Open In App

Memory Leaks in Android

Improve
Improve
Like Article
Like
Save
Share
Report

A memory leak is basically a failure of releasing unused objects from the memory. As a developer one does not need to think about memory allocation, memory deallocation, and garbage collection. All of these are the automatic process that the garbage collector does by itself, but the situation becomes difficult for the garbage collector when the user is referencing the object, which is in not use anymore, but since that object is being referenced by another object, garbage collector feels that the unused object is being used by another object and due to this garbage collector does not free-up the memory of that object, due to which available heap memory get decreases leading to memory shortage and memory leak. The unfreed object is basically called as leaks.

Memory leaks are the common causes of application crashes in android apps. Every developer must know how to avoid memory leaks and what are the circumstances which can lead to memory leaks in android applications. When the RAM resources are not released when they are no longer needed and if this is done several times, the part of the memory that the operating system has assigned to that particular application may exceed the upper limit and the system can terminate the application causing the application to crash. Holding the references of the object and resources that are no longer needed is the main cause of the memory leaks in android applications.  As it is known that the memory for the particular object is allocated within the heap and the object point to certain resources using some object reference. But when the work is completed, the object references should be freed. But when it is not done, the heap space increases continuously and the rest of the application has to run on whatever heap space that is left, and ultimately there are high chances that it can lead to memory leaks in the application. Thus it can be said that memory leaks are a term used when our app goes short of the memory because some object which is not used but still there are being pointed out by the references and continually occupying the heap space, which ultimately leads to a shortage of space for the other components of the application and thus eventually causing the app to crash.

Note: One needs to remember that whenever there is a shortage of space in the heap and the system needs to allocate space for some new objects, the garbage collector is being called in the frequent intervals, causing to slow down of the application or sometime the application may crash.

Causes of Memory Leaks and Their Solutions

1. Using Static Views

One should not use static views while developing the application, as static views are never destroyed.

2. Using Static Context

One should never use the Context as static, because that context will be available through the life of the application, and will not be restricted to the particular activity.

public class MainActivity extends AppCompatActivity {

     // this should not be done

     private static Button button;

}

3. Using Code Abstraction Frequently 

Developers often take the advantage of the abstraction property because it provides the code maintenance and flexibility in code, but using abstraction might cost a bit to the developer, as while using abstraction one needs to write more code, more code means more time to execute the space and more RAM. So whenever one can avoid the abstraction in the code, it is code as it can lead to fewer memory leaks.

4. Unregistered  Listeners

When the developer uses any kind of listener in his application code, then the developer should not forget to unregister the listener.

5. Unregistered  Receivers

Many times the developer needs to register the local broadcast receiver in an activity. However, if the developer does not unregisters the broadcast receiver there is a strong chance that our app can lead to the memory leak problem as the receiver will hold a strong reference to the activity. So even if the activity is of no use, it will prevent the garbage collector to collect the activity for garbage collection, which will ultimately cause the memory leak.

Kotlin




// sample kotlin program for broadcast receiver 
import android.app.Activity
import android.content.BroadcastReceiver
import android.content.Context
import android.content.Intent
import android.content.IntentFilter
import android.os.Bundle
import com.example.myapplication.R
  
class LocalBroadcastReceiverActivity : Activity() {
  
    private var localBroadcastReceiver: BroadcastReceiver? = null
  
    override fun onCreate(savedInstanceState: Bundle?) {
        super.onCreate(savedInstanceState)
        setContentView(R.layout.activity_main)
    }
  
    private fun registerBroadCastReceiver() {
        localBroadcastReceiver = object : BroadcastReceiver() {
            override fun onReceive(context: Context, intent: Intent) {
                // Write your code here
            }
        }
        registerReceiver(localBroadcastReceiver, IntentFilter("android.net.conn.CONNECTIVITY_CHANGE"))
    }
  
    override fun onStart() {
        super.onStart()
        // registering the broadcast receiver
        registerBroadCastReceiver()
    }
  
    override fun onStop() {
         super.onStop()
          
         // Broadcast receiver holds the implicit reference of Activity.
           // Therefore even if activity is destroy,
         // garbage collector will not be able to remove its instance.
         if (localBroadcastReceiver != null) {
            unregisterReceiver(localBroadcastReceiver)
        }
    }
}


6. Inner class Reference

An inner class is often used by the android developers within their code. However, the nonstatic class will hold the implicit reference of the parent class which can cause the memory leak problem. So there can be two solutions that can be proposed to solve this problem:

  • One can make the inner class static
  • If one wants to pass the reference of the non-static inner class, then it can be passed using weak reference.

Tools to Detect Memory Leak

As it is known that when something cannot be seen, it is very difficult to fix it, same is the case with the memory leak, it cannot be seen, so it is very difficult to fix. But there are some tools that help us to detect the memory leaks in the android application and helps us to fix it. Let’s see some of the most popular tools:

1. Leak Canary

Leak Canary is a memory detection library in Android. It is developed by a square cup company. This library has a unique ability to decrease down the memory leak and helping developers to get less “MemoryOutOfError”. Leak canary even helps us to notify where the leak is actually happening. To use Leak-Canary, add the following dependency in the build.gradle(app level file).

dependencies {

 // debugImplementation because LeakCanary should only run in debug builds.

 debugImplementation ‘com.squareup.leakcanary:leakcanary-android:2.4’

}

Once the leak canary is installed it automatically detects and reports memory leaks in 4 steps:

  • Detecting retained objects.
  • Dumping the heap.
  • Analyzing the heap.
  • Categorizing leaks.

If one wants to dig deeper and learn how to leak canary report memory leaks can refer to the official documentation of leak canary.

2. Android Profiler

It is basically a tool that helps to keep track of memory usage of every application in android. It replaced Android Monitor in the android version 3.0 and higher. The Android Profiler is compatible with Android 5.0 (API level 21) and higher. Android Profiler detects the performance of the application in the real-time on the parameters like:

  • Battery
  • Memory (In MB)
  • CPU Usage (In %)
  • Network Rate(Rate of uploading and receiving)

To open the android profiler within the Android project in android studio, perform the following steps: Choose View > Tool Windows > Profiler > Select deployment target and choose the device. The performance of the app will be listed as shown as in the image below: 

Image Showing App Usage Data Using  Andorid Profiler in various respects

To know more about how the android profiler works, refer to the official documentation of the android profiler.



Last Updated : 25 Sep, 2020
Like Article
Save Article
Previous
Next
Share your thoughts in the comments
Similar Reads