Quick answer: Only one peer should be the authority for each node, and all peers must agree on who that is. Set authority on the server and use @rpc("authority", "call_local", "reliable") for state-changing calls. Verify with is_multiplayer_authority() before writing replicated values.

Here is how to fix Godot multiplayer authority desync. You have a shooter prototype: server plus two clients. Everyone spawns a player. They see each other move. Then player A shoots player B, and on A’s screen B dies; on B’s screen A dies; on C’s screen both are still alive. The shots registered, but the damage applied differently on each peer. Authority in Godot 4 is per-node, and disagreement about authority is the root of most multiplayer desync.

The Symptom

Clients display different game states. Possible manifestations:

Logs on different peers show different current values. No RPC errors in the console.

What Causes This

Authority not explicitly set. New nodes default to authority = 1 (server). If your peer IDs are not what you expect, or if a client-spawned node assumes authority it should not have, writes happen on the wrong peer.

Multiple peers claim authority. If client A calls set_multiplayer_authority(A_id) on a shared node and client B does the same with B_id, both now try to drive the same node. Whoever sends last wins on each receiver — inconsistent state emerges.

MultiplayerSynchronizer replication paths mismatched. A Synchronizer replicates specific properties. If the property path is typoed or refers to a node that differs between peers (e.g. different scene instance structure), replication silently fails for that property.

RPC without call_local. @rpc("authority") without "call_local" only runs on remote peers. If the authority needs to execute the same logic locally, it skipped. Add "call_local" for write calls that affect all peers including the caller.

Scene spawned only on some peers. If you add_child a scene on one peer without MultiplayerSpawner or equivalent replication, other peers do not have that node. Calls referencing it fail silently.

The Fix

Step 1: Use MultiplayerSpawner for networked scenes. A MultiplayerSpawner node handles spawning scenes across all peers automatically:

extends Node # Game root

@onready var spawner = $MultiplayerSpawner

func _ready():
    spawner.spawn_path = ^"Players"
    spawner.spawn_limit = 16
    # Register player scene for replication
    spawner.add_spawnable_scene("res://player.tscn")

func spawn_player(peer_id: int):
    if not multiplayer.is_server():
        return
    var p = preload("res://player.tscn").instantiate()
    p.name = str(peer_id)
    p.set_multiplayer_authority(peer_id) # peer owns their player
    $Players.add_child(p, true)

Spawning only on the server with add_child propagates to all peers automatically via the spawner. Authority is set before replication so all peers receive the correct ownership info.

Step 2: Guard writes with is_multiplayer_authority. Always check authority before writing to replicated state:

func take_damage(amount: int):
    if not is_multiplayer_authority():
        return # only authority can change health
    health -= amount
    if health <= 0:
        die.rpc() # broadcast death to all

@rpc("authority", "call_local", "reliable")
func die():
    queue_free()

Non-authority peers early-return. The authority applies changes, then RPC-broadcasts events. All peers run the RPC (including the authority via call_local), so die() runs everywhere.

Step 3: Use MultiplayerSynchronizer for continuous state. Add a MultiplayerSynchronizer child to your player. Configure Replication Config with paths to properties you want synced:

# In editor: MultiplayerSynchronizer > Replication Config
# Add properties:
#   position (spawn + sync)
#   rotation (sync only)
#   health (sync only)

# At runtime, only the authority writes these values
func _physics_process(delta):
    if not is_multiplayer_authority():
        return

    # authority applies input and physics
    var input_vec = Input.get_vector(...)
    velocity = input_vec * speed
    move_and_slide()

The Synchronizer at network rate (default 30 Hz) pushes authority’s property values to all peers. Non-authority peers just display what they receive.

Step 4: Diagnose with debug prints. On every peer, log authority state:

func _ready():
    print("Peer ", multiplayer.get_unique_id(),
        " says: this node authority = ", get_multiplayer_authority())

Run with three peers (server + 2 clients). All three should report the same authority for the same node. If they disagree, authority was set inconsistently — usually because clients set it locally without waiting for server.

High-Latency Smoothing

For smooth motion despite network jitter, enable interpolation on MultiplayerSynchronizer via its replication config. Or implement client-side prediction: authority sets “target position,” non-authority lerps current position toward target over multiple frames.

Understanding the issue

Multiplayer code has a different correctness model than single-player code. It must tolerate latency, packet loss, and out-of-order delivery while preserving game-state consistency. Each tolerance is engineering work; you choose which network conditions to handle.

The specific bug described above is the kind that surfaces during integration rather than unit testing. It depends on a combination of factors: the asset configuration, the runtime state, the platform's specific behavior. In isolation, each piece looks correct; in combination, the bug emerges. This is why thorough integration testing - playing the actual game in realistic conditions - catches things that automated tests miss.

