Firebase-ios-sdk: Crash app [GDTUploadCoordinator packageExpired:]

[zhuravel-sergey]

• Xcode version: 10.3
• Firebase SDK version: 6.6.0
• Firebase Component:
  Using FirebaseABTesting (3.0.0)
  Using FirebaseAnalytics (6.1.0)
  Using FirebaseCore (6.2.0)
  Using FirebaseCoreDiagnostics (1.0.1)
  Using FirebaseCoreDiagnosticsInterop (1.0.0)
  Using FirebaseInstanceID (4.2.3)
  Using FirebaseRemoteConfig (4.2.1)
  Using Google-Mobile-Ads-SDK (7.48.0)
  Using GoogleAppMeasurement (6.1.0)
  Using GoogleDataTransport (1.1.0)
  Using GoogleDataTransportCCTSupport (1.0.1)
  Using GoogleUtilities (6.2.4)

[REQUIRED] Step 3: Describe the problem

Some users have a crash on:
GDTUploadCoordinator.m line 239
-[GDTUploadCoordinator packageExpired:]

Only about ~20% of users have this problem, and on all devices and on all versions of iOS

In the attachment, a log file with Crashlytics

com.shmatchenko.vpm-locker_issue_crash_444193d2ab66446e921b3d63a3caa5e2_DNE_0_v2.txt

[google-oss-bot]

I found a few problems with this issue:

• I couldn't figure out how to label this issue, so I've labeled it for a human to triage. Hang tight.
• This issue does not seem to follow the issue template. Make sure you provide all the required information.

[paulb777]

Thanks for the report. We'll investigate.

In the meantime, a workaround would be to revert to Firebase 6.5.0.

[mikehaney24]

Looking into it.

[zhuravel-sergey]

Guys, there is such a problem on my second application too :(
com.lock.master_issue_crash_dcee41f2caad42f0a3539351ea108052_DNE_0_v2.txt

[mrgrauel]

Reverting to 6.5.0 doesn't help cause there is the FIRInstanceID Crash.
https://github.com/firebase/firebase-ios-sdk/issues/3439

The last versions of Firebase aren't as stable as before. Each release has a few problems.

[mikehaney24]

If you're also running into this issue, please post any diagnostics information you have.

[stsandro]

This might be connected to this issue, although I'm using an older version of GoogleDataTransport:

• Firebase/Core (6.6.0):
  
  
  
  • GoogleDataTransport (1.0.0)
  
  
  • GoogleDataTransportCCTSupport (1.0.0)
  
  

com.mycompany.myapp_issue_crash_e3be9ca665f245fcb8e8d9486459586a_DNE_0_v2.txt

[mikehaney24]

Thanks for posting that--is anyone able to reproduce this locally (and willing to make a project that demonstrates the issue)? Or is it just showing up in crash reports?

In either case, I'll add checks to make this bit of code unlikely to crash, but I'd like to be able to pinpoint the cause.

[iwheelbuy]

Or is it just showing up in crash reports?

Just crash reports...

[zhuravel-sergey]

Or is it just showing up in crash reports?

Yes, I can’t reproduce locally, because not all users have this problem, in my applications only about ~20%

[iwheelbuy]

Wow. 20% is a lot. I have only 0.1% of crashed sessions, which is 1 crash for 1000 app launches.

[zhuravel-sergey]

I have about ~2000/day new installations and ~2000 daily users, my number of crashes in the screenshot. This is only one application, on others, the situation is about the same. :(

[flocbit]

Experiencing this issue as well (except that the crash happens on line 224). Fortunately, we have not released anything yet but the error has already affected two of our 100 beta testers. Let me where I can send the crash logs to...

[mikehaney24]

If you're fine posting them here, please do so. If you prefer to keep them private, email them to me at lowlevel@ google dot com

[flocbit]

I have emailed them to you.

