New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 694992 link

Starred by 4 users

Issue metadata

Status: Verified
Owner:
Closed: Mar 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug-Regression



Sign in to add a comment

Regression: Print preview crashes in Guest mode.

Project Member Reported by jbanavatu@chromium.org, Feb 22 2017

Issue description

Chrome Version: 58.0.3015.0/9305.0.0 dev-channel Wolf,Glimmer and Peach-pit
OS:Chrome

What steps will reproduce the problem?
(1)Browse as guest>> Open NTP or navigate to any page>> Hit Ctrl+P to open print preview and Observe

Expected: Print preview of the page should be seen.
Actual: Instead, print preview crashes.

This is regression issue as no crash is seen in 57.0.2987.72/9202.35.0
beta channel peppy.

Crash id: f48a1aef40000000

Stack trace:

Thread 0 CRASHED [SIGILL @ 0xb4502e54 ] MAGIC SIGNATURE THREAD
Stack Quality95%Show frame trust levels
0xb4502e54	(chrome -focus_manager.cc:306)	views::FocusManager::SetFocusedViewWithReason(views::View*, views::FocusManager::FocusChangeReason)
0xb4503355	(chrome -focus_manager.cc:64)	views::FocusManager::OnKeyEvent(ui::KeyEvent const&)
0xb450e4e5	(chrome -widget.cc:1150)	views::Widget::OnKeyEvent(ui::KeyEvent*)
0xb3ea5aeb	(chrome -event_dispatcher.cc:191)	ui::EventDispatcher::DispatchEvent(ui::EventHandler*, ui::Event*)
0xb3ea5de1	(chrome -event_dispatcher.cc:139)	ui::EventDispatcher::ProcessEvent(ui::EventTarget*, ui::Event*)
0xb3ea5e6b	(chrome -event_dispatcher.cc:86)	ui::EventDispatcherDelegate::DispatchEventToTarget(ui::EventTarget*, ui::Event*)
0xb3ea5f49	(chrome -event_dispatcher.cc:58)	ui::EventDispatcherDelegate::DispatchEvent(ui::EventTarget*, ui::Event*)
0xb5c6991f	(chrome -event_processor.cc:35)	ui::EventProcessor::OnEventFromSource(ui::Event*)
0xb4a8d707	(chrome -ash_window_tree_host_platform.cc:118)	ash::AshWindowTreeHostPlatform::DispatchKeyEventPostIME(ui::KeyEvent*)
0xb4a8b00d	(chrome -window_tree_host_manager.cc:832)	ash::WindowTreeHostManager::DispatchKeyEventPostIME(ui::KeyEvent*)
0xb5f579a1	(chrome -input_method_base.cc:130)	ui::InputMethodBase::DispatchKeyEventPostIME(ui::KeyEvent*) const
0xb5f58d77	(chrome -input_method_chromeos.cc:383)	ui::InputMethodChromeOS::ProcessUnfilteredKeyPressEvent(ui::KeyEvent*)
0xb5f591df	(chrome -input_method_chromeos.cc:96)	ui::InputMethodChromeOS::DispatchKeyEvent(ui::KeyEvent*, std::unique_ptr<base::Callback<void (bool), (base::internal::CopyMode)1, (base::internal::RepeatMode)1>, std::default_delete<base::Callback<void (bool), (base::internal::CopyMode)1, (base::internal::RepeatMode)1> > >)
0xb5f5922b	(chrome -input_method_chromeos.cc:151)	ui::InputMethodChromeOS::DispatchKeyEvent(ui::KeyEvent*)
0xb4a8e987	(chrome -input_method_event_handler.cc:33)	ash::InputMethodEventHandler::OnKeyEvent(ui::KeyEvent*)
0xb3ea5aeb	(chrome -event_dispatcher.cc:191)	ui::EventDispatcher::DispatchEvent(ui::EventHandler*, ui::Event*)
0xb3ea5cb1	(chrome -event_dispatcher.cc:170)	ui::EventDispatcher::DispatchEventToEventHandlers(std::vector<ui::EventHandler*, std::allocator<ui::EventHandler*> >*, ui::Event*)
0xb3ea5db9	(chrome -event_dispatcher.cc:127)	ui::EventDispatcher::ProcessEvent(ui::EventTarget*, ui::Event*)
0xb3ea5e6b	(chrome -event_dispatcher.cc:86)	ui::EventDispatcherDelegate::DispatchEventToTarget(ui::EventTarget*, ui::Event*)
0xb3ea5f49	(chrome -event_dispatcher.cc:58)	ui::EventDispatcherDelegate::DispatchEvent(ui::EventTarget*, ui::Event*)
0xb5c6997f	(chrome -event_processor.cc:35)	ui::EventProcessor::OnEventFromSource(ui::Event*)
0xb3ea6415	(chrome -event_source.cc:73)	ui::EventSource::DeliverEventToProcessor(ui::Event*)


