1157269 - ANR Executing service org.mozilla.firefox_beta/org.mozilla.gecko.background.healthreport.prune.HealthReportPruneService

Status: RESOLVED INVALID

(Android Background Services Graveyard :: Firefox Health Report Service, defect)

Description

[Aaron Train [:aaronmt]]

Attachment: allthreads.log

This is a high volume ANR reported on the Play console dashboard


---- pid 19220 at 2015-01-11 00:37:45 -----
Cmd line: org.mozilla.firefox_beta

JNI: CheckJNI is off; workarounds are off; pins=0; globals=429

DALVIK THREADS:
(mutexes: tll=0 tsl=0 tscl=0 ghl=0)

"main" prio=5 tid=1 TIMED_WAIT
| group="main" sCount=1 dsCount=0 obj=0x41793ea0 self=0x4177a028
| sysTid=19220 nice=-11 sched=0/0 cgrp=apps handle=1074508116
| state=S schedstat=( 0 0 0 ) utm=1359 stm=355 core=0
at java.lang.Object.wait(Native Method)
- waiting on <0x41fa2ea8> (a java.lang.Object)
at java.lang.Object.wait(Object.java:401)
at org.mozilla.gecko.GeckoAppShell.sendEventToGeckoSync(GeckoAppShell.java:513)
at org.mozilla.gecko.gfx.GLController.serverSurfaceDestroyed(GLController.java:146)
at org.mozilla.gecko.gfx.LayerView.access$300(LayerView.java:52)
at org.mozilla.gecko.gfx.LayerView$SurfaceListener.surfaceDestroyed(LayerView.java:547)
at android.view.SurfaceView.updateWindow(SurfaceView.java:586)
at android.view.SurfaceView.onWindowVisibilityChanged(SurfaceView.java:240)
at android.view.View.dispatchWindowVisibilityChanged(View.java:8509)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewGroup.dispatchWindowVisibilityChanged(ViewGroup.java:1096)
at android.view.ViewRootImpl.performTraversals(ViewRootImpl.java:1537)
at android.view.ViewRootImpl.doTraversal(ViewRootImpl.java:1254)
at android.view.ViewRootImpl$TraversalRunnable.run(ViewRootImpl.java:6637)
at android.view.Choreographer$CallbackRecord.run(Choreographer.java:814)
at android.view.Choreographer.doCallbacks(Choreographer.java:614)
at android.view.Choreographer.doFrame(Choreographer.java:584)
at android.view.Choreographer$FrameDisplayEventReceiver.run(Choreographer.java:800)
at android.os.Handler.handleCallback(Handler.java:733)
at android.os.Handler.dispatchMessage(Handler.java:95)
at android.os.Looper.loop(Looper.java:146)
at android.app.ActivityThread.main(ActivityThread.java:5602)
at java.lang.reflect.Method.invokeNative(Native Method)
at java.lang.reflect.Method.invoke(Method.java:515)
at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:1283)
at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:1099)
at dalvik.system.NativeStart.main(Native Method)

Comment 1

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

Aaron, I don't see anything in the logs indicating this is from the HealthReportPruneService - how do you know that? Was it from information in the Google Play dashboard? Also, do you know how long this ANR has been visible for? I haven't made any changes to the PruneService recently.

The logs indicate the issue in GeckoAppShell.sendEventToGeckoSync, which states [1]:

 // Block the current thread until the Gecko event loop is caught up

I wonder if Gecko just isn't running (e.g. it's killed! But then maybe everything Fennec would be killed). I don't think the PruneService relies on Gecko though (e.g. the ProfileInformationCache) so this interaction is altogether weird.

Note that there is no direct use of sendEventToGeckoSync from PruneService [2]. sendEventToGeckoSync seems to be used by shutdown routines (GeckoAppShell), low memory routines (MemoryMonitor), and gfx routines (GLController). If this is actually related to PruneService, my best guess is that our SQL queries are allocating a lot of memory and we're getting a low memory warning.

Might be good to try to figure out if the browser is or is not running when this happens.

[1]: https://mxr.mozilla.org/mozilla-beta/source/mobile/android/base/GeckoAppShell.java#501
[2]: https://mxr.mozilla.org/mozilla-beta/search?string=sendeventtogeckosync&filter=^[^\0]*%24

Comment 2

[Aaron Train [:aaronmt]]

(In reply to Michael Comella (:mcomella) from comment #1)
> Aaron, I don't see anything in the logs indicating this is from the
> HealthReportPruneService - how do you know that? Was it from information in
> the Google Play dashboard? Also, do you know how long this ANR has been
> visible for? I haven't made any changes to the PruneService recently.

This is from the dashboard, these reports re grouped under that signature. Earliest builds reported are on 34.0 so this is nothing new.

Comment 3

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

13:44 <@rnewman> mcomella: re that ANR: presumably main-thread SQL, or we're blocking the healthreport DB and thus stalling out the main activity

Comment 4

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

Taking so I don't lose this but not actively working on it so take it if you'd like to work on it.

Comment 5

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

Not actively working on this.

Comment 6

[Michael Comella (:mcomella) [NI reported issues only: ex-Mozilla]]

We don't do FHR anymore.