Our app is crashing with the fatal exception:
Fatal Exception: NSInternalInconsistencyException
Modifications to the layout engine must not be performed from a background thread after it has been accessed from the main thread.
Our app used to not throw this exception on previous versions of iOS (12 and below) and when I would compile on Xcode 10. Now, on iOS 13.3.1 and compiling on Xcode 11.3.1 our app crashes 10 seconds after opening every time. I looked into this exception and it seems Apple has decided to prevent apps from accessing the UIKit on background threads (please correct me if I'm wrong here). After enabling the Main Thread Checker I was able to pinpoint the line that was causing this crash. Backtracing the code led me to this line:
dispatch_async(dispatch_get_main_queue(), ^{
//Code that uses the StoreKit to retrieve purchase data,
//then updates the UI using UIKit based on this information
});
This code is fine though right? Because even though it's dispatching this block of code, it's doing it to the main thread, so this shouldn't be the issue, correct? Well, even after moving this block of code out of the dispatch_async() function, it still crashes. So then I looked at the singleton for this class, which looks like this:
+(instancetype)sharedPurchaseManager
{
static PurchaseManager * sharedPurchaseManager = nil;
static dispatch_once_t onceToken;
dispatch_once(&onceToken, ^{
sharedPurchaseManager = [[PurchaseManager alloc] _init];
});
return sharedPurchaseManager;
}
And I wondered if the _init() for the PurchaseManager was happening in a background thread. So I tried moving the allocation for PurchaseManager out of the dispatch_once() block and the app stopped crashing. Furthermore, the Main Thread Checker stopped interrupting the program at the line that uses UIKit. I'm pretty sure that moving my _init() code out of the dispatch_once() block is not the correct way to go about fixing my issue, as this is the standard singleton design pattern (please correct me if I'm wrong). So then how should I go about fixing this issue? Furthermore, we have about 10+ crashes in our code that is due to the same exception, and with a similar design pattern to our PurchaseManager, so we would want a solution for all of these exceptions.
It sounds like [[PurchaseManager alloc] _init] contains some UI code, but +(instancetype)sharedPurchaseManager is accessed first time from background thread.
So you can try two approaches:
1) find the way explicitly initialize sharedPurchaseManger on the main thread (eg. in applicationDidFinishLaunching)
2) try to use the following
dispatch_once(&onceToken, ^{
dispatch_sync(dispatch_get_main_queue(), ^{ // << intentionally synchronous!
sharedPurchaseManager = [[PurchaseManager alloc] _init];
}
});
In hockey app dashboard, I'm getting crashes with following hints:
Exception Type: SIGTRAP
Exception Codes: #0 at 0x1943f61e8
Crashed Thread: 7
Application Specific Information:
Selector name found in current argument registers: retain
Exception Type: SIGSEGV
Exception Codes: SEGV_ACCERR at 0x568855a90
Crashed Thread: 18
Application Specific Information:
objc_msgSend() selector name: retain
Both crash groups are hitting same code but somehow they had different crash description which I believe should be same root cause.
This is what my code looks like:
thread 1:
dispatch_queue_t queue;
#synchronized (self) {
queue = _mySerialQueue;
}
dispatch_async(queue, ^{ // Crash happens here
if (_ivar) {
...
}
});
thread 2:
#synchronized (self) {
_mySerialQueue = dispatch_queue_create(...);
}
Can this code run into ARC issue?
On the premise that (please verify these)
_mySerialQueue is an instance variable
_mySerialQueue is only accessed within #synchronized(self) blocks
You don't access _mySerialQueue via any property or via key value coding ([theObject valueForKey:"mySerialQueue"])
All code you posted runs with ARC
I see only one reason why the above code could crash the way it does. I'm still not 100% sure about that, but here is my shot:
self is in the process of being deallocated when the crash happens. It can be checked if you clean the iVar in dealloc:
- (void)dealloc {
#synchronized(self) {
_mySerialQueue = nil;
}
// ...
}
these crashes would be extremely rare, e.g. one in a hundred at least. If it happens more often, I think the problem is somewhere else.
After the change, the app would still crash, but the EXC_BAD_ACCESS would be at a lower memory location because you are dereferencing a NULL pointer when you call dispatch_async(nil, ...). The crash location will most likely be 0x00000050, or a similarily low address.
Why do I think this?
When an object is being destructed, the iVars are destroyed in a separate pass by a method named .cxx_destruct. When this happens, the iVar can be in the process of being deallocated, but at the same time the variable still points to the object because it is never cleared.
