ANR Application Not Responding

ANR Full name Application Not Responding

One 、ANR Cause of occurrence .

    Only when the application UI Thread response timeout will cause ANR, The causes of timeout are generally 2 Kind of .
    ● The current event has no chance to be handled
    ● The current event is being processed , But it took too long to finish in time

Two 、ANR The classification of （ Three ）.

    1.KeyDispatchTimeout: The most common type , as a result of View The key event or touch event of 5 No response in seconds .
    2.BroadcastTimout: The reason is that the broadcast receiver (BrocastReceiver) Of onReceive() Function at a specific time (10 second ) Unable to complete processing .
    3.ServiceTimeout: The reason is service (Service) At a specific time (20 second ) Processing cannot be completed within .

 a.View Click events or touch events at a specific time （5s） Unable to get a response within 

 b. The main thread is executing BroadcastReceiver Of onReceive() Function time 10 Not processed in seconds 

 c. The main thread is in Service Each life cycle function of 20 Not processed in seconds .

3、 ... and 、 Typical ANR Problem scenario .

    1. Applications UI Threads have time-consuming operations , For example, in UI Network request in thread , Database operation or file operation, etc , May lead to UI Threads cannot process user input in time .
    2. Applications UI A thread waits for a child thread to release a lock , Thus, the input requested by the user cannot be processed .
    3. Time consuming animation requires a lot of computational work , May lead to CPU Too much load .

Four 、ANR Positioning and analysis of .

    1.LogCat Log information .
    2. Inside the cell phone anr file ( be located /data/anr/). for example anr_2022-06-27-12-47-52-079

----- pid 8418 at 2022-06-27 12:47:52 ----- pid  When does it appear anr
Cmd line: com.example.testdemo               Corresponding   Package name 


"main" prio=5 tid=1 Sleeping
  | group="main" sCount=1 dsCount=0 flags=1 obj=0x71b2a1f0 self=0xf1e5ce00
  | sysTid=8418 nice=-10 cgrp=default sched=0/0 handle=0xf238bdc0
  | state=S schedstat=( 1176699784 74358829 585 ) utm=110 stm=6 core=3 HZ=100
  | stack=0xff2d2000-0xff2d4000 stackSize=8192KB
  | held mutexes=
  at java.lang.Thread.sleep(Native method)
  - sleeping on <0x0d9beb8e> (a java.lang.Object)
  at java.lang.Thread.sleep(Thread.java:440)
  - locked <0x0d9beb8e> (a java.lang.Object)
  at java.lang.Thread.sleep(Thread.java:356)
  at com.example.testdemo.MainActivity$1.onClick(MainActivity.java:19)
  at android.view.View.performClick(View.java:7140)
  at com.google.android.material.button.MaterialButton.performClick(MaterialButton.java:1194)
  at android.view.View.performClickInternal(View.java:7117)
  at android.view.View.onKeyUp(View.java:14165)
  at android.widget.TextView.onKeyUp(TextView.java:8543)
  at android.view.KeyEvent.dispatch(KeyEvent.java:2825)
  at android.view.View.dispatchKeyEvent(View.java:13374)
  at android.view.ViewGroup.dispatchKeyEvent(ViewGroup.java:1922)
  at android.view.ViewGroup.dispatchKeyEvent(ViewGroup.java:1922)
  at android.view.ViewGroup.dispatchKeyEvent(ViewGroup.java:1922)
  ... repeated 2 times

"main" prio=5 tid=1 Sleeping
Represent the thread name, thread Priority, DVM thread id, DVM thread status
"main" :main thread -> activity thread
prio :java thread priority default is 5, （ The normal area is 1-10）
tid: yes DVM thread id, No linux thread id（ Next line sysTid It's just ）
Native:DVM thread Status These states are normal （ZOMBIE, RUNNABLE, TIMED_WAIT, MONITOR, WAIT, INITALIZING,STARTING, NATIVE, VMWAIT, SUSPENDED,UNKNOWN）

group="main" sCount=1 dsCount=0 flags=1 obj=0x416eaf18 self=0x416d8650
representative DVM thread status.
group: Is the thread group of the thread default is “main”
sCount: The number of times the thread is normally suspended 1 (thread suspend count)
dsCount: Number of times the thread was suspended due to debugging 0 (thread dbg suspend count)
obj: Associated with the current thread java Thread object 0x75720fb8 (thread obj address)
sef: The address of the thread itself 0x7f7e8af800 (thread point address)

