Most recent event fired is never flushed from Appboy cache

[appboy]

Most recent event fired is never flushed from Appboy cache

[briancaw]

Hi @manderson-productions

Thanks very much for reporting this. We've been unable to reproduce this so far, but we're actively investigating and will continue to do so.

Would it be possible for you to share that sample app with us? Also, if you could reach out to me at [email protected] we'd like to inquire more about your setup/environment and we can avoid publishing any sensitive info if we chat over email.

Thanks,
Brian

[manderson-productions]

Great, thanks Brian. I just sent you an email.

Mark Anderson
Software Engineer

On Tue, Apr 26, 2016 at 10:57 AM, briancaw [email protected] wrote:

Hi @manderson-productions https://github.com/manderson-productions

Thanks very much for reporting this. We've been unable to reproduce this
so far, but we're actively investigating and will continue to do so.

Would it be possible for you to share that sample app with us? Also, if
you could reach out to me at [email protected] we'd like to inquire more
about your setup/environment and we can avoid publishing any sensitive info
if we chat over email.

Thanks,
Brian

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#64 (comment)

[billmag]

Hey Mark-

Thanks for reporting this. We've identified the issue as triggering with a combination of a race condition related to a network flush and a force quit (or crash) which happens when you hit "Stop" in XCode.

The exact situation that would need to occur is after a regular network flush happens, the app would need to crash or be force closed before anything else is recorded by the SDK (thus any of these things would keep it from happening: the user fires another event or purchase, an update is made to a user attribute, a location ping is recorded, your customer locks/sleeps their phone, or your customer closes the app in the normal fashion by hitting the home button).

As an example, in your steps to repro, if you were to hit Analytic TWO after the network flush, but before force closing the app, you will not see the issue (we confirmed this in the sample app you sent over). Similarly, if you hit the home button (command-shift-h in the simulator) at any time, you will also not see the issue.

Brian will update you soon with our plan to release an update to fix this issue. We were planning on releasing a minor version bump before the end of the week with a couple of other issues fixed, so we can bundle this in and move the schedule forward to get it out promptly.

-Bill

[briancaw]

Hey @manderson-productions,

We will be releasing tonight with this update. Please let us know if you have any further comments/questions; we'll ping again when we've released.

Thanks,
Brian

[manderson-productions]

@briancaw Great thanks for the speedy fix! Will do some testing later this week once we update and let you know how everything is going.

[Wenzhi]

Hi @manderson-productions,

The new SDK 2.19.3 with the fix for this issue is released. I'm going to close the issue for now. Please feel free to re-open this issue if you still see it happen in the new SDK.

Thanks,
Wenzhi

[manderson-productions]

@Wenzhi @briancaw @billmag We tested the updated sdk and so far everything is looking good. Thanks for the fix!

[billmag]

Awesome, thanks for the follow up (and finding the issue in the first place)!