philor spotted this once: https://tbpl.mozilla.org/php/getParsedLog.php?id=23281618&tree=Mozilla-Inbound PROCESS-CRASH | java-exception | java.lang.NullPointerExceptionat org.mozilla.gecko.health.BrowserHealthRecorder$1$1.run(BrowserHealthRecorder.java:340) which occurs when initialization finds a null storage. That implies that init (which recently switched to a background thread run) and shutdown (which nulls-out storage) are racing.

Comment on attachment 753040 [details] [diff] [review] Proposed patch. v1 [Approval Request Comment] Bug caused by (feature/regressing bug #): Bug 858742. User impact if declined: No FHR in 23. Testing completed (on m-c, etc.): Baking on m-c for some time. QA overview written up; waiting for QA to execute. Manual testing and automated unit tests. Risk to taking this patch (and alternatives if risky): Slim (improves safety). String or IDL/UUID changes made by this patch: None.

Comment on attachment 753040 [details] [diff] [review] Proposed patch. v1 Approving for uplift as part of the body of work for FHR's first revision on Android.