Your ViewModel Won't Save You: Designing for Android Process Death

  How to use SavedStateHandle and the Single Source of Truth pattern to build "death-proof" apps that never lose user progress.

Your ViewModel Won't Save You: Designing for Android Process Death

TL;DR:

  • ViewModel survives rotations, but not process death.
  • LMK (Low Memory Killer) will kill your background app to reclaim RAM.
  • SavedStateHandle is your tool for restoring small UI states (IDs, flags).
  • Room/DataStore is required for critical business data persistence.
  • Test using adb shell am kill <package> to see your app's true resilience.

The Ghost in the Machine

You’re in a banking app, halfway through a wire transfer. You switch to your email to copy an OTP. When you return, the app has reset to the login screen, and your progress is gone.

To the user, this is a “glitch.” To a seasoned Android developer, this is a failure to architect for Process Death. Android enforces strict process management regardless of available RAM. If you want to build professional software, you must stop assuming your app will stay alive in the background.

1. The Visual Mental Model: How State is Lost & Found

Understanding the timeline of a “Background Kill” is the first step to fixing it:

The Process Death Flow:

  1. User leaves app → App enters the background.
  2. Memory Pressure → System (LMK) kills the app process.
  3. User returns → Android recreates the Activity and ViewModel.
  4. The Gap → Standard variables in ViewModel are now null or reset.
  5. The Rescue → SavedStateHandle restores UI state → Repository re-fetches data.

2. The Bouncer: How LMK Really Works

The Low Memory Killer Daemon (lmkd) is the system’s “bouncer.” Modern Android (10+) utilizes PSI (Pressure Stall Information) alongside OOM (Out Of Memory) Adjustment scores.

PSI allows the system to detect memory pressure before the device starts “thrashing.” If your app is in the background and consuming significant resources (like unoptimized bitmaps), it becomes the LMK’s primary target.

Pro Tip: Visibility is not a guarantee of survival. Process death can occur even while your app is visible under extreme system pressure — though it is far more common once the app is sent to the background.

3. ❌ Common Mistakes Developers Make

Before we look at the solution, are you committing these architectural “sins”?

  • Storing API responses in ViewModel variables: These vanish instantly during process death.
  • Assuming high-end devices are “safe”: Even a phone with 16GB of RAM will kill background processes to prioritize a 4K video recording or a heavy gaming session.
  • Using Singletons as state holders: Process death wipes the entire memory heap, including your static Singletons.
  • Saving large objects in the Bundle: This leads to the dreaded TransactionTooLargeException.

4. The Solution: Resilience via SavedStateHandle

SavedStateHandle is backed by the system’s saved instance state (Bundle) mechanism. It allows your ViewModel to access the bundle data during recreation without needing to reference the Activity directly.

💻 Production-Grade Kotlin Implementation

Here is how to structure a Search feature using a Single Source of Truth (SSOT) pattern.

5. Testing the “Recreation” Flow

Don’t rely on the “Don’t keep activities” toggle in Developer Options. It only destroys the Activity, not the process. To simulate a real-world LMK event, use this ADB command while your app is in the background:

🙋 Frequently Asked Questions (FAQs)

Does this happen if the user swipes the app away?

Yes — but with a key difference. When the user swipes the app away from “Recents,” the process is killed and the saved state is discarded. The app starts completely fresh next time. For data that must survive a swipe-away (like a shopping cart), always use Room or DataStore.

Is rememberSaveable the same thing?

In Jetpack Compose, rememberSaveable uses the same underlying Bundle mechanism as SavedStateHandle. It’s perfect for UI-only state like scroll positions within a Composable.

How much data is “too much” for the Bundle?

As a rule of thumb, keep it well under the ~1MB Binder transaction limit (ideally <100KB). If you find yourself saving large objects, save an ID instead and re-fetch the data from your database during recreation.

🏁 Final Thoughts

Android memory management is aggressive by design. Process death isn’t a bug you should try to prevent; it’s a lifecycle event you must embrace. By separating your Transient UI State (SavedStateHandle) from your Persistent Business Data (Room), you ensure that your users never lose their place again.

If your app can’t survive process death, it isn’t production-ready.

💬 Join the Conversation

  • Which famous app have you seen fail the process death test? (OTP screens are the biggest culprits!)
  • How do you handle the “Loading” state when a process is recreated?
  • Next Up: Deep Dive: SavedStateHandle vs. rememberSaveable — Which should you choose?

Stop assuming your app will stay alive. Build for recreation.

📘 Master Your Next Technical Interview

Since Java is the foundation of Android development, mastering DSA is essential. I highly recommend “Mastering Data Structures & Algorithms in Java”. It’s a focused roadmap covering 100+ coding challenges to help you ace your technical rounds.


Comments

Post a Comment

Popular posts from this blog

No More _state + state: Simplifying ViewModels with Kotlin 2.3

Image
 Clean up your Android code with Explicit Backing Fields—the modern way to handle read-only state without the boilerplate. TL;DR:  Kotlin 2.3 introduces  explicit backing fields , allowing you to replace the classic  _state  +  state  boilerplate with a single, cleaner property—provided you're okay with a few experimental trade-offs. If you’ve been developing Android apps, you know the “Backing Property Dance.” It’s that repetitive ritual where we create a private  MutableStateFlow  and a public  StateFlow  just to ensure encapsulation. The “Old” Way (The Boilerplate) private val _uiState = MutableStateFlow(UiState.Loading) // Exposing a read-only view of the mutable internal state val uiState: StateFlow<UiState> = _uiState.asStateFlow() We’ve done this for years to prevent external classes from mutating our state. While effective, it pollutes our IDE suggestions with underscores and adds unnecessary ceremony to every singl...
Read more

Why You Should Stop Passing ViewModels Around Your Compose UI Tree 🚫

Image
 Master State Hoisting and the "Route-level" pattern to build scalable, testable, and preview-friendly Android apps. If you’ve been building with Jetpack Compose for a while, you know how tempting it is to just pass a  ViewModel  instance down through five layers of UI components. It’s convenient—one object gives you all the data and all the functions you need. But here is the architectural reality:  Passing ViewModel instances deep into your UI tree is an architectural anti-pattern.  It couples your visual components to business logic and makes your code significantly harder to maintain, test, and preview. The Core Problem: Lifecycle and Scoping ViewModels are designed to be state holders tied to a specific lifecycle (usually a Navigation Destination). When you pass a ViewModel into a small UI component, you break the  Unidirectional Data Flow (UDF) . ❌ The “Bad” Way (Tight Coupling) // ❌ BAD: This component is now impossible to preview or reuse @Composabl...
Read more

Is Jetpack Compose Making Your APK Fatter? (And How to Fix It)

Image
 Unpacking the Compose Tax, compiler-driven DEX optimization, and why your resource strategy must evolve for modern UI. TL;DR DEX Bloat:  Compose’s compiler injects “Restartable” and “Skippable” logic for memoization; R8 minification is effectively required. Code vs. Assets:  Moving iconography to  ImageVector  constants allows R8 to prune unused assets as code. Tooling Leak:   ui-tooling  in release builds is the #1 cause of "false" bloat. The Scale Factor:  Compose has a higher “size floor” but a much lower “size ceiling” than the legacy View system. The promise of Jetpack Compose was simple: “Write less code, ship faster.” But as teams migrate, a quiet concern often echoes in code reviews:  “Why did our APK size just jump?” While Compose removes the weight of deep XML hierarchies and the overhead of ViewBinding, it introduces a unique “Compose Tax.” For a senior engineer, the goal isn’t just to accept this tax, but to optimize the runtime...
Read more