Quick answer: Write to the NetworkVariable on the server (or grant owner write permission), and assign a whole new value rather than mutating a struct's fields in place so dirty tracking fires.
A NetworkVariable in Netcode for GameObjects only replicates when the authorized writer assigns it and the value is detected as changed. Mutating a struct in place or writing from an unauthorized side silently drops the update.
How to fix it
1. Write from the authorized side
By default only the server can write a NetworkVariable. Set NetworkVariableWritePermission.Owner in the constructor if the owner must write, otherwise change the value inside a ServerRpc so the server applies it.
2. Assign a new value, do not mutate
For struct-typed NetworkVariables, read the value into a local, change it, then assign it back: hp.Value = new Health{cur=5}. Editing hp.Value.cur directly does not mark the variable dirty and never replicates.
3. Confirm spawn before writing
Writes before NetworkObject.Spawn() are lost. Set initial values in OnNetworkSpawn on the server, and subscribe to OnValueChanged there so late state is applied.
Catching the ones you can't reproduce
The hardest version of this to fix is the one you can't reproduce — it only happens on a player's hardware, OS, driver, or save state, under conditions that simply aren't present on your machine. A report that says “it crashed” or “it froze” gives you nothing to act on, so the bug survives release after release while quietly costing you players.
Automatic error capture closes that gap. Each failure arrives with its full stack trace, the device and OS, the build number, and a breadcrumb trail of what the player did right before it broke, so even a failure you have never seen becomes a specific, reproducible issue. Fold identical failures into one signature ranked by how many players each hits, and your worklist sorts itself worst-first instead of arriving as a stream of vague complaints.
This is where a tool like Bugnet earns its place. Its SDK captures every Unity error automatically with the full stack trace plus device, OS, memory, build, and game-state context, folds duplicates into one grouped issue with an occurrence count, and ties each to the build it first appeared on — so you fix the problem that hurts the most players first and confirm it is gone when its signature disappears from the next release.
The bug you can't reproduce isn't gone — it's just invisible until you capture it from the player's device.