java.lang.NullPointerException: Attempt to read from field 'org.chromium.chrome.browser.omaha.RequestGenerator org.chromium.chrome.browser.omaha.OmahaDelegate.mRequestGenerator' on a null object reference |
|||
Issue descriptionI got this Chrome crash while trying to repro crbug.com/686662 with the instructions provided on that bug: 02-08 11:54:05.125 19548 20236 E AndroidRuntime: FATAL EXCEPTION: IntentService[omaha] 02-08 11:54:05.125 19548 20236 E AndroidRuntime: Process: com.chrome.canary, PID: 19548 02-08 11:54:05.125 19548 20236 E AndroidRuntime: java.lang.NullPointerException: Attempt to read from field 'org.chromium.chrome.browser.omaha.RequestGenerator org.chromium.chrome.browser.omaha.OmahaDelegate.mRequestGenerator' on a null object reference 02-08 11:54:05.125 19548 20236 E AndroidRuntime: at org.chromium.chrome.browser.omaha.OmahaClient.onHandleIntent(OmahaClient.java:3034) 02-08 11:54:05.125 19548 20236 E AndroidRuntime: at android.app.IntentService$ServiceHandler.handleMessage(IntentService.java:66) 02-08 11:54:05.125 19548 20236 E AndroidRuntime: at android.os.Handler.dispatchMessage(Handler.java:102) 02-08 11:54:05.125 19548 20236 E AndroidRuntime: at android.os.Looper.loop(Looper.java:148) 02-08 11:54:05.125 19548 20236 E AndroidRuntime: at android.os.HandlerThread.run(HandlerThread.java:61) dfalcantara: Could you take a look?
,
Feb 8 2017
I am synced to Cr-Commit-Position: refs/heads/master@{#448969}
,
Feb 8 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/0e1252cce0bae1c59bfd1c1c998f8206258b3123 commit 0e1252cce0bae1c59bfd1c1c998f8206258b3123 Author: nyquist <nyquist@chromium.org> Date: Wed Feb 08 23:09:26 2017 Revert of [Omaha] Make the whole pipeline run each time (patchset #4 id:60001 of https://codereview.chromium.org/2665133005/ ) Reason for revert: Causes crashes on startup in some cases. Original issue's description: > [Omaha] Make the whole pipeline run each time > > * Instead of making OmahaClient run as an IntentService that handles individual > actions, run the whole pipeline, including registering new requests and > POSTing to the server. This has several advantages, including that the > pipeline is somewhat easier to understand. > > * The pipeline now uses one single-shot alarm to determine when to either > generate a new active user request or to trigger a POST. This requires > canceling existing repeating alarms. > > * Redoes the "isOverude" check in handleRegisterActiveRequest to just check > current timestamps instead of also checking that a request doesn't exist. > This is more logical for an "isOverdue" check, and prevents the OmahaClient > from trying to reschedule itself repeatedly until an existing request's > timestamps are screwed up. This should be alright because the request > timestamp and the timestamp for the new request should match in all cases > where the SharedPrerences are valid. > > * Introduces the OmahaDelegate, which masks out calls to various classes that > OmahaClient requires. > > * Replaces all old OmahaClientTests that tested the old pipeline with one that > is more black-boxy and tests the new pipeline. Significant cleanups are still > required here, but the pipeline will be changing further and basic tests are > better than no tests. > > BUG=680817 > > Review-Url: https://codereview.chromium.org/2665133005 > Cr-Commit-Position: refs/heads/master@{#448803} > Committed: https://chromium.googlesource.com/chromium/src/+/c9625f069b73578ab45f49be70a830d48f6141bc TBR=dfalcantara@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=680817, 690054 , 689850, 689894 Review-Url: https://codereview.chromium.org/2685053002 Cr-Commit-Position: refs/heads/master@{#449127} [modify] https://crrev.com/0e1252cce0bae1c59bfd1c1c998f8206258b3123/chrome/android/java/src/org/chromium/chrome/browser/omaha/ExponentialBackoffScheduler.java [modify] https://crrev.com/0e1252cce0bae1c59bfd1c1c998f8206258b3123/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaClient.java [delete] https://crrev.com/613cdf8eeeaac4830b55c4b55b4377269faa2e35/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaDelegate.java [delete] https://crrev.com/613cdf8eeeaac4830b55c4b55b4377269faa2e35/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaDelegateImpl.java [modify] https://crrev.com/0e1252cce0bae1c59bfd1c1c998f8206258b3123/chrome/android/java_sources.gni [modify] https://crrev.com/0e1252cce0bae1c59bfd1c1c998f8206258b3123/chrome/android/javatests/src/org/chromium/chrome/browser/omaha/OmahaClientTest.java