Important: If this works, it will just replace one rare crash with another rare crash, because it would mean that there is a race condition on self somewhere.
Another possible solution would be to try to reproduce the problem with the thread sanitizer. It can track down such problems pretty fast in most cases.
I'm using PLCrashReporter in my iOS project and I'm curious, is it possible to use Core Foundation code in my custom crash callback. The thing, that handle my needs is CFPreferences.Here is part of code, that I create:
void LMCrashCallback(siginfo_t* info, ucontext_t* uap, void* context) {
CFStringRef networkStatusOnCrash;
networkStatusOnCrash = (CFStringRef)CFPreferencesCopyAppValue(networkStatusKey, kCFPreferencesCurrentApplication);
CFStringRef additionalInfo = CFStringCreateWithFormat(
NULL, NULL, CFSTR( "Additional Crash Properties:[Internet: %#]", networkStatusOnCrash);
CFPreferencesSetAppValue(additionalInfoKey, additionalInfo,
kCFPreferencesCurrentApplication);
CFPreferencesAppSynchronize(kCFPreferencesCurrentApplication);
}
My target is to collect some system information, just in time when app crashed, e.g Internet connection type.
I know it is not good idea to create own crash callback due to async-safe functions, but this can help.
Also as other option: Is there a way to extend somehow PLCrashReportSystemInfo class?
This is very dangerous. In particular the call to CFStringCreateWithFormat allocates memory. Allocating memory in the middle of a crash handler can lead to battery-draining deadlock (yep; had that bug…) For example, if you were in the middle of free() (which is not an uncommon place to crash), you may already be holding a spinlock on the heap. When you call malloc to get some memory, you may spinlock the heap again and deadlock in a tight-loop. The heap needs to be locked so often and for such short periods of time that it doesn't use a blocking lock. It does the equivalent of while (locked) {}.
You seem to just be reading a preference and copying it to another preference. There's no reason to do that inside a crash handler. Just check hasPendingCrashReport during startup (which I assume you're doing already), and read the key then. It's not clear what networkStatusKey is, but it should still be there when you start up again.
If for any reason it's modified very early (before you call hasPendingCrashReport), you can grab it in main() before launching the app. Or you can grab it in a +load method, which is called even earlier.
When deploying the application to the device, the program will quit after a few cycles with the following error:
Program received signal: "EXC_BAD_ACCESS".
The program runs without any issue on the iPhone simulator, it will also debug and run as long as I step through the instructions one at a time. As soon as I let it run again, I will hit the EXC_BAD_ACCESS signal.
In this particular case, it happened to be an error in the accelerometer code. It would not execute within the simulator, which is why it did not throw any errors. However, it would execute once deployed to the device.
Most of the answers to this question deal with the general EXC_BAD_ACCESS error, so I will leave this open as a catch-all for the dreaded Bad Access error.
EXC_BAD_ACCESS is typically thrown as the result of an illegal memory access. You can find more information in the answers below.
Have you encountered the EXC_BAD_ACCESS signal before, and how did you deal with it?
From your description I suspect the most likely explanation is that you have some error in your memory management. You said you've been working on iPhone development for a few weeks, but not whether you are experienced with Objective C in general. If you've come from another background it can take a little while before you really internalise the memory management rules - unless you make a big point of it.
Remember, anything you get from an allocation function (usually the static alloc method, but there are a few others), or a copy method, you own the memory too and must release it when you are done.
But if you get something back from just about anything else including factory methods (e.g. [NSString stringWithFormat]) then you'll have an autorelease reference, which means it could be released at some time in the future by other code - so it is vital that if you need to keep it around beyond the immediate function that you retain it. If you don't, the memory may remain allocated while you are using it, or be released but coincidentally still valid, during your emulator testing, but is more likely to be released and show up as bad access errors when running on the device.
The best way to track these things down, and a good idea anyway (even if there are no apparent problems) is to run the app in the Instruments tool, especially with the Leaks option.
A major cause of EXC_BAD_ACCESS is from trying to access released objects.
To find out how to troubleshoot this, read this document:
DebuggingAutoReleasePool
Even if you don't think you are "releasing auto-released objects", this will apply to you.
This method works extremely well. I use it all the time with great success!!
In summary, this explains how to use Cocoa's NSZombie debugging class and the command line "malloc_history" tool to find exactly what released object has been accessed in your code.
