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

Issue 740754 link

Starred by 3 users

Issue metadata

Status: Available
Owner: ----
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 2
Type: Bug



Sign in to add a comment

Browser reliably crashes when viewing a link in Chrome + OS X + web telegram client

Reported by bdow...@gmail.com, Jul 10 2017

Issue description

UserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.115 Safari/537.36

Steps to reproduce the problem:
1. Install telegram web client
2. This is going to be weird:

In telegram, you can 'reply' to a message with a location. By default it picks your initial position, but it gives you a marker to select coordinates. It then renders this as a google map insert -- which when copied is just a google maps link. But when *this widge is clicked*
3. Crashes chrome

I tested this on three friends, all running modern OS X flavors + Chrome + telegram web. 

3. All chrome instances crash

3. 

What is the expected behavior?
This is a google maps link. In context, it should load a maps link with these coordinates.

What went wrong?
Chrome crashed.

Crashed report ID: bc6dfde5-2eea-422e-b762-674c9281a50c

How much crashed? Whole browser

Is it a problem with a plugin? Yes https://github.com/zhukov/webogram 0.5.7.1

Did this work before? N/A 

Chrome version: 59.0.3071.115  Channel: stable
OS Version: OS X 10.12.5
Flash Version: 

Happy to help replicate -- since this is weird.
 
Screen Shot 2017-07-10 at 12.05.29 PM.png
51.2 KB View Download
Labels: Needs-Feedback
Hi bdowney, thanks for reporting this. Can we also get the Server ID of this crash? It's also in chrome://crashes, in parens next to the Crash ID you posted above.

Thanks!

Comment 2 by bdow...@gmail.com, Jul 11 2017

Server ID: 0c1aad3708000000
Project Member

Comment 3 by sheriffbot@chromium.org, Jul 11 2017

Cc: patricia...@chromium.org
Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "patricialor@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Cc: asvitk...@chromium.org
Components: Internals>Core Internals>Sandbox>SiteIsolation
[mac bug triage]

Adding Internals > Core and Internals > Sandbox > SiteIsolation (the components that RenderWidgetHelper and RenderFrameHostImpl belong to), and CCing asvitkine for help in triaging - would you know who this should go to? Thanks.

Stack trace:

0x000000010e5d5def	(Google Chrome Framework -web_contents_impl.cc:2093 )	content::WebContentsImpl::CreateNewWindow(content::SiteInstance*, int, int, int, content::mojom::CreateNewWindowParams const&, content::SessionStorageNamespace*)
0x000000010e371893	(Google Chrome Framework -render_frame_host_impl.cc:1135 )	content::RenderFrameHostImpl::OnCreateNewWindow(int, int, int, content::mojom::CreateNewWindowParams const&, content::SessionStorageNamespace*)
0x000000010e4edd71	(Google Chrome Framework -render_widget_helper.cc:139 )	content::RenderWidgetHelper::OnCreateNewWindowOnUI(mojo::StructPtr<content::mojom::CreateNewWindowParams>, int, int, int, content::SessionStorageNamespace*)
0x000000010e4ee1cf	(Google Chrome Framework -bind_internal.h:214 )	base::internal::Invoker<base::internal::BindState<void (content::RenderWidgetHelper::*)(mojo::StructPtr<content::mojom::CreateNewWindowParams>, int, int, int, content::SessionStorageNamespace*), scoped_refptr<content::RenderWidgetHelper>, base::internal::PassedWrapper<mojo::StructPtr<content::mojom::CreateNewWindowParams> >, int, int, int, base::internal::RetainedRefWrapper<content::SessionStorageNamespace> >, void ()>::Run(base::internal::BindStateBase*)
0x000000010f7f4b10	(Google Chrome Framework -callback.h:91 )	base::debug::TaskAnnotator::RunTask(char const*, base::PendingTask*)
0x000000010f81d4fa	(Google Chrome Framework -message_loop.cc:423 )	base::MessageLoop::RunTask(base::PendingTask*)
0x000000010f81d84b	(Google Chrome Framework -message_loop.cc:434 )	base::MessageLoop::DeferOrRunPendingTask(base::PendingTask)
0x000000010f81dc02	(Google Chrome Framework -message_loop.cc:527 )	base::MessageLoop::DoWork()
0x000000010f821019	(Google Chrome Framework -message_pump_mac.mm:420 )	base::MessagePumpCFRunLoopBase::RunWork()
0x000000010f811a19	(Google Chrome Framework + 0x01a70a19 )	base::mac::CallWithEHFrame(void () block_pointer)
0x000000010f820a3e	(Google Chrome Framework -message_pump_mac.mm:396 )	base::MessagePumpCFRunLoopBase::RunWorkSource(void*)
0x00007fff9d7ff320	(CoreFoundation + 0x000a7320 )	__CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__
0x00007fff9d7e021c	(CoreFoundation + 0x0008821c )	__CFRunLoopDoSources0
0x00007fff9d7df715	(CoreFoundation + 0x00087715 )	__CFRunLoopRun
0x00007fff9d7df113	(CoreFoundation + 0x00087113 )	CFRunLoopRunSpecific
0x00007fff9cd40ebb	(HIToolbox + 0x00030ebb )	RunCurrentEventLoopInMode
0x00007fff9cd40cf0	(HIToolbox + 0x00030cf0 )	ReceiveNextEventCommon
0x00007fff9cd40b25	(HIToolbox + 0x00030b25 )	_BlockUntilNextEventMatchingListInModeWithFilter
0x00007fff9b2d9a53	(AppKit + 0x00046a53 )	_DPSNextEvent
0x00007fff9ba557ed	(AppKit + 0x007c27ed )	-[NSApplication(NSEvent) _nextEventMatchingEventMask:untilDate:inMode:dequeue:]
0x000000010f3a9a0f	(Google Chrome Framework -chrome_browser_application_mac.mm:187 )	__71-[BrowserCrApplication nextEventMatchingMask:untilDate:inMode:dequeue:]_block_invoke
0x000000010f811a19	(Google Chrome Framework + 0x01a70a19 )	base::mac::CallWithEHFrame(void () block_pointer)
0x000000010f3a9953	(Google Chrome Framework -chrome_browser_application_mac.mm:186 )	-[BrowserCrApplication nextEventMatchingMask:untilDate:inMode:dequeue:]
0x00007fff9b2ce3da	(AppKit + 0x0003b3da )	-[NSApplication run]
0x000000010f82185d	(Google Chrome Framework -message_pump_mac.mm:755 )	base::MessagePumpNSApplication::DoRun(base::MessagePump::Delegate*)
0x000000010f820e7b	(Google Chrome Framework -message_pump_mac.mm:292 )	base::MessagePumpCFRunLoopBase::Run(base::MessagePump::Delegate*)
0x000000010f83efd2	(Google Chrome Framework -run_loop.cc:37 )	base::RunLoop::Run()
0x000000010f3af734	(Google Chrome Framework -chrome_browser_main.cc:1977 )	ChromeBrowserMainParts::MainMessageLoopRun(int*)
0x000000010e270a63	(Google Chrome Framework -browser_main_loop.cc:1179 )	content::BrowserMainLoop::RunMainMessageLoopParts()
0x000000010e2732b1	(Google Chrome Framework -browser_main_runner.cc:140 )	content::BrowserMainRunnerImpl::Run()
0x000000010e26c93b	(Google Chrome Framework -browser_main.cc:46 )	content::BrowserMain(content::MainFunctionParams const&)
0x000000010f363c9a	(Google Chrome Framework -content_main_runner.cc:740 )	content::ContentMainRunnerImpl::Run()
0x0000000110b2165b	(Google Chrome Framework -main.cc:179 )	service_manager::Main(service_manager::MainParams const&)
0x000000010f363073	(Google Chrome Framework -content_main.cc:19 )	content::ContentMain(content::ContentMainParams const&)
0x000000010dda4576	(Google Chrome Framework -chrome_main.cc:123 )	ChromeMain
0x00000001098e4d99	(Google Chrome -chrome_exe_main_mac.c:85 )	main
0x00007fffb2f59234	(libdyld.dylib + 0x00005234 )	start
0x00007fffb2f59234	(libdyld.dylib + 0x00005234 )	start
Cc: michaeln@chromium.org mek@chromium.org
Components: -Internals>Core Blink>Storage>DOMStorage
Given it's hitting:

CHECK(session_storage_namespace_impl->IsFromContext(dom_storage_context));

cc'ing content/browser/dom_storage/OWNERS

Comment 6 by jsb...@chromium.org, Jul 12 2017

Naively, it seems to me that the code determining the site instance in render_frame_host_impl.cc:2725 and web_contents_impl.cc:2227 need to match, and they don't in this case?

Cc: -patricia...@chromium.org
Labels: Needs-Milestone

Comment 9 by ajha@chromium.org, Aug 4 2017

Cc: ajha@chromium.org
Labels: Needs-Feedback
Link to the list of the builds with this magic signature on Mac:
=================================================================
https://goto.google.com/apvqm

As per the crash server data not seeing any crashes on Mac on the latest M-60,M-61 or M-62 builds.

bdowney@: Could you please check this on the latest stable(60.0.3112.90) and confirm if this crash is seen as per the scenario mentioned in C#0.

Note: Unable to test this manually, as don't have telephone number(for testing purpose) available as of now to complete the registration on telegram web client.

Comment 10 by bdow...@gmail.com, Aug 4 2017

Confirm crash still on 60.0.03112.90.

Repro by using the original crashing telegram 'location', and then I
generated a new one on the mobile client (telegram for android 4.1.1).

- Brandon
Project Member

Comment 11 by sheriffbot@chromium.org, Aug 4 2017

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "ajha@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Cc: pnangunoori@chromium.org
Labels: Needs-Feedback
Tested on latest stable #60.0.3112.90 and Canary #62.0.3178.0 on Mac 10.12.5 and unable to reproduce the issue mentioned. Please refer the screencast links (https://drive.google.com/open?id=0B3Nz6WzbhmLRTlJxM1NDbVBrNlE).

Observations -- 
1. When tried to reply, it seems that automatically device location is sent but there is no option to edit location.
2. When Google link is copied and pasted no crash is observed (https://drive.google.com/open?id=0B3Nz6WzbhmLRLWstMm9MZGZzOFk).

@bdowney -- Could you please share your device GPU and provide screen-cast for the further investigation of the issue.

Please let us know if we have missed anything.

Thanks in advance.
Status: Untriaged (was: Unconfirmed)
Mac triage: this crash still appears as recently as M64 (i.e., current stable). For example, see report ID 0be9a19da948d25a.

I'm considering this confirmed by the crash reports we have and marking for triage.
Owner: mek@chromium.org
Status: Assigned (was: Untriaged)
Mac triage: assigning directly to mek@ (one of the dom_storage OWNERs)

Comment 15 by mek@chromium.org, Mar 12 2018

Owner: dmu...@chromium.org
dmurph@ is probably a better owner as he is actively working refactoring session storage code. Although unfortunately I don't think his refactoring would actually fix this...
Cc: dmu...@chromium.org
Owner: ----
Status: Available (was: Assigned)

Sign in to add a comment