sysTid=30307 nice=0 sched=0/0 cgrp=apps handle=1074565528
representative Linux thread status Display thread scheduling information
sysTId: linux Local threads under the system id linux thread tid
Nice: Thread scheduling has priority linux thread nice value
cgrp: Priority group genus c group
sched: Scheduling strategy cgroup policy/gourp id
handle: Handle function address handle address

state=S schedstat=( 0 0 0 ) utm=5 stm=4 core=3
representative CPU Sched stat Show more of the thread's current context
state: Scheduling status process/thread state （ Normal yes "R (running)", "S (sleeping)", "D (disk sleep)", "T (stopped)", "t (tracing stop)", "Z (zombie)", "X (dead)", "x (dead)", "K (wakekill)", "W (waking)",）, Usually average Process The state of being is S (sleeping), And if you find yourself in such a D (disk sleep), T (stopped), Z (zombie) It will be examined carefully after a while .
schedstat (Run CPU Clock/ns, Wait CPU Clock/ns, Slice times) The thread running information
utm: utime, user space time The time value used in thread user mode ( The unit is jiffies）
stm: stime, kernel space time The scheduling time value in kernel mode
core: now running in cpu. Finally, run the thread change cup identification

stack=0x7f7dc93000-0x7f7dc95000 stackSize=1020KB
Represents the stack address area and size

held mutexes=
Represents whether it is locked , Normal has four properties (mutexes: tll=0 tsl=0 tscl=0 ghl=0),0 Express unlock, All other values represent being lock,
tll: thread List Lock,
tsl: thread Suspend Lock,
tscl: thread Suspend Count Lock
ghl: gc Heap Lock

The rest is just some Call Stack

5、 ... and 、ANR To deal with

Three different situations , The general treatment is as follows
1. Main thread blocking

Open up separate sub threads to handle time-consuming blocking transactions .

2.CPU Full load , I/O Blocking

I/O Generally speaking, blocking means that file read / write or database operations are executed in the main thread , It can also be executed asynchronously by opening up sub threads .

3. Out of memory

increase VM Memory , Use largeHeap attribute , Check for memory leaks .

6、 ... and 、ANR Detection of

Use StrictMode

Strict mode StrictMode yes Android SDK A tool class provided to detect whether there are illegal operations in the code ,StrictMode It mainly detects two types of problems .

• Thread strategy ThreadPolicy
  • detectCustomSlowCalls： Detect custom time-consuming actions
  • detectDiskReads： Detect if there is a disk read operation
  • detectDiskWrites： Check whether there is a disk write operation
  • detectNetWork： Detect whether there is network operation
• Virtual machine policy VmPolicy
  • detectActivityLeaks： Detect the presence of Activity Let the cat out of the
  • detectLeakedClosableObjects： Check whether there are unclosed Closeable Object leakage
  • detectLeakedSqlLiteObjects： Detect the presence of Sqlite Object leakage
  • setClassInstanceLimit： Check whether the number of class instances exceeds the limit

You can see ,ThreadPolicy It can be used to detect possible mainline time-consuming operations , It should be noted that we can only Debug Use it in version , The version released to the market should be closed .StrictMode It's easy to use , We only need to initialize the application, for example Application perhaps MainActivity Class onCreate Method ：

StrictMode.setThreadPolicy(new StrictMode.ThreadPolicy.Builder()
                .detectAll().penaltyLog().penaltyDialog().build());
        
StrictMode.setVmPolicy(new StrictMode.VmPolicy.Builder().detectAll()
.penaltyLog().build());

penaltyLog It means that Logcat Print log in ,
detectAll Method means to start all detection policies 

BlockCanary

BlockCanary Is a non intrusive performance monitoring function library , Its usage and leakCanary similar , But the latter monitors memory leaks in applications , and BlockCanary It is mainly used to monitor the application main thread . Its basic principle is to use the message queue processing mechanism of the main thread , Determine whether the set time is exceeded by comparing the start and end time points of message distribution , If it is , It is judged that the main route is stuck . Its integration is simple

1. stay build.gradle Introduce dependency in

implementation 'com.github.markzhai:blockcanary-android:1.5.0'

2. stay Application Class

BlockCanary.install(this,new MyBlockCanaryContext()).start();