I actually tested both 2170 and 2179 and you’re right joe528, i cant seem to break the app anymore using version 2170… However, when i used 2179 it would crash when calling my upload function the second time or sometimes even calling the upload function once.
Corona Dev’s there is deff something off
@eja Do you tested it with actual Android device or Simulator? How about iOS device?
When the crash happens, what’s printed out in the console? (I cannot reproduce it. I only got Google Play crash report from end users which doesn’t show much)… wondering what’s printed out in the console?
I tested with an actual Android device (Galaxy S3)… I have also tested on IOS and everything seems fine. As far as console reports i get none, i do however get a message when it crashes that states “Unfortunately, the process com.(app package name) has stopped.”
So we both see a situation that there is no crash with build 2170 and there is a network-related crash in build 2179 for Android devices.
Corona may want to dig into this. It’s quite a severe bug and might lose track if you don’t fix it as soon as possible.
joe, did you report this bug officially yet or no?
Here are more updates about this bug:
(1) After I re-built my app with Corona 2170 and published (without any source code change in my part), it has been over 2 days, there is no single crash report.
(2) I still received 6 more crash reports from Google Play for those users who are still using my app built with Corona 2179. So the number of this crash report is 12+6=18 in total now. Although the number is not many, it’s still the greatest number for a single problem so far since my app is published.
All my users are hitting this all of a sudden when they use the app built with 2179, and no single crash like this if 2170 is used.
And don’t forget that my app built with 2179 was in published state for fewer than 10 hours. In such a short time frame, I got 18 crash reports for the same bug. It’s odd enough and unprecedented.
It’s just not a coincidence.
My app uses a lot of network requests to pull data from the Internet. So this bug might not happen to a normal app built with Corona but there is something wrong fundamentally in network handling that should be heeded immediately. This definitely hurts the stability of Corona. And if you want to reproduce this bug first, you may never be able to identify the bug & fix it.
Hi,
I have received the same crash reports as well.
The app was built with latest version 2179.
9 crash reports have been received within past few hours.
So far these 9 reports all came from Android 4.3.
Possible steps to reproduce the crash:
I am using Sony Xperia Z1 Android 4.3
Open your app
- do not do anything and wait until your device go into sleepmode/standby/autolock
wake up your phone and resume your app
game crash
run your app as normal
press home button after some minutes
game crash
Here is my build.settings
[lua]-- Supported values for orientation:
– portrait, portraitUpsideDown, landscapeLeft, landscapeRight
settings = {
orientation = {
default = “portrait”,
supported = { “portrait”, }
},
iphone = {
plist = {
UIStatusBarHidden = false,
UIPrerenderedIcon = true, – set to false for “shine” overlay
UIApplicationExitsOnSuspend = false, – uncomment to quit app on suspend
UIAppFonts =
{
“slkscr.ttf”,
},
FacebookAppID = “XXX”,
CFBundleURLTypes =
{
{
CFBundleURLSchemes =
{
“fbXXX”, – example scheme for facebook
}
}
}
}
},
android =
{
usesPermissions =
{
“android.permission.INTERNET”,
“android.permission.ACCESS_WIFI_STATE”,
“android.permission.READ_PHONE_STATE”,
“android.permission.ACCESS_NETWORK_STATE”,
“android.permission.WRITE_EXTERNAL_STORAGE”,
},
},
plugins =
{
[“facebook”] =
{
publisherId = “com.coronalabs”,
supportedPlatforms = { iphone = true },
},
[“CoronaProvider.native.popup.social”] =
{
– required
publisherId = “com.coronalabs”,
},
[“CoronaProvider.ads.admob”] =
{
– required
publisherId = “com.coronalabs”,
},
},
}
[/lua]
＠snagonstudio
Could you reiterate followings so Corona can start to believe something wrong between 2170~2179
(1) Could you post your crash report, is it exactly the same as what I posted in my first post?
(2) Did you receive this crash report before you use build 2179? Ever?
java.lang.NullPointerException
at network.NetworkRequest$AsyncNetworkRequestRunnable.run(NetworkRequest.java:1839)
at java.lang.Thread.run(Thread.java:841)
Last reported
Today, 3:34 PM
Reports this week
7
Reports total
7
Application version
1.0.67
Android version
Android 4.37
Device
Xperia Z1 (C6903)3
2.
Galaxy Note3 (hlte)2
3.
Xperia Z (C6603)1
4.
Galaxy Nexus (maguro)1
When I switch back to 2170 build. The crash is gone.
Hope that helps!
Wow, Corona
based on the information above,
I really don’t know what else you need to believe something is just off between build 2170~2179
I am going to create a sample app just so I can submit this bug right now so this can get resolved, this is not good.
Please try with 2182. We’ve fixed a case that could be causing it, and it should produce more useful information in the log files. If you can try with 2182 and let us know if it fixes it, or more likely the crash will stop but the network error will still be there and info from the logs will be most helpful in helping is make this solid.
Thanks
Rob
In response to both yes I was able to test using build 2183 and everything now seems fine. Thanks Rob
@eja Thanks for your testing and it’s a good news.
However, Rob said earlier: “but the network error will still be there and info from the logs will be most helpful in helping is make this solid.”
Do you see any “extra network error” even there is no more crash?
Is it possible to tell us a bit more what kind of logs should be expected? and in what condition this bug would be triggered? and what causes the bug and what has been fixed?
I will see if I can play around with the app to hit the conditions that may cause the bug.
I’m not sure. What I got from engineering what that the exception was in a block of code that was triggered on a network error. This causes the app to not crash any more, but it doesn’t fix the fact that you got an error from the server side. You need to make sure you’re trapping your network errors and handling those conditions.
Based on Rob last post, there is some extra log when the exception is caught.
Have you seen the “extra log output” while you are trying it?
It can help Corona to make the network handling more solid.
@joe528, I am sorry after this issue was no longer replicated on builds 2182+ I have moved on, but if its any help I have a retry network function in place and the crash was happening when the network would fail and not even run through my “retry” function. Everything is 100% fine now, no more crashes so I didn’t take the time to check the logs since it was fixed.