Attaching screen-cast for reference.
 
Actual.mp4
3.5 MB View Download

Comment 1 by ajha@chromium.org, Feb 22 2017

Owner: osh...@chromium.org
Status: Assigned (was: Untriaged)
Based on code search on 'focus_manager.cc', could be related to https://codereview.chromium.org/2691183002.

oshima@: Could you please take a look at this.

Thank you!

Comment 2 by osh...@chromium.org, Feb 24 2017

Owner: skau@chromium.org
That stack must be from different crash. I've got this

#0 0x7ff6343b38d7 base::debug::StackTrace::StackTrace()
#1 0x7ff6343b344f base::debug::(anonymous namespace)::StackDumpSignalHandler()
#2 0x7ff63452e330 <unknown>
#3 0x7ff629ec4c37 gsignal
#4 0x7ff629ec8028 abort
#5 0x7ff6343b17c2 base::debug::BreakDebugger()
#6 0x7ff6343d57ac logging::LogMessage::~LogMessage()
#7 0x7ff63560dad7 chromeos::PrintersManagerFactory::BuildServiceInstanceFor()
#8 0x7ff62fa232f0 BrowserContextKeyedServiceFactory::BuildServiceInstanceFor()
#9 0x7ff63055ffd1 KeyedServiceFactory::GetServiceForContext()
#10 0x7ff636e3b1ce printing::PrinterBackendProxy::Create()
#11 0x7ff636e3636b PrintPreviewHandler::printer_backend_proxy()
#12 0x7ff636e35a3e PrintPreviewHandler::HandleGetInitialSettings()
#13 0x7ff63205b89c content::WebUIImpl::ProcessWebUIMessage()
#14 0x7ff63205a8e3 _ZN3IPC8MessageTI26ViewHostMsg_WebUISend_MetaSt5tupleIJ4GURLSsN4base9ListValueEEEvE8DispatchIN7content9WebUIImplESA_vMSA_FvRKS3_RKSsRKS5_EEEbPKNS_7MessageEPT_PT0_PT1_T2_
#15 0x7ff63205a7d3 content::WebUIImpl::OnMessageReceived()
#16 0x7ff63202e60f content::WebContentsImpl::OnMessageReceived()
#17 0x7ff631f2978a content::RenderViewHostImpl::OnMessageReceived()
#18 0x7ff631f2e622 content::RenderWidgetHostImpl::OnMessageReceived()
#19 0x7ff631f1d0fa content::RenderProcessHostImpl::OnMessageReceived()
#20 0x7ff6347f8725 IPC::ChannelProxy::Context::OnDispatchMessage()

It's crashing because sync_service is null for guest profile.


Comment 3 by skau@chromium.org, Feb 24 2017

Status: Started (was: Assigned)

Comment 4 by skau@chromium.org, Feb 24 2017

I have a patch for this and should have it out today.

