I am getting this log when using realm in my app during development.
E/REALM: /home/jenkins/workspace/m_realm-core_release_5.10.2-JLYB4463STYEHDZYG5MCNCHSR6P7KGQAINRPSTQDTS37Q72ZT66Q@6/src/realm/util/thread.cpp:185: [realm-core-5.10.2] Destruction of mutex in use
2019-11-27 19:24:09.369 2779-2831/com.fhs A/libc: Fatal signal 6 (SIGABRT), code -6 in tid 2831 (RealmFinalizing), pid 2779.
I am using following configurations for Realm.
App level Gradle
kapt 'androidx.databinding:databinding-compiler:3.5.1'
kapt "io.realm:realm-annotations:6.0.1"
kapt "io.realm:realm-annotations-processor:6.0.1"
Project level Gradle
classpath "io.realm:realm-gradle-plugin:6.0.1"
I would like to know what causes this issue to reproduce not everytime, but after heavy usage of app, & steps with which we can handle this?
This is fixed in Realm Core 5.23.7
If you ask nicely, maybe @cmelchior will update Realm Core version in the next release, because that's not yet released on Realm-Java side unfortunately
5.23.7 introduces a regression (sorting properties which have negative floats is broken) so we actually need a 5.23.8 release, which has been delayed by the holidays.
So we have Realm version 7.0.0-beta out now. Reading some other answers for same issue, I am sure Realm version 7.0 would include 5.23.8 release of Realm-core which would ideally fix this issue (Correct me if I am wrong here). When can we expect this new release.
The part of the code resulting in this error has been completely rewritten in 7.0.0, but hopefully, we can have a 6.1 release out by the end of this week.
Facing similar issue. It will be great help if we get release as soon as possible. Appreciate all your efforts and great product!
@cmelchior @Zhuinden One thing I have observed is, this bug which causes above crash does not get logged to Google Crashlytics. Also the app is not fully closed, it throws user to last screen (activity in my case). How can we log this issue to Google crashlytics & can we close the app altogether once this exception occurs ?
This is assumed to have been fixed with 10.1.1 . If this is still happening, please reopen this issue.