• After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.

Help Major problem with Android System Webview"

markdoc

markdoc

Android Expert
It has been posted in a number of forums about this app. If your phone started having problems with apps closing for unknown reasons and they were working fine before, most likely it is caused by Google's "Android System WebView". Usually, no error reports are generated either. To fix the problem, you need to uninstall all the updates and then unchecked autoupdate on Google Play. You can read the details at http://www.asus-zenfone.com/2015/04/android-system-webview-may-cause-apps-crash.html. I have tried to post this twice on Google Play, but it was deleted. Props to the original poster in my forum, Spoiter. Any thoughts would be welcome. I have an HTC One M8 with Android 5.0.1. (I should have added this before. Thanks PatticakeUS)
 
Last edited:
Most of the posts I've seen on this forum & another about this issue, are by HTC M8 owners. I don't have this issue on my LG G3. OP what phone do you have?
 
I have an HTC One M8 also, but the original web post was about Asus. It didn't mention HTC at all. Check it out. I just reported it to Google Apps. We'll see what happens.
 
I wasn't aware Asus even makes Android phones (but there's a lot I don't know).

As for the M8's & any other affected phones, I hope they fix that problem soon.
 
Last edited:
Yep, I have Lollipop 5.0.1 (Sprint). I've had no errors, crashes, or weird things at all - except the one difference in how things work with Llama since the update, that I posted about a while ago. And my Webview app is up to date, too.
 
There have been a number if articles in the net about this. If you Google "Android System Webview Problems" you"ll see a number of them. There is one technical article that's over my head at:
http://stackoverflow.com/questions/...ndroid-apps-tombstones-are-disabled-on-jb-mr2
Also, on my Verizon M8, it effected a number of games, my email app, one web browser and a few others. I just hope they fix it. Like I said earlier, I wrote to Google earlier this evening but haven't heard back.
 
Last edited:
Uninstalling the update kept apps from crashing - but did you open yourself up to a security vulnerability? Do the other apps crash because they're non-compliant with how WebView ought to be used?

The problem appeared with the WebView update - maybe it is the problem.

And maybe it's not.

If they release a WebView fix, it was the problem.

If they don't, maybe it's the other apps.

Anyway - don't keep the update off forever - that's not in your best interest.

As for the stackexchange link, let's look at this -

"We found out that in our case it crashed when an exception was thrown. If we used Java.Lang.Exception instead of the C# Exceptions, it stopped crashing."

OK, let me put that in layman's terms.

An exception gets thrown when the code hits an unexpected error condition.

The simple name for that is buggy code.

Android apps can be written in a number of languages - Java is massively the most common - C# is the darling for iOS apps.

So - buggy code using methods common to iOS crash when an unexpected bug hits the app (the app - not WebView).

That case is not WebView's fault.
 
markdoc said:
Earlymon,
I just found this article that may interest you. It says Google knows about the problem but won't fix it. (A very short synopsis) the url is: http://arstechnica.com/security/2015/01/13/google-wont-fix-bug-hitting-60-percent-of-android-phones/
Click to expand...
Yeah.

Not news and Ars Technica sensationalized that to no end.

Let's review -

That's not the same thing being discussed now.

Contrary to articles at the time, the defect in question did not affect all Android before a certain version - it hit specific versions of Android.

So let's look at the title of the article you linked -

"Google won’t fix bug hitting 60 percent of Android phones"

Google said something entirely different - they fixed the security defect in a later version of the operating system (prior to the press getting ahold of it) and that was the solution.

The press went on a feeding frenzy because 1) Apple didn't have the problem (total nonsense, Safari is a WebView/WebKit browser and the security firm they all quoted said Safari failed at a certain revision level, just like Android because WebView is cross platform) and 2) Google refused to update legacy users.

Oh the humanity.

Except when WebView was bound to the operating system Google could not update that for you anyway.

If you have a carrier branded phone in the US, you get your updates from your carrier, not Google - and if it's not carrier branded and not a Nexus, you get your updates from your phone manufacturer - again not Google.

