@sharonlee218: Thanks for the report!!

Could you please help us with a sample file, .apk where you're seeing this issue, OS and details of device and if possible attach a screencast as well for further triaging of the issue?

Thanks!!

@sandeepkumars, this error could be thrown from any WebView, and since ERROR_TOO_MANY_REQUESTS is spilt into three errors:
  1. NetError.ERR_HOST_RESOLVER_QUEUE_TOO_LARGE
  2. NetError.ERR_INSUFFICIENT_RESOURCES
  3. NetError.ERR_OUT_OF_MEMORY
The only thing we could tell is we are seeing NetError.ERR_INSUFFICIENT_RESOURCES the most in the log. 
Another thing is actually ERR_INSUFFICIENT_RESOURCES maps to over 30 errors cases(https://cs.chromium.org/search/?q=ERR_INSUFFICIENT_RESOURCES&sq=package:chromium&type=cs). This makes it even harder to diagnose and repro.
For repro, please try on an Android app with WebView. However, as I mentioned above, there are so many scenarios could trigger the ERROR_TOO_MANY_REQUESTS.

Thank you for providing more feedback. Adding the requester to the cc list.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Any net owner aware of anything that went into m66? Particularly did anything get flipped for chrome but not for webview?

This is kind of impossible for webview dev to make progress.

Another team has noticed regression due to v66 update as I mentioned in the report(https://bugs.chromium.org/p/chromium/issues/detail?id=839524). 
Our team only have metrics for WebView. We don't own any metric regarding chrome. 

Adding a few folks from net/OWNERS. Do you know why WebView might see an increase in ERR_INSUFFICIENT_RESOURCES in M-66?

Hmm, I feel like ERR_INSUFFICIENT_RESOURCES came up in another context with similar timing...

Hi @morlovich, is there a issue related to the context you mentioned? Would you please point me to it?

Any relation to  issue 835340  or  issue 829146 ?

Is it possible that this commit(https://chromium.googlesource.com/chromium/src/+/42cd4a84003a14c2098bc919534a912caaf86bce%5E%21/) limiting cache size that caused the ERR_INSUFFICIENT_RESOURCES error?

That commit (comment #13) landed in 67.0.3385.0. Your initial report implies this regressed in M66.

Hi, @ntfschr. The ERR_INSUFFICIENT_RESOURCES started in v66 and now still going up. Is there any changes in v66 possibly related to causing this error?

No changes which webview team is aware of. morlovich@ did you remember the other context in which ERR_INSUFFICIENT_RESOURCES came up?

Not sure how much I can say on an open bug, but... metrics for Net.ErrorCodesForMainFrame3 on Android Chrome(!) M66(!!).


https://uma.googleplex.com/timeline_v2?sid=3643cdb1e7cb2bace72a3d9461388034 for Googlers here, link courtesy of Brad Lassey. (But do note the scale..)

Here's WebView's dashboard (thanks, paulmiller@):

https://uma.googleplex.com/p/android_webview/timeline_v2/?sid=86c70ea3d4ad48fd194f678e7fb7b28b

We can't directly compare counts between webview and chrome, but the shapes of the graphs differ significantly--I'm actually quite curious as to why that might be happening.

sharonlee218@ do you have metrics going back to early-March? How do these compare with mid-April (before the spike)? How does early-March compare with today?

Hi @ntfschr. Before Apr 18, the error rate is fairly low. The significant increase starts from Apr 18(rollout date of v66.106), and spikes up on Apr 23(rollout date of v66.126). And on May 3 and May 10, it shows spikes again.

Don't think the net stack team can help much here.

When I zoom out and look over the past 90 days, I still don't see anything too out-of-character in recent metrics [1]. I don't know what action webview should take, so I'm going to close this. Please ping this bug if you see further spikes.

[1] https://uma.googleplex.com/p/android_webview/timeline_v2/?sid=9c9e4b204af7e76c4e989db4e74f168d