Sidenote:
Running Instruments and checking for leaks will not help troubleshoot EXC_BAD_ACCESS. I'm pretty sure memory leaks have nothing to do with EXC_BAD_ACCESS. The definition of a leak is an object that you no longer have access to, and you therefore cannot call it.
UPDATE:
I now use Instruments to debug Leaks. From Xcode 4.2, choose Product->Profile and when Instruments launches, choose "Zombies".
An EXC_BAD_ACCESS signal is the result of passing an invalid pointer to a system call. I got one just earlier today with a test program on OS X - I was passing an uninitialized variable to pthread_join(), which was due to an earlier typo.
I'm not familiar with iPhone development, but you should double-check all your buffer pointers that you're passing to system calls. Crank up your compiler's warning level all the way (with gcc, use the -Wall and -Wextra options). Enable as many diagnostics on the simulator/debugger as possible.
In my experience, this is generally caused by an illegal memory access. Check all pointers, especially object pointers, to make sure they're initialized. Make sure your MainWindow.xib file, if you're using one, is set up properly, with all the necessary connections.
If none of that on-paper checking turns anything up, and it doesn't happen when single-stepping, try to locate the error with NSLog() statements: sprinkle your code with them, moving them around until you isolate the line that's causing the error. Then set a breakpoint on that line and run your program. When you hit the breakpoint, examine all the variables, and the objects in them, to see if anything doesn't look like you expect.I'd especially keep an eye out for variables whose object class is something you didn't expect. If a variable is supposed to contain a UIWindow but it has an NSNotification in it instead, the same underlying code error could be manifesting itself in a different way when the debugger isn't in operation.
I just spent a couple hours tracking an EXC_BAD_ACCESS and found NSZombies and other env vars didn't seem to tell me anything.
For me, it was a stupid NSLog statement with format specifiers but no args passed.
NSLog(#"Some silly log message %#-%#");
Fixed by
NSLog(#"Some silly log message %#-%#", someObj1, someObj2);
The 2010 WWDC videos are available to any participants in the apple developer program.
There's a great video: "Session 311 - Advanced Memory Analysis with Instruments" that shows some examples of using zombies in instruments and debugging other memory problems.
For a link to the login page click HERE.
Not a complete answer, but one specific situation where I've received this is when trying to access an object that 'died' because I tried to use autorelease:
netObjectDefinedInMyHeader = [[[MyNetObject alloc] init] autorelease];
So for example, I was actually passing this as an object to 'notify' (registered it as a listener, observer, whatever idiom you like) but it had already died once the notification was sent and I'd get the EXC_BAD_ACCESS. Changing it to [[MyNetObject alloc] init] and releasing it later as appropriate solved the error.
Another reason this may happen is for example if you pass in an object and try to store it:
myObjectDefinedInHeader = aParameterObjectPassedIn;
Later when trying to access myObjectDefinedInHeader you may get into trouble. Using:
myObjectDefinedInHeader = [aParameterObjectPassedIn retain];
may be what you need. Of course these are just a couple of examples of what I've ran into and there are other reasons, but these can prove elusive so I mention them. Good luck!
I find it useful to set a breakpoint on objc_exception_throw. That way the debugger should break when you get the EXC_BAD_ACCESS.
Instructions can be found here DebuggingTechniques
Just to add another situation where this can happen:
I had the code:
NSMutableString *string;
[string appendWithFormat:#"foo"];
Obviously I had forgotten to allocate memory for the string:
NSMutableString *string = [[NSMutableString alloc] init];
[string appendWithFormat:#"foo"];
fixes the problem.
Another method for catching EXC_BAD_ACCESS exceptions before they happen is the static analyzer, in XCode 4+.
Run the static analyzer with Product > Analyze (shift+cmd+B).
Clicking on any messages generated by the analyzer will overlay a diagram on your source showing the sequence of retains/releases of the offending object.
Use the simple rule of "if you didn't allocate it or retain it, don't release it".
Run the application and after it fails (Should display "Interrupted" rather than "EXC_BAD_ACCESS"... Check the Console (Run > Console)... There should be a message there now telling what object it was trying to access.
I've been debuging, and refactoring code to solve this error for the last four hours. A post above led me to see the problem:
Property before:
startPoint = [[DataPoint alloc] init] ;
startPoint= [DataPointList objectAtIndex: 0];
.
.
.
x = startPoint.x - 10; // EXC_BAD_ACCESS
Property after:
startPoint = [[DataPoint alloc] init] ;
startPoint = [[DataPointList objectAtIndex: 0] retain];
Goodbye EXC_BAD_ACCESS
Hope you're releasing the 'string' when you're done!
