C# New Added 9 September 2026

EF Core: another instance with the same key value is already being tracked

The change tracker already holds an entity with that primary key and Attach or Update tried to add a second copy. It normally follows a tracked read earlier in the request, then an update built from the request body. The first query is the cause even though the second call is what throws.

Quick fix

Read the commands before running them. Anything that restarts a service, deletes data or changes permissions should be tried on a non-production system first.

Quick fix

// Read only queries should not track
var order = await db.Orders.AsNoTracking().FirstAsync(o => o.Id == id);

// Update the entity that is already tracked instead of attaching a new one
var tracked = await db.Orders.FindAsync(id);
db.Entry(tracked).CurrentValues.SetValues(dto);
await db.SaveChangesAsync();

// See what the context is holding when it throws
foreach (var e in db.ChangeTracker.Entries())
    Console.WriteLine($"{e.Entity.GetType().Name} {e.State}");

// A long lived context makes this constant: it is meant to be per request
builder.Services.AddDbContext<AppDb>(o => o.UseNpgsql(cs));

How to diagnose C# errors

C# errors divide into compile-time and restore-time problems (NuGet resolution, target framework mismatches) and runtime problems dominated by null references and async misuse. The classic .NET production hazard is the sync-over-async deadlock, calling .Result or .Wait() on a task in a context with a synchronisation context, which manifests as a hang rather than an exception and is therefore much harder to spot.

If the quick fix above does not resolve it, work through these steps. They apply to this whole class of error, not just to this one message, which is usually what saves the time.

Run dotnet restore --verbosity detailed to see which feed each package resolved from. Most restore failures are a private feed with expired credentials, not a missing package.
Enable nullable reference types (<Nullable>enable</Nullable>) and treat the warnings as errors. This converts a whole class of production NullReferenceExceptions into compile failures.
For async hangs, search the codebase for .Result, .Wait() and .GetAwaiter().GetResult(). Replace with await all the way up, or use ConfigureAwait(false) in library code.
For EF Core, run dotnet ef migrations list to see which migrations the database believes are applied before generating a new one.
Capture a dump with dotnet-dump and inspect it with dotnet-gcdump / dotnet-counters when the process misbehaves without throwing.

Tools worth reaching for

dotnet restore -v detailed
dotnet ef migrations list
dotnet-counters
dotnet-dump

Authoritative references

Primary documentation for this error, worth reading before applying any fix in production.

learn.microsoft.com

Related C# errors

See all 12 C# errors →

Browse other categories

Something missing or wrong?

This entry is maintained by hand. If the fix is out of date, incomplete, or you have a better one, email a correction and it will be reviewed.