If you were running KitKat at the time that article came out, the problem was already fixed by Google - and if you were on Gingerbread or earlier, the bug hadn't been introduced yet.

If you were affected, rooted and no carrier update was available, a simple root modification fixed it.

If you were vulnerable and not rooted and not getting an update, then using a non-WebView browser solved the problem also.

The site with the test for the vulnerability, the link to the actual security article, and the community of users at Androidforums.com coming together to share security test results is here, contemporary to the time frame of the Ars Technica hype - last January -

http://androidforums.com/threads/google-has-thrown-android-users-under-the-bus.895215/page-3

How does it relate to the issue you have with apps today that WebView is being blamed for?

Users that understood all of the above complained that they ought not be held hostage to carrier OTA updates for important web security bugs. Carriers are very slow and people ought not get screwed because their devices are past the normal update/support lifetime.

That's an excellent point.

Google agreed.

So at the next operating system opportunity - Lollipop - they decoupled WebView from the rom and made it an app that they can control individually from the Play Store - without having to go through carriers or manufacturers to get out security fixes as rapidly as possible.

And it's that update mechanism for WebView that the web today is saying that everyone ought to turn off.

So - Google refused to fix a bug - not true when claimed - problem was Android only - not true when claimed - Google could only solve future problems if they controlled all rom updates like Apple - not true when claimed (remember - news bloggers are not software engineers) - Google threw everyone under the bus - not true when claimed.

You have a WebView that is very particular about inputs and outputs being correct in number, nature and kind - more strict than before.

Something that you would expect apps to respect. Indeed, it's an absolute requirement for all software functions.

And you have apps that use the standard safety net - "If I as a programmer screw that up, I'll just have the code jump over here and do stuff."

That's absolutely common under the theory that you can't test for everything - but it doesn't make it ok, it's just a bandaid until you can find and fix the bug.

I would agree with the thread title here wholeheartedly if it said this -

Major problem with Android System, Some Apps Suck and the latest WebView proves it


EDIT and PS - this quote by Ars Technica - "Android 4.4 and 5.0, which use Blink rather than WebKit for their WebView, are unaffected." - is simply misleading.

In true Ars Technica fashion, they were simply clueless as to what Blink was and where it stood in the food chain.

This article explains - from 2013 - what Blink is - and it's what's being used on desktop Chrome, not Android - http://thenextweb.com/google/2013/0...-on-all-platforms-in-10-weeks-with-chrome-28/

Parts of Blink were included in WebView starting in KitKat - yes. But that didn't magically make the bug go away.

Ars Technica really is a bunch of monkeys with typewriters trying to reproduce Shakespeare.
 
Last edited:
Earlymon,
I can't thank you enough. Clear, concise explanation of a complex situation to most. (Including me) I'm no newbie, but I'm not familiar with the ins & outs of the Android world. BTW, someone deleted 4 of my reviews warning about the issue. When I changed it to "It's okay" 3 stars, it's been left alone. I have to believe that Google is censoring me. ???????
 
As I mentioned in the other thread - it's possible that they are.

Most devs would simply respond to the negative comments - as we all know, Google responds to nothing publicly except to squirt ink in the water like a squid.
 
Cross posting from another thread -

markdoc said:
I think they released another update of Webview. The version is the same but the release date changed from April 24 to 28. I tried it and a lot fewer apps are malfunctioning now. I'll leave the interpreting to EarlyMon.
Click to expand...
Nothing to interpret. :)
EarlyMon said:
The problem appeared with the WebView update - maybe it is the problem.

And maybe it's not.

If they release a WebView fix, it was the problem.

If they don't, maybe it's the other apps.

Anyway - don't keep the update off forever - that's not in your best interest.
Click to expand...

We already know that some of this has been the fault of some app approaches.

Clearly we can now conclude that it wasn't 100% of the apps' faults.

And the best part is - despite causing a lot of grief - it was short term and you got the update without having to wait on HTC and carrier, in addition to Google.

I'd say that the reasonable guess on apps left crashing - is those apps at fault, as described earlier.
 
You must log in or register to reply here.
Back
Top Bottom