I forgot to return self in an init-Method... ;)
This is an excellent thread. Here's my experience: I messed up with the retain/assign keyword on a property declaration. I said:
#property (nonatomic, assign) IBOutlet UISegmentedControl *choicesControl;
#property (nonatomic, assign) IBOutlet UISwitch *africaSwitch;
#property (nonatomic, assign) IBOutlet UISwitch *asiaSwitch;
where I should have said
#property (nonatomic, retain) IBOutlet UISegmentedControl *choicesControl;
#property (nonatomic, retain) IBOutlet UISwitch *africaSwitch;
#property (nonatomic, retain) IBOutlet UISwitch *asiaSwitch;
I encountered EXC_BAD_ACCESS on the iPhone only while trying to execute a C method that included a big array. The simulator was able to give me enough memory to run the code, but not the device (the array was a million characters, so it was a tad excessive!).
The EXC_BAD_ACCESS occurred just after entry point of the method, and had me confused for quite a while because it was nowhere near the array declaration.
Perhaps someone else might benefit from my couple of hours of hair-pulling.
Forgot to take out a non-alloc'd pointer from dealloc. I was getting the exc_bad_access on my rootView of a UINavigationController, but only sometimes. I assumed the problem was in the rootView because it was crashing halfway through its viewDidAppear{}. It turned out to only happen after I popped the view with the bad dealloc{} release, and that was it!
"EXC_BAD_ACCESS" [Switching to process 330] No memory available to program now: unsafe to call malloc
I thought it was a problem where I was trying to alloc... not where I was trying to release a non-alloc, D'oh!
How i deal with EXC_BAD_ACCESS
Sometimes i feel that when a EXC_BAD_ACCESS error is thrown xcode will show the error in the main.m class giving no extra information of where the crash happens(Sometimes).
In those times we can set a Exceptional Breakpoint in Xcode so that when exception is caught a breakpoint will be placed and will directly intimate the user that crash has happened in that line
NSAssert() calls to validate method parameters is pretty handy for tracking down and avoiding passing nils as well.
I just had this problem. For me the reason was deleting a CoreData managed object ans trying to read it afterwards from another place.
I've been debuging, and refactoring code to solve this error for the last four hours. A post above led me to see the problem:
Property before:
startPoint = [[DataPoint alloc] init] ;
startPoint= [DataPointList objectAtIndex: 0];
x = startPoint.x - 10; // EXC_BAD_ACCESS
Property after:
startPoint = [[DataPoint alloc] init] ;
startPoint = [[DataPointList objectAtIndex: 0] retain];
Goodbye EXC_BAD_ACCESS
Thank you so much for your answer. I've been struggling with this problem all day. You're awesome!
Just to add
Lynda.com has a fantastic DVD called
iPhone SDK Essential Training
and Chapter 6, Lesson 3 is all about EXEC_BAD_ACCESS and working with Zombies.
It was great for me to understand, not just the error code but how can I use Zombies to get more info on the released object.
To check what the error might be
Use NSZombieEnabled.
To activate the NSZombieEnabled facility in your application:
Choose Project > Edit Active Executable to open the executable Info window.
Click Arguments.
Click the add (+) button in the “Variables to be set in the environment” section.
Enter NSZombieEnabled in the Name column and YES in the Value column.
Make sure that the checkmark for the NSZombieEnabled entry is selected.
I found this answer on iPhoneSDK
I realize this was asked some time ago, but after reading this thread, I found the solution for XCode 4.2:
Product -> Edit Scheme -> Diagnostics Tab -> Enable Zombie Objects
Helped me find a message being sent to a deallocated object.
When you have infinite recursion, I think you can also have this error. This was a case for me.
Even another possibility: using blocks in queues, it might easily happen that you try to access an object in another queue, that has already been de-allocated at this time. Typically when you try to send something to the GUI.
If your exception breakpoint is being set at a strange place, then this might be the cause.
I got it because I wasn't using[self performSegueWithIdentifier:sender:] and -(void) prepareForSegue:(UIstoryboardSegue *) right
Don't forget the # symbol when creating strings, treating C-strings as NSStrings will cause EXC_BAD_ACCESS.
Use this:
#"Some String"
Rather than this:
"Some String"
PS - typically when populating contents of an array with lots of records.
XCode 4 and above, it has been made really simple with Instruments. Just run Zombies in Instruments. This tutorial explains it well: debugging exc_bad_access error xcode instruments