Comment 5 by skau@chromium.org, Feb 28 2017

Chrome Sync team might have a better solution.  Will update soon.

Comment 6 by skau@chromium.org, Feb 28 2017

Cc: skau@chromium.org
Owner: pav...@chromium.org
do we know what might have regressed this? When I search for BuildServiceInstanceFor, there seem to be several new crashes with that in the stack (for various keyed services). See for example issue 696993

Comment 8 by skau@chromium.org, Mar 1 2017

When PrintersManager gets created, it makes a call to a ProfileSyncService which is null.  The actual line, for this particular crash, is here: https://cs.chromium.org/chromium/src/chrome/browser/chromeos/printing/printers_manager_factory.cc?rcl=c1517f207db34e85c4c1744aa7c06b0b6eb24fee&l=66

This occurs for Guest mode because Guest mode is run with the --disable-sync flag which causes the above problem.
Project Member

Comment 9 by bugdroid1@chromium.org, Mar 2 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/233dc1e93732142695fe2407f1ed882454cd1eab

commit 233dc1e93732142695fe2407f1ed882454cd1eab
Author: skau <skau@chromium.org>
Date: Thu Mar 02 21:47:17 2017

Use an in memory store until  crbug.com/688533  is fixed.

In ChromeOS Guest Mode, Chrome runs with --disable-sync which causes
ProfileSyncService to always be null.  Thus, we are unable to create a
ModelTypeStore.  Until this can be fixed, use the in-memory store that's
normally used for testing.

BUG= 694992 

Review-Url: https://codereview.chromium.org/2716973002
Cr-Commit-Position: refs/heads/master@{#454387}

[modify] https://crrev.com/233dc1e93732142695fe2407f1ed882454cd1eab/chrome/browser/chromeos/printing/printers_manager_factory.cc

Comment 10 by skau@chromium.org, Mar 2 2017

Owner: skau@chromium.org
Status: Fixed (was: Started)
This is now mitigated.  A proper fix will land in M59.
Labels: Merge-TBD
[Auto-generated comment by a script] We noticed that this issue is targeted for M-58; it appears the fix may have landed after branch point, meaning a merge might be required. Please confirm if a merge is required here - if so add Merge-Request-58 label, otherwise remove Merge-TBD label. Thanks.

Comment 12 by skau@chromium.org, Mar 3 2017

Labels: -Merge-TBD
Commit was present at branch point and does not require merge.
Status: Verified (was: Fixed)
Verified on 9334.2.0/58.0.3029.6 
Status: Assigned (was: Verified)
Before the branch open the print view it would crash immediately but not now it does still crash but after ~50 seconds so re-opening 

Test build - 9334.2.0/58.0.3029.6 

Comment 15 by skau@chromium.org, Mar 7 2017

Any chance you have a crash log?  I'm trying to repro on a celes.

Comment 16 by skau@chromium.org, Mar 7 2017

It seems to be fine on 9347.0.0/59.0.3033 and we haven't patched anything recently.  Is it possible the crash was due to a different issue?
In guest mode does not collect the crash ids. Attaching the debug logs

It not specific to print view.Open any browser let to stand for 50sec it crashes
only in guest mode
debug-logs_20170307-142520.tgz
2.5 MB Download
This could be because the chrome crash 
https://bugs.chromium.org/p/chromium/issues/detail?id=698057

Will verify again once the above mentioned bug fix is in.

Components: -UI
The fix for the crash mentioned in comment 18 is in, so can we check if this is still happening?
Status: Verified (was: Assigned)
 9334.9.0/58.0.3029.18 
Labels: Merge-TBD
[Auto-generated comment by a script] We noticed that this issue is targeted for M-58; it appears the fix may have landed after branch point, meaning a merge might be required. Please confirm if a merge is required here - if so add Merge-Request-58 label, otherwise remove Merge-TBD label. Thanks.
Labels: -Merge-TBD

Sign in to add a comment