[iDevid]

Same here.
"""
Crashed: com.apple.main-thread
0 libdispatch.dylib 0x2151035d0 dispatch_async$VARIANT$mp + 160
1 AppName 0x1012967e4 -[GDTUploadCoordinator packageExpired:] + 236 (GDTUploadCoordinator.m:236)
2 Foundation 0x2161a48a4 __NSFireTimer + 84
3 CoreFoundation 0x2156af650 __CFRUNLOOP_IS_CALLING_OUT_TO_A_TIMER_CALLBACK_FUNCTION__ + 28
4 CoreFoundation 0x2156af380 __CFRunLoopDoTimer + 864
5 CoreFoundation 0x2156aebb4 __CFRunLoopDoTimers + 248
6 CoreFoundation 0x2156a9b04 __CFRunLoopRun + 1844
7 CoreFoundation 0x2156a90b0 CFRunLoopRunSpecific + 436
8 GraphicsServices 0x2178a979c GSEventRunModal + 104
9 UIKitCore 0x241edf978 UIApplicationMain + 212
10 AppName 0x100aa9140 main + 21 (AppDelegate.swift:21)
11 libdyld.dylib 0x21516e8e0 start + 4
"""

[mikehaney24]

This should be fixed by #3555, I'm working on creating a new release that contains the fix.

[leoabortolotti]

Experiencing the same issue here...

[mikehaney24]

We've released GoogleDataTransport v1.1.1, which _should_ fix this issue. Can someone update their pods and let me know if the issue is fixed?

[leoabortolotti]

Hi @mikehaney24 I updated my pods... How can I test if it fixed the issues?

[mikehaney24]

Unless you're able to reproduce it locally, you'd have to ship a new version of your app with the updated pods and monitor crash reports on your new app version.

[ptsochantaris]

Seeing this too, it’s quite worrying. We’ve updated pods now to include the new GDT 1.1.1 pod and have queued a hotfix for review and release. Will advise on progress once that’s out.

[flocbit]

We'll implement 1.1.1 in our next beta release and will let you know if the issue still appears, thanks for the quick update!

[subhojitGoqii]

I tried pod repo update and then install pod but it still shows the installed version of GTD as 1.1.0. How to get the GTD v1.1.1 ?

[ptsochantaris]

@subhojitGoqii pod install will only install the currently selected list of pods from your Podfile.lock

You need to run pod update in order for your local pods to be updated (be careful, this may update other pods too, not just this one, so you may want to restrict that with options to just update this pod, if you don't want the others updated).

The pod repo update command you ran will only update the pods your local installation knows about. When you run pod update it will compare your installed pods to that registry and update them if needed.

[subhojitGoqii]

Thanks @ptsochantaris that helped. Thanks for the clarity about update and repo update too.

[ptsochantaris]

Rollout is quite early and numbers are very limited, but yesterday at around the equivalent user/sessions level we had seen a few instances of this crash, while now there are none, so that's encouraging. Will update again once we have larger numbers.

[stsandro]

I've also released an update to my app yesterday evening (around 9:00 PM CEST). As far as I can tell now this crash seems fixed.

[leoabortolotti]

I've updated my app yesterday too, and by now, no crashes. It seems fixed. Thanks!

[ptsochantaris]

Just as a followup, our user/session numbers are definitely higher and still no crash, at this point I'd say it's fixed for us too 👍

[mikehaney24]

That's great to hear! Glad we took care of this! Thanks everyone for being so communicative and following up!

[mars-lan]

Start seeing a different kind of crash after the update. Suspect it's related to GoogleDataTransport v1.1.1

Crashed: com.google.GDTStorage
EXC_BREAKPOINT 0x000000018b77a734
-[GDTStorage encodeWithCoder:]

[paulb777]

@mars-lan Thanks for the report. Please open another issue with the full backtrace.

[gauravhomedepot]

0

Crashed: com.apple.main-thread
EXC_BAD_ACCESS KERN_INVALID_ADDRESS 0x0000000000000054

GDTUploadCoordinator.m line 239
-[GDTUploadCoordinator packageExpired:] + 239