Realm-java: Unrecoverable error. Bad transaction log in io_realm_internal_SharedGroup.cpp

Created on 7 Apr 2016  路  16Comments  路  Source: realm/realm-java

Transferred from #2402 No.2

Crash log at
http://crashes.to/s/24c6157fa0e

Exception see below

Fatal Exception: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in io_realm_internal_SharedGroup.cpp line 157
       at io.realm.internal.SharedGroup.nativeAdvanceRead(SharedGroup.java)
       at io.realm.internal.SharedGroup.advanceRead(SharedGroup.java:86)
       at io.realm.internal.ImplicitTransaction.advanceRead(ImplicitTransaction.java:33)
       at io.realm.internal.SharedGroupManager.advanceRead(SharedGroupManager.java:74)
       at io.realm.HandlerController.realmChanged(HandlerController.java:333)
       at io.realm.HandlerController.handleMessage(HandlerController.java:100)
       at android.os.Handler.dispatchMessage(Handler.java:98)
       at android.os.Looper.loop(Looper.java:136)
       at android.app.ActivityThread.main(ActivityThread.java:5120)
       at java.lang.reflect.Method.invokeNative(Method.java)
       at java.lang.reflect.Method.invoke(Method.java:515)
       at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:792)
       at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:608)
       at dalvik.system.NativeStart.main(NativeStart.java)
Reproduction-Required T-Bug-Crash
Source
picture beeender
👍3

Most helpful comment

The issue still occurs in Realm 2.1.1. The app crashes because of this issue at about 1% of all sessions, which is pretty serious. Restarting the app seems to clear the transaction log, as our users are only one time affected. The crashes are grouped in Fabric into two stacktraces:

Fatal Exception: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/zaki/fromgit/realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_SharedRealm.cpp line 113
       at io.realm.internal.SharedRealm.nativeBeginTransaction(SharedRealm.java)
       at io.realm.internal.SharedRealm.beginTransaction(SharedRealm.java:219)
       at io.realm.BaseRealm.beginTransaction(BaseRealm.java:328)
       at io.realm.Realm.beginTransaction(Realm.java:127)
       at io.realm.Realm.executeTransaction(Realm.java:1253)
       [...]

Fatal Exception: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/zaki/fromgit/realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_SharedRealm.cpp line 210
       at io.realm.internal.SharedRealm.nativeRefresh(SharedRealm.java)
       at io.realm.internal.SharedRealm.refresh(SharedRealm.java:281)
       at io.realm.HandlerController.realmChanged(HandlerController.java:450)
       at io.realm.HandlerController.handleMessage(HandlerController.java:121)
       at android.os.Handler.dispatchMessage(Handler.java:98)
       at android.os.Looper.loop(Looper.java:150)
       at android.app.ActivityThread.main(ActivityThread.java:5665)
       at java.lang.reflect.Method.invoke(Method.java)
       at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:799)
       at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:689)
picture vonox7 on 15 Nov 2016
👍3

All 16 comments

One of my users reported the above crash.

Caused by io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in io_realm_internal_SharedGroup.cpp line 178
io.realm.internal.SharedGroup.nativePromoteToWrite (SharedGroup.java)
io.realm.internal.e.b (ImplicitTransaction.java:1145)

looks like it happens on the realm call beginTransaction()

Mobile: Samsung SM-G903F on Android 6.0.1

This happened in a doInBackground() method from an AsyncTask

wbervoets picture wbervoets on 3 Nov 2016

The issue still occurs in Realm 2.1.1. The app crashes because of this issue at about 1% of all sessions, which is pretty serious. Restarting the app seems to clear the transaction log, as our users are only one time affected. The crashes are grouped in Fabric into two stacktraces:

Fatal Exception: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/zaki/fromgit/realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_SharedRealm.cpp line 113
       at io.realm.internal.SharedRealm.nativeBeginTransaction(SharedRealm.java)
       at io.realm.internal.SharedRealm.beginTransaction(SharedRealm.java:219)
       at io.realm.BaseRealm.beginTransaction(BaseRealm.java:328)
       at io.realm.Realm.beginTransaction(Realm.java:127)
       at io.realm.Realm.executeTransaction(Realm.java:1253)
       [...]

Fatal Exception: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/zaki/fromgit/realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_SharedRealm.cpp line 210
       at io.realm.internal.SharedRealm.nativeRefresh(SharedRealm.java)
       at io.realm.internal.SharedRealm.refresh(SharedRealm.java:281)
       at io.realm.HandlerController.realmChanged(HandlerController.java:450)
       at io.realm.HandlerController.handleMessage(HandlerController.java:121)
       at android.os.Handler.dispatchMessage(Handler.java:98)
       at android.os.Looper.loop(Looper.java:150)
       at android.app.ActivityThread.main(ActivityThread.java:5665)
       at java.lang.reflect.Method.invoke(Method.java)
       at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:799)
       at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:689)
vonox7 picture vonox7 on 15 Nov 2016
👍3

So that means the errored methods are these

refresh: https://github.com/realm/realm-object-store/blob/master/src/shared_realm.cpp#L559-L584

beginTransaction: https://github.com/realm/realm-object-store/blob/master/src/shared_realm.cpp#L442-L455

The question is, why can the transaction log be corrupted?

@vonox7 My question is this: do you use synchronous queries, or asynchronous queries? Or both?

