1509342 - Crash in java.lang.IllegalStateException: at android.app.ContextImpl.startServiceCommon(ContextImpl.java)

Status: RESOLVED FIXED

(Firefox for Android Graveyard :: General, defect, P1)

Description

[Calixte Denizet (:calixte)]

This bug was filed from the Socorro interface and is
report bp-1e0258ad-63fb-4d18-853b-83eaa0181122.
=============================================================

Java stack trace:

java.lang.IllegalStateException
	at android.app.ContextImpl.startServiceCommon(ContextImpl.java:1577)
	at android.app.ContextImpl.startService(ContextImpl.java:1532)
	at android.content.ContextWrapper.startService(ContextWrapper.java:664)
	at org.mozilla.gecko.BrowserApp.onCreate(BrowserApp.java:620)
	at android.app.Activity.performCreate(Activity.java:7136)
	at android.app.Activity.performCreate(Activity.java:7127)
	at android.app.Instrumentation.callActivityOnCreate(Instrumentation.java:1271)
	at android.app.ActivityThread.performLaunchActivity(ActivityThread.java:2893)
	at android.app.ActivityThread.handleLaunchActivity(ActivityThread.java:3048)
	at android.app.ActivityThread.handleRelaunchActivityInner(ActivityThread.java:4784)
	at android.app.ActivityThread.handleRelaunchActivity(ActivityThread.java:4693)
	at android.app.servertransaction.ActivityRelaunchItem.execute(ActivityRelaunchItem.java:69)
	at android.app.servertransaction.TransactionExecutor.executeCallbacks(TransactionExecutor.java:108)
	at android.app.servertransaction.TransactionExecutor.execute(TransactionExecutor.java:68)
	at android.app.ClientTransactionHandler.executeTransaction(ClientTransactionHandler.java:55)
	at android.app.ActivityThread.handleRelaunchActivityLocally(ActivityThread.java:4743)
	at android.app.ActivityThread.access$3200(ActivityThread.java:199)
	at android.app.ActivityThread$H.handleMessage(ActivityThread.java:1818)
	at android.os.Handler.dispatchMessage(Handler.java:106)
	at android.os.Looper.loop(Looper.java:193)
	at android.app.ActivityThread.main(ActivityThread.java:6669)
	at java.lang.reflect.Method.invoke(Native Method)
	at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:493)
	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:858)

There are 3 crashes (from 3 installations) in nightly 65 with buildid 20181121100030. In analyzing the backtrace, the regression may have been introduced by patch [1] to fix bug 1494748.

[1] https://hg.mozilla.org/mozilla-central/rev?node=555521c70c0f

Comment 1

[Jan Henning [:JanH]]

Presumably the background service was start prohibitions on O, i.e. maybe if we get backgrounded again while still starting up. Can I ask to reconsider a solution that doesn't involve a service?

Comment 2

[Dylan Roeh (:droeh) (he/him)]

Yeah, this looks like a regression from 1494748; I'll figure something out ASAP.

(In reply to Jan Henning [:JanH] from comment #1)
> Presumably the background service was start prohibitions on O, i.e. maybe if
> we get backgrounded again while still starting up. Can I ask to reconsider a
> solution that doesn't involve a service?

Just to be clear, snorp and I are talking about possible approaches to handle this generally in GeckoView (which should also hopefully fix the issue in Fennec) that won't involve running a service, but that's not likely to land and get uplifted in time to unblock Fennec 64. The service approach is just a temporary fix.

Comment 3

[Jan Henning [:JanH]]

Fair enough, thank you.

Comment 4

[Dylan Roeh (:droeh) (he/him)]

Attachment: Move startService call to onResume

I was looking into JobIntentService, but frankly it seems like the simplest thing to do here is just move the startService() call to onResume() to guarantee we're foregrounded when it's called.

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/c59f82a36a28
Move startService() call to BrowserApp.onResume() to guarantee we're foregrounded. r=snorp

Comment 6

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/c59f82a36a28

Comment 7

[Julien Cristau [:jcristau]]

https://crash-stats.mozilla.com/report/index/ae7658ec-36e0-4be7-ac34-216d20181128 crashed on 20181128100103 which has this fix; I guess that's for a separate bug since it's OnResume rather than OnCreate?

Comment 8

[Dylan Roeh (:droeh) (he/him)]

(In reply to Julien Cristau [:jcristau] from comment #7)
> https://crash-stats.mozilla.com/report/index/ae7658ec-36e0-4be7-ac34-
> 216d20181128 crashed on 20181128100103 which has this fix; I guess that's
> for a separate bug since it's OnResume rather than OnCreate?

Reopening this for further investigation; as far as I know we should be guaranteed to be foregrounded in onResume(), so something weird is going on here.

Comment 9

[Dirkjan Ochtman (:djc)]

I just ran into this directly after unlocking my phone (Pixel 2):

https://crash-stats.mozilla.com/report/index/5b7a2896-a20c-48d3-90e0-ce8220181129

Comment 10

[Jan Henning [:JanH]]

Fixed by the backout in bug 1510587:
https://hg.mozilla.org/mozilla-central/rev/e9a34518f078