index 9e9c719b5..12ad37330 100644
--- a/weed/operation/lookup.go
+++ b/weed/operation/lookup.go
@@ -4,12 +4,13 @@ import (
 	"context"
 	"errors"
 	"fmt"
-	"github.com/seaweedfs/seaweedfs/weed/pb"
-	"google.golang.org/grpc"
 	"math/rand/v2"
 	"strings"
 	"time"
 
+	"github.com/seaweedfs/seaweedfs/weed/pb"
+	"google.golang.org/grpc"
+
 	"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
 )
 
@@ -101,7 +102,7 @@ func LookupVolumeIds(masterFn GetMasterFn, grpcDialOption grpc.DialOption, vids
 					GrpcPort:   int(loc.GrpcPort),
 				})
 			}
-			if vidLocations.Error != "" {
+			if vidLocations.Error == "" {
 				vc.Set(vidLocations.VolumeOrFileId, locations, 10*time.Minute)
 			}
 			ret[vidLocations.VolumeOrFileId] = &LookupResult{