Zhuinden picture Zhuinden on 15 Nov 2016

The insert is done via realm.executeTransaction { action(realm) }, the queries are all synchronous queries.

vonox7 picture vonox7 on 15 Nov 2016

Android 4x, 5x
Realm 2.2.0
Use synchronous queries and asynchronous queries

Fatal Exception: java.lang.Error: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/cm/Realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_TableQuery.cpp line 1285 at io.realm.AndroidNotifier.throwBackgroundException(AndroidNotifier.java:138) at io.realm.internal.async.QueryUpdateTask.run(QueryUpdateTask.java:128) at io.realm.internal.async.BgPriorityRunnable.run(BgPriorityRunnable.java:34) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:422) at java.util.concurrent.FutureTask.run(FutureTask.java:237) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1112) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:587) at java.lang.Thread.run(Thread.java:841) Caused by io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /Users/cm/Realm/realm-java/realm/realm-library/src/main/cpp/io_realm_internal_TableQuery.cpp line 1285 at io.realm.internal.TableQuery.nativeBatchUpdateQueries(TableQuery.java) at io.realm.internal.TableQuery.batchUpdateQueries(TableQuery.java:500) at io.realm.internal.async.QueryUpdateTask.run(QueryUpdateTask.java:87) at io.realm.internal.async.BgPriorityRunnable.run(BgPriorityRunnable.java:34) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:422) at java.util.concurrent.FutureTask.run(FutureTask.java:237) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1112) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:587) at java.lang.Thread.run(Thread.java:841)

LPFinch picture LPFinch on 23 Nov 2016

Our investigation showed, that the error occurs only on encrypted realms and when writing concurrently from multiple threads. Can we send you a private demo-project that reproduces that (and many other) realm crashes?

vonox7 picture vonox7 on 3 Jan 2017

@vonox7 YES PLEASE!! Please send the demo project to [email protected] and let us know the steps to reproduce it. Thanks A lot!

beeender picture beeender on 3 Jan 2017

@beeender We sent you a mail containing the information ;)

rekt0x picture rekt0x on 3 Jan 2017

INTERNAL LINK: https://secure.helpscout.net/conversation/299975517/8277/?folderId=366141

beeender picture beeender on 4 Jan 2017

We are also facing this issue with our customers. We are using v2.2.1.

Our use case very similar to @wbervoets where we are updating realm (encrypted) in the doInBackground method of AsyncTask.

mehulsutariya picture mehulsutariya on 21 Jan 2017

also having this issue. It does happens in this scenario:

  • Encryption is On

Imagining an activity that, at start, load data from realm and display it in a view (in my case, async on main thread). But at the same time, it also performs a network call (bg), get some data (bg), and insert in realm with insertall (bg).

The crash happens when the user quickly start the activity, and then press back. Repeat multiple times and it can be reproducible.

recoverrelax picture recoverrelax on 27 Jan 2017

Xiaomi Redmi 6.0.1
Realm 2.3.0, encryption is on
Sync and async queries

02-06 14:03:17.077 25645-25645/? E/AndroidRuntime: FATAL EXCEPTION: main PID: 25645 java.lang.Error: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /home/cc/repo/realm/realm-java-release/realm/realm-library/src/main/cpp/io_realm_internal_TableQuery.cpp line 1297 at io.realm.AndroidNotifier.throwBackgroundException(Unknown Source) at io.realm.internal.async.QueryUpdateTask.run(Unknown Source) at io.realm.internal.async.BgPriorityRunnable.run(Unknown Source) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:423) at java.util.concurrent.FutureTask.run(FutureTask.java:237) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1113) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:588) at java.lang.Thread.run(Thread.java:818) Caused by: io.realm.exceptions.RealmError: Unrecoverable error. Bad transaction log in /home/cc/repo/realm/realm-java-release/realm/realm-library/src/main/cpp/io_realm_internal_TableQuery.cpp line 1297 at io.realm.internal.TableQuery.nativeBatchUpdateQueries(Native Method) at io.realm.internal.TableQuery.batchUpdateQueries(Unknown Source) at io.realm.internal.async.QueryUpdateTask.run(Unknown Source)袙 at io.realm.internal.async.BgPriorityRunnable.run(Unknown Source)袙 at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:423)袙 at java.util.concurrent.FutureTask.run(FutureTask.java:237)袙 at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1113)袙 at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:588)袙 at java.lang.Thread.run(Thread.java:818)袙

LPFinch picture LPFinch on 6 Feb 2017

Potential fix for this https://github.com/realm/realm-core/pull/2426

beeender picture beeender on 7 Feb 2017

The 2.3.2 release of Realm Java (containing the 2.3.2 core) fixed the issue for us. 馃憤

rekt0x picture rekt0x on 15 Mar 2017

Thanks for fixing this bug, our crash rate dropped due to this fix by 80% :)

vonox7 picture vonox7 on 15 Mar 2017

We are so happy to see this issue gets fixed!! Also I am very sorry that it took such a long time ... Anyway, thanks a lot for all your help to identify the issue!

beeender picture beeender on 15 Mar 2017
👍1
Was this page helpful?
0 / 5 - 0 ratings

Related issues

jjorian picture jjorian  路  3Comments
bryanspano picture bryanspano  路  3Comments
mithrann picture mithrann  路  3Comments
Merlin1993 picture Merlin1993  路  3Comments
CNyezi picture CNyezi  路  3Comments