Search Issue Tracker
Fixed in 2019.3
Fixed in 2019.1, 2019.2
Votes
49
Found in
2019.1.4f1
2019.1.7f1
2019.2.0a1
2019.3.0a1
Issue ID
1163795
Regression
Yes
[iOS] UnityWebRequest test fails sending large data
To reproduce:
1. Open user attached project "iosWebRequestFailures.zip"
2. Open "SmapleScene.unity" scene
3. Enter Play mode
4. Observe in the Game window as all tests are passing ( it should cycle 5 times )
5. Build for iOS
6. Deploy the Xcode project to an iOS device
Expected result: all the test are passing as in the Editor
Actual result: tests start to fail almost instantly
Reproduced on:
2019.1.4f1 2019.1.8f1 2019.2.0b7 2019.3.0a7
Doesn't reproduce on:
2018.4.3f1 2019.1.3f1
Devices reproduced on:
- iPhone 8 Plus iOS 12.0
- iPhone XR iOS 12.0
Device not reproduced on:
- VLNQA00263, Google Pixel 3 (Pixel 3), Android 9, CPU: Snapdragon 845, GPU: Adreno (TM) 630
- macOS 10.14 machine
Notes:
- In 2019.3.0a7 on Editor tests fail on the last tests because of a timeout, changing line 32 in "NetworkTester.cs" script from WaitForSeconds(20.0f); to 30.0f all the tests pass
Add comment
All about bugs
View bugs we have successfully reproduced, and vote for the bugs you want to see fixed most urgently.
Latest issues
- MissingMethodException: Method not found: 'UnityEngine.UnityAPICompatibilityVersionAttribute..ctor
- [URP] Cannot assign overlay cameras to a camera stack while in Prefab Mode.
- "Go to previous Keyframe" button/shortcut does not work in the Animation Window
- Passing Scene names to LoadScene(Async)InPlayMode instead of full path loads empty Scenes
- [Linux] [Editor] Input.GetAxis("Mouse ScrollWheel") always returns 0 when Cursor.lockState is set to CursorLockMode.Locked
johndd110
Feb 13, 2020 11:23
If you set out to make me think today; mission accomplished! I really like your writing style and how you express your ideas. Thank you.
is it safe to travel to taiwan now https://www.islandlifetaiwan.com/taiwan-safe-to-visit-during-coronavirus-in-mainland-china/
johndd110
Feb 12, 2020 15:19
You make so many great points here that I read your article a couple of times. Your views are in accordance with my own for the most part. This is great content for your readers.
latest cryptocurrency news https://cryptorising.news
razirbel66
Jul 08, 2019 19:23
Another 'ThankYou' to @RCAMDUELL. We were stuck with send failures for large data and this at least keeps things working until a permanent solution happens.
andreyul
Jul 04, 2019 18:32
Delete the line
[self writeMoreBody];
from
- (void)URLSession:(NSURLSession *)session task:(NSURLSessionTask *)task didSendBodyData:...
in Classes/Unity/UnityWebRequest.mm
to fix this problem.
sier_ua
Jun 29, 2019 09:51
I'm experiencing the same issue on iOs devices. However I was able to temporarily work around the problem by using System.Net.Http.HttpClient which sends large amount of data from iOs devices without any issues.
nia_ru
Jun 28, 2019 09:49
Having the same problem with sending big `byte[] postData` using WWW on iOS platform. And even switching to the recommended UnityWebRequest does not help, so the problem is common to both. We too have reproduced steadily this on all released versions of 2019.1.* branch (.1.4 through .1.8) on iPad, and our customers have also the same problem on a wide variety of iOS devices (hopefully other problems are not affected).
This critically affects our app. Since the app is cross-platform, we currently have no idea how to address this issue without using external libraries that can limit our cross-platformness. Please pay as much attention as possible to the problem. Having this still not fixed after 8 patches is unbelievable!
vl_avan
Jun 28, 2019 07:19
We have the same issue when sending large (6K+) base-64 encoded tokens to our server. 40% of the data is corrupted. Please fix this bug as soon as possible.
tcvpromo
Jun 27, 2019 06:14
Fix it please! Thats a big problem for us
UnityDev_
Jun 25, 2019 16:11
@RCAMDUELL Thank you, you're a life saver!
rcamduell
Jun 25, 2019 16:08
I found that the method "- (void)writeMoreBody" in Classes/Unity/UnityWebRequest.mm is being called on the same instance from multiple threads when the issue happens. I added @synchronized (self) {<method body>} and it seems to have worked without any obvious performance issues. I assume this method was developed to be thread safe, but something broke that.