Why this happens

Bugs of this class are particularly easy to ship past internal QA because they often depend on specific runtime conditions - hardware combinations, network states, or asset configurations that QA didn't reproduce. Players hit them in the wild, file reports that are hard to repro, and the bug accumulates negative reviews while engineering tries to recreate the failure mode.

At the engine level, the behavior comes from a deliberate design decision in Godot. The engine team chose a particular trade-off - usually performance versus convenience, or generality versus specificity - and that trade-off has consequences when you push against it. Understanding the trade-off is what turns 'this bug is mysterious' into 'this bug is the expected consequence of this design'.

Verifying the fix

Verifying this fix in isolation is straightforward: reproduce the bug, apply the change, confirm the bug no longer reproduces. The harder verification is regression - did this fix introduce a new bug elsewhere? Run your standard regression suite, plus any tests that exercise the same code path with different inputs.

Reproducibility is the prerequisite for verification. If you can't reliably reproduce the bug pre-fix, you can't reliably verify it post-fix. Spend time getting a clean reproduction before you write any fix code. The fix is fast once you understand the reproduction; the reproduction is the slow part.

Variations to watch for

Related bug classes often share the same root cause. If you find yourself fixing this issue, look for cousins: similar symptoms in adjacent systems, the same data flow but a different value, or the same fix pattern in another module. The catalog of 'we've seen this before' becomes valuable institutional knowledge.

Adjacent bugs often share a root cause. After fixing the case you've found, spend an hour searching the codebase for similar patterns. What's the same call with different arguments? The same data flow with a different entity type? The same lifecycle issue in a sibling system? Each match is a candidate for the same fix, or a related fix that prevents future bugs of the same class.

In production

In shipping builds, this issue may interact with other production-only behavior. Stripping, encryption, asset bundling, and platform-specific code paths can each modify the symptoms. When players report a related issue, capture build SHA, platform, and any feature flags - those three fields cover most of the production-only variations.

When triaging a similar issue in production, prioritize gathering data over hypothesizing causes. A player report describes a symptom; what you need is a build SHA, a session timestamp, and ideally a screen recording or session replay. With those, the bug becomes tractable. Without them, you're guessing at hypothetical reproductions that may not match what the player actually hit.

Performance considerations

Performance implications matter when this bug class scales with player count or asset count. A bug that fires once per session is annoying; a bug that fires once per frame compounds. After fixing, profile the affected code path under realistic load. The fix that's correct for one entity may be too slow for ten thousand.

Diagnostic approach

Diagnosing this class of bug benefits from a structured approach: confirm the symptom, isolate the variables, hypothesize the cause, and verify the hypothesis before writing fix code. Skipping the isolation step is the most common mistake; without it, fixes often address symptoms while the underlying cause continues to produce other variations.

For Godot-specific diagnostics, the editor's profiler is the canonical starting point. Capture a representative frame with the symptom present; compare against a frame without the symptom; the diff often points directly at the cause. If the symptom is non-deterministic, capture multiple frames and look for the pattern - the cause is usually a state transition or a specific input value rather than a continuous effect.

Tooling and ecosystem

Third-party plugins often provide better diagnostics for their own behavior than the engine does. If the affected code is in a plugin, check the plugin's documentation for debug modes, verbose logging, or inspector tools - these can save hours of investigation when they exist.

Within Godot, the relevant diagnostic surfaces include the standard frame debugger, memory profiler, and engine-specific debug overlays. Each one shows a different facet of what's happening. The frame debugger reveals draw call ordering and state transitions; the memory profiler shows allocation patterns; the debug overlay reveals per-system state. Bugs that resist one tool usually surrender to another - the trick is knowing which tool to reach for first.

Edge cases and pitfalls

Platform-specific edge cases are worth enumerating explicitly. iOS handles backgrounding differently than Android; Windows handles focus changes differently than macOS. A fix that works on the development platform may not work on every target. Test on each shipping platform deliberately.

When writing a regression test for this fix, focus on the boundary conditions that surfaced the original bug. Tests that exercise the happy path catch obvious regressions; tests that exercise the boundary catch the subtler regressions that look like new bugs but are really the original returning. The latter are the tests that earn their keep over the long life of the project.

Team communication

When this bug class affects multiple teams (often the case for cross-system issues), early communication prevents duplicate work. The team that owns the symptom may not own the cause. A 15-minute conversation at the start of triage often saves hours of independent investigation.

If this fix touches a system several engineers work in, a short writeup in the team's engineering channel helps. Not a full design doc - a paragraph explaining what was wrong, what's fixed, and what to watch for. Future engineers encountering similar symptoms will search for the fix; making it findable is a small investment that pays back later.

“One authority per node. All peers agree. Set by the server, never by individual clients. Multiplayer desync dies.”

Related Issues

For signal handling in multiplayer, see Godot Await Signal. For general multiplayer testing, Testing Cross-Platform Multiplayer covers related debug techniques.

Server spawns, sets authority. is_multiplayer_authority() guards writes. Synchronizer for state, RPCs for events.