-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Crash: InflateException: Binary XML file line #26: Error inflating class org.wordpress.android.l... #9905
Crash: InflateException: Binary XML file line #26: Error inflating class org.wordpress.android.l... #9905
Comments
30-day impact: ~73 per day https://sentry.io/share/issue/9eaadccac3be4c259943d72c46427792/ |
It seems that we are dealing with the same issue as WOOCOMMERCE is dealing with. Thanks to @anitaa1990 and @0nko's comment: woocommerce/woocommerce-android#1141 we saved some time here. We need to make some decision what would be next steps in solving this issue. Suggestion from the comment is to add this: https://developer.android.com/guide/app-bundle/sideload-check |
I think @daniloercoli may have some insights too. |
Restrict to Google Play only devices is super sad. Not sure if we can fix this issue here by adding run-time checks about the availability of Google Play Services, like it was done here #9831 |
Summary This one is really difficult to be resolved as it's hard to find a scenario which will crash the application. Based on the latest reports from the Both Not sure if that is a potential problem, but as I can see the version And newest version: Not sure if it's safe to use the newest library (regarding our UI) but maybe if we update it, it will affect to crashes disappear? @daniloercoli @hypest @maxme @mzorz Any thoughts? |
Been following this interesting issue here! In 1.1.0-alpha-104 they've added support for a custom end icon for the That said, the issues described about sideloading IIUC wouldn't be totally circumvented by this, but given the crashes are not caused by missing WP resources at all I'm pretty confident packaging the resource ourselves should help alleviate the problem. We'd be targeting an alpha product that could have other consequences, but it may be worth a try. Wdyt @marecar3 ? Would you like to try that? |
Hey @mzorz thanks for the information. I am on it. |
90-day impact: ~113 per day Moving this back to the Prioritized Android list in the Groundskeeping project and noting that the open PR needs a review or a nudge. |
90-day impact: ~167 per day |
90-day impact: ~189 per day |
Users affected in the last 90 days: 2400 The open PR for this needs follow-up. |
Users affected in the last 90 days: 1900 (Also see internal reference: paqN3M-7F-p2) |
Users affected in the last 90 days: 1200 |
Not good. @oguzkocer I believe you are in groundskeeping next week? Can you please take a look at this one? |
@elibud I'd be happy to check it out, however there is already a PR for it with quite a bit discussion. From what I understand, it looks like there is still an issue with Here is the PR in question: #10279 |
Got it, thank you @oguzkocer. @khaykov @marecar3 any updates there? |
Hey @elibud |
Maybe we shouldn't wait for them 😕 |
I have gone through the discussions across several issues and had a look at the changes in the PR. I can't help but feel like due to the confusing nature of the issue and uncertainty about the android material library update, we might have lost sight of the main issue. First of, here is my understanding of the situation:
Please correct me if I am wrong about any of these assumptions and ignore the rest of the comment in that case. With these assumptions in mind here are my suggestions:
Hope all this makes sense and let me know if you have any questions or concerns @elibud @marecar3. |
Hey @oguzkocer, Thanks for sharing your thoughts! The main problem for me was how can we use our own drawable for a password as they are using their own drawable - which is crashing the app, in the constructor of the component : @oguzkocer maybe I am missing something but I don't see which approach can be used in order to override the code from the constructor? Thanks! |
@marecar3 I might be missing something as well, but can we not do something like this:
When I check the decompiled code for |
Hey, @oguzkocer thanks for the great suggestion! Let's do it in that way and see what will happen. Here is the new PR: #11172 cc: @elibud @designsimply |
Unfortunately the solution in #11172 did not work for all users. We have one user crashing multiple times in It's very early for The device that crashed is |
Hey @oguzkocer, Thanks for tracking this was. I was also wondering did we manage to solve it. I saw this error in the stack trace: This means that the app can't find a selector that was created in a recent PR fix. |
It seems that users can also experience a side effect of the fix. I assume that there is also a case when Android can't find a drawable reference and somehow it connects other drawable instead. Here is the issue: #11307 |
Thanks for looking into this again @marecar3! The side effect is pretty weird, I honestly don't understand what's going on with these drawables in certain phones. Also, the fact that it's happening for a password field is making me feel uneasy. |
Zero events have been tracked for this crash in 14.6.1 since it was released 9d ago on Apr 22.
Number of users affected in the last 90d from May 19: 384 @oguzkocer the latest version we're seeing this crash happen for in the last 90d is 14.0, which makes me feel it's no longer happening in current releases, but is it still worth keeping the issue open because it's related to the password field (which you mentioned in your latest comment here back in Feb)? |
@designsimply The issue number you used for that query is incorrect and there has been 10 events in |
@oguzkocer thanks for catching that! 🙇 On a closer look, I see that I got the event number correct for the 14.6.1 query but used the wrong link for the number of users affected in the last 90d. I fixed the error in the comment above, and here's another look at the numbers drilling into the 14.7 release: Zero events have been tracked for this crash in 14.7 since it was released 15d ago on May 4. Events in the last 90d: 23,000 The total number of issues is still high, but if you look at the Tags section of the Sentry issue, you can see that it's only happening to version 14.0 and below of the WP Android app. In this case do you think it's okay to close or should we look into trying to resolve this for older versions as well? Note that the number of users affected is probably in the middle for the crashes we're tracking for WPAndroid in GitHub at the moment, which would lead me to prioritize this as medium. wdyt? |
@designsimply That sounds reasonable and when in doubt I almost always prefer to resolve the crash in Sentry since it'll just come back as regression if it's not actually fixed. However, I believe we have several issues for this crash in Sentry, so let me look through them in my next crash monitoring session to just verify that it's not happening in the newer releases. If that's the case, I'll close this and resolve the crashes in Sentry. |
As mentioned in my previous comment, we have a lot of different issues in Sentry for this same crash and unfortunately some of it seems to be still happening in the latest releases. I've merged relevant issues and there seems to be a significant downward trend in the number of crashes in the latest releases, but somehow some of these crashes are ending up in the crash reported in #11966 which is a completely unrelated issue. So working on this issue right now is tricky and the impact of it is unclear. My proposed plan of action is:
If we are not planning to work on this issue in the next couple weeks, which I don't think we should because the numbers are kind of low in Sentry and it's hard to fix this crash, then I think this plan would work. If we want to fix the issue as soon as possible, I think whoever is working on this needs to do a little bit of investigation which I'd be happy to help with. cc @designsimply @yaelirub (as triager for the week, I think you might want to know that the linked issue is related) |
Hey @oguzkocer @khaykov @designsimply 👋 |
I am wondering as I am getting the same crash only periodically by users that we have 100K users per day, we maybe get only 1 per day. Only thing I see is that the line number in our code for XML points to "android:importantForAccessibility="no" on the image file. Could it be something with that attribute? |
Sentry Issue: https://sentry.io/share/issue/136cee38cb79491db456071ee0c686a0/
The text was updated successfully, but these errors were encountered: