From 058569c77b82bf4d30454ddd74baba45117a5498 Mon Sep 17 00:00:00 2001
From: Chris Lu <chrislusf@users.noreply.github.com>
Date: Sun, 7 Jun 2026 11:46:57 -0700
Subject: [PATCH] operation: index VidCache by map instead of slice (#9853)

VidCache.cache was a []VidInfo indexed directly by volume id, so caching
one volume with a large id grew the backing array to that many entries
(each 48 bytes), allocating a zeroed slot for every unused id below it. A
single id of 32M cost ~1.5GB resident, plus geometric realloc churn as the
append loop doubled the array.

Use map[uint32]VidInfo so memory scales with the number of volumes actually
cached rather than the largest id seen. Parse ids with ParseUint(.,32) so
values outside the uint32 volume-id range are rejected instead of silently
wrapping into a key.
---
 weed/operation/lookup_vid_cache.go      | 41 +++++++++++++------------
 weed/operation/lookup_vid_cache_test.go | 22 +++++++++++++
 2 files changed, 44 insertions(+), 19 deletions(-)

diff --git a/weed/operation/lookup_vid_cache.go b/weed/operation/lookup_vid_cache.go
index 248fc17de..e2ff405a9 100644
--- a/weed/operation/lookup_vid_cache.go
+++ b/weed/operation/lookup_vid_cache.go
@@ -17,43 +17,46 @@ type VidInfo struct {
 }
 type VidCache struct {
 	sync.RWMutex
-	cache []VidInfo
+	cache map[uint32]VidInfo
 }
 
 func (vc *VidCache) Get(vid string) ([]Location, error) {
-	id, err := strconv.Atoi(vid)
+	id, err := strconv.ParseUint(vid, 10, 32)
 	if err != nil {
 		glog.V(1).Infof("Unknown volume id %s", vid)
 		return nil, err
 	}
+	if id == 0 {
+		return nil, ErrorNotFound
+	}
 	vc.RLock()
 	defer vc.RUnlock()
-	if 0 < id && id <= len(vc.cache) {
-		if vc.cache[id-1].Locations == nil {
-			return nil, errors.New("not set")
-		}
-		if vc.cache[id-1].NextRefreshTime.Before(time.Now()) {
-			return nil, errors.New("expired")
-		}
-		return vc.cache[id-1].Locations, nil
+	info, found := vc.cache[uint32(id)]
+	if !found || info.Locations == nil {
+		return nil, ErrorNotFound
 	}
-	return nil, ErrorNotFound
+	if info.NextRefreshTime.Before(time.Now()) {
+		return nil, errors.New("expired")
+	}
+	return info.Locations, nil
 }
+
 func (vc *VidCache) Set(vid string, locations []Location, duration time.Duration) {
-	id, err := strconv.Atoi(vid)
+	id, err := strconv.ParseUint(vid, 10, 32)
 	if err != nil {
 		glog.V(1).Infof("Unknown volume id %s", vid)
 		return
 	}
+	if id == 0 {
+		return
+	}
 	vc.Lock()
 	defer vc.Unlock()
-	if id > len(vc.cache) {
-		for i := id - len(vc.cache); i > 0; i-- {
-			vc.cache = append(vc.cache, VidInfo{})
-		}
+	if vc.cache == nil {
+		vc.cache = make(map[uint32]VidInfo)
 	}
-	if id > 0 {
-		vc.cache[id-1].Locations = locations
-		vc.cache[id-1].NextRefreshTime = time.Now().Add(duration)
+	vc.cache[uint32(id)] = VidInfo{
+		Locations:       locations,
+		NextRefreshTime: time.Now().Add(duration),
 	}
 }
diff --git a/weed/operation/lookup_vid_cache_test.go b/weed/operation/lookup_vid_cache_test.go
index 9c9e2affb..a6ca0a556 100644
--- a/weed/operation/lookup_vid_cache_test.go
+++ b/weed/operation/lookup_vid_cache_test.go
@@ -24,3 +24,25 @@ func TestCaching(t *testing.T) {
 		t.Fatal("Not found vid 123")
 	}
 }
+
+// a single large volume id must not allocate an entry per id below it
+func TestCachingLargeVolumeId(t *testing.T) {
+	var vc VidCache
+	locations := []Location{{Url: "a.com:8080"}}
+	vc.Set("32000000", locations, time.Minute)
+	if got := len(vc.cache); got != 1 {
+		t.Fatalf("expected 1 cached entry, got %d", got)
+	}
+	if ret, _ := vc.Get("32000000"); ret == nil {
+		t.Fatal("Not found vid 32000000")
+	}
+
+	// ids beyond uint32 are not real volume ids and must not wrap into the cache
+	vc.Set("4294967296", locations, time.Minute)
+	if got := len(vc.cache); got != 1 {
+		t.Fatalf("out-of-range id must not be cached, got %d entries", got)
+	}
+	if ret, _ := vc.Get("4294967296"); ret != nil {
+		t.Fatal("out-of-range vid 4294967296 should not be found")
+	}
+}