,
Feb 8 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/d92b8f8279b057dee4285b668ec1ade6d52af68e commit d92b8f8279b057dee4285b668ec1ade6d52af68e Author: Tommy Nyquist <nyquist@chromium.org> Date: Wed Feb 08 23:21:59 2017 Revert of [Omaha] Make the whole pipeline run each time (patchset #4 id:60001 of https://codereview.chromium.org/2665133005/ ) Reason for revert: Causes crashes on startup in some cases. Original issue's description: > [Omaha] Make the whole pipeline run each time > > * Instead of making OmahaClient run as an IntentService that handles individual > actions, run the whole pipeline, including registering new requests and > POSTing to the server. This has several advantages, including that the > pipeline is somewhat easier to understand. > > * The pipeline now uses one single-shot alarm to determine when to either > generate a new active user request or to trigger a POST. This requires > canceling existing repeating alarms. > > * Redoes the "isOverude" check in handleRegisterActiveRequest to just check > current timestamps instead of also checking that a request doesn't exist. > This is more logical for an "isOverdue" check, and prevents the OmahaClient > from trying to reschedule itself repeatedly until an existing request's > timestamps are screwed up. This should be alright because the request > timestamp and the timestamp for the new request should match in all cases > where the SharedPrerences are valid. > > * Introduces the OmahaDelegate, which masks out calls to various classes that > OmahaClient requires. > > * Replaces all old OmahaClientTests that tested the old pipeline with one that > is more black-boxy and tests the new pipeline. Significant cleanups are still > required here, but the pipeline will be changing further and basic tests are > better than no tests. > > BUG=680817 > > Review-Url: https://codereview.chromium.org/2665133005 > Cr-Commit-Position: refs/heads/master@{#448803} > Committed: https://chromium.googlesource.com/chromium/src/+/c9625f069b73578ab45f49be70a830d48f6141bc TBR=dfalcantara@chromium.org NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=680817, 690054 , 689850, 689894 Review-Url: https://codereview.chromium.org/2685053002 Cr-Commit-Position: refs/heads/master@{#449127} (cherry picked from commit 0e1252cce0bae1c59bfd1c1c998f8206258b3123) Review-Url: https://codereview.chromium.org/2686783003 . Cr-Commit-Position: refs/branch-heads/3006@{#3} Cr-Branched-From: d6822b0656d0a9a13ca7926806e22b3afa4d2bf9-refs/heads/master@{#448862} [modify] https://crrev.com/d92b8f8279b057dee4285b668ec1ade6d52af68e/chrome/android/java/src/org/chromium/chrome/browser/omaha/ExponentialBackoffScheduler.java [modify] https://crrev.com/d92b8f8279b057dee4285b668ec1ade6d52af68e/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaClient.java [delete] https://crrev.com/4b43540ef1ff4016c6ed9853ce868c9b33d05b94/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaDelegate.java [delete] https://crrev.com/4b43540ef1ff4016c6ed9853ce868c9b33d05b94/chrome/android/java/src/org/chromium/chrome/browser/omaha/OmahaDelegateImpl.java [modify] https://crrev.com/d92b8f8279b057dee4285b668ec1ade6d52af68e/chrome/android/java_sources.gni [modify] https://crrev.com/d92b8f8279b057dee4285b668ec1ade6d52af68e/chrome/android/javatests/src/org/chromium/chrome/browser/omaha/OmahaClientTest.java
,
Feb 9 2017
|
|||
►
Sign in to add a comment |
|||
Comment 1 by asvitk...@chromium.org
, Feb 8 2017