geektutu/7days-golang · error

RegisterPeerPicker called more than once

Error message

RegisterPeerPicker called more than once

What it means

Group.RegisterPeers panics if it is called twice on the same Group: g.peers is already non-nil. Peer selection (the PeerPicker) is intended to be configured exactly once per group; re-registering would silently replace the peer topology mid-flight, so the library panics to surface the double initialization.

Source

Thrown at gee-cache/day5-multi-nodes/geecache/geecache.go:77

// Get value for a key from cache
func (g *Group) Get(key string) (ByteView, error) {
	if key == "" {
		return ByteView{}, fmt.Errorf("key is required")
	}

	if v, ok := g.mainCache.get(key); ok {
		log.Println("[GeeCache] hit")
		return v, nil
	}

	return g.load(key)
}

// RegisterPeers registers a PeerPicker for choosing remote peer
func (g *Group) RegisterPeers(peers PeerPicker) {
	if g.peers != nil {
		panic("RegisterPeerPicker called more than once")
	}
	g.peers = peers
}

func (g *Group) load(key string) (value ByteView, err error) {
	if g.peers != nil {
		if peer, ok := g.peers.PickPeer(key); ok {
			if value, err = g.getFromPeer(peer, key); err == nil {
				return value, nil
			}
			log.Println("[GeeCache] Failed to get from peer", err)
		}
	}

	return g.getLocally(key)
}

func (g *Group) populateCache(key string, value ByteView) {

View on GitHub (pinned to cf36443821)

Solutions

1. Call RegisterPeers exactly once per Group, right after NewGroup
2. Recreate the Group (or guard registration with a sync.Once / nil check) instead of re-registering
3. Refactor startup so peer setup happens in a single code path

Example fix

// before
grp.RegisterPeers(peers)
// on reload:
grp.RegisterPeers(newPeers) // panic: RegisterPeerPicker called more than once
// after
if grp.GetPeers() == nil { // or use sync.Once
    grp.RegisterPeers(newPeers)
}

var peersRegistered sync.Once
func registerPeers(g *geecache.Group, p geecache.PeerPicker) {
    peersRegistered.Do(func() { g.RegisterPeers(p) })
}

defer func() {
    if r := recover(); r == "RegisterPeerPicker called more than once" {
        log.Println("peers already registered, ignoring")
    }
}()
g.RegisterPeers(peers)

When it happens

Trigger: Calling RegisterPeers twice on one Group — e.g. startCacheServer executed more than once for the same group in tests, re-running registration on config reload, or registering the same pool from two init paths.

Common situations: Test harness reusing a Group across subtests; hot-reload code re-applying peer config without recreating the Group; duplicated initialization in main and a helper.

Related errors

• RegisterPeerPicker called more than once
• RegisterPeerPicker called more than once
• nil Getter
• nil Getter
• nil Getter

AI-assisted analysis of geektutu/7days-golang@cf36443821 (2026-09-03). Data as JSON: /api/errors/ca2356c7a8e9576a. Report an issue: GitHub.