Issue metadata
Sign in to add a comment
|
Heap corruption after the update
Reported by
s.h.h.n....@gmail.com,
Mar 10 2018
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.146 Safari/537.36 Steps to reproduce the problem: Not sure. Just updated the Chrome dev version and it's constantly crashing at the boot time. Reproed in Windows 10 RS3. What is the expected behavior? What went wrong? Heap corruption. Attaching what I saw in windbg. I have a crash dump, but it's too big to attach. Did this work before? N/A Chrome version: 66 Channel: canary OS Version: 10.0 Flash Version: Let me know what information you need more. It's not possible to send crash ID since I can't boot up the Chrome dev version.
,
Mar 10 2018
I installed Chrome dev version in Win10 RS4 build and it's working fine. So it might be caused by restoring previous state of my Chrome dev version after the update.
,
Mar 10 2018
,
Mar 13 2018
I left Chrome dev for few days and it started without a problem today. This is most probably because it didn't try to restore previous state (restore previously opened website button didn't appear). I can't check anything now because history is also gone. You can close this case if you can't get any information from crash dump.
,
Mar 15 2018
+wfh: Could you take a quick look? Per #4, you might be able to just close it.
,
Mar 16 2018
wfh might (?) be OOO, adding pamg who might be able to suggest another owner.
,
Mar 16 2018
He'll be back Monday, but in the meantime maybe Bruce can take a look.
,
Mar 16 2018
Heap corruption is tricky because the place where it is detected is not necessarily anywhere near where it happened. It's possible to run Chrome with Application Verifier enabled (run Application Verifier, add chrome.exe, turn off Handle and TLS checking, then run chrome with --no-sandbox) and that may give much more useful crash reports. It would be great if the original reporter could try that. The heap corruption may still be happening but just not being detected anymore. I loaded the crash dump with Chrome's symbol server enabled (https://www.chromium.org/developers/how-tos/debugging-on-windows) in order to get a proper call stack. Perhaps a bug in URL parsing? Here is the stack: 00 ntdll!RtlReportCriticalFailure 01 ntdll!RtlpHeapHandleError 02 ntdll!RtlpLogHeapFailure 03 ntdll!RtlFreeHeap 04 chrome!network::`anonymous namespace'::SimpleURLLoaderImpl::RequestState::~RequestState 05 chrome!std::default_delete<network::(anonymous namespace)::SimpleURLLoaderImpl::RequestState>::operator() 06 chrome!std::unique_ptr<network::(anonymous namespace)::SimpleURLLoaderImpl::RequestState,std::default_delete<network::(anonymous namespace)::SimpleURLLoaderImpl::RequestState> >::~unique_ptr 07 chrome!network::`anonymous namespace'::SimpleURLLoaderImpl::~SimpleURLLoaderImpl 08 chrome!network::`anonymous namespace'::SimpleURLLoaderImpl::~SimpleURLLoaderImpl 09 chrome!std::default_delete<network::SimpleURLLoader>::operator() 0a chrome!std::unique_ptr<network::SimpleURLLoader,std::default_delete<network::SimpleURLLoader> >::~unique_ptr 0b chrome!media_router::DialURLFetcher::~DialURLFetcher 0c chrome!media_router::DialURLFetcher::~DialURLFetcher 0d chrome!std::default_delete<media_router::DialURLFetcher>::operator() 0e chrome!std::unique_ptr<media_router::DialURLFetcher,std::default_delete<media_router::DialURLFetcher> >::reset 0f chrome!media_router::DeviceDescriptionFetcher::ProcessResponse 10 chrome!base::OnceCallback<void (const std::basic_string<char,std::char_traits<char>,std::allocator<char> > &)>::Run 11 chrome!media_router::DialURLFetcher::ProcessResponse 12 chrome!base::internal::FunctorTraits<void (DomainReliabilityInternalsUI::*)(std::unique_ptr<base::Value,std::default_delete<base::Value> >) const,void>::Invoke 13 chrome!base::internal::InvokeHelper<0,void>::MakeItSo 14 chrome!base::internal::Invoker<base::internal::BindState<void (DomainReliabilityInternalsUI::*)(std::unique_ptr<base::Value,std::default_delete<base::Value> >) const,base::internal::UnretainedWrapper<const DomainReliabilityInternalsUI> >,void (std::unique_ptr<base::Value,std::default_delete<base::Value> >)>::RunImpl 15 chrome!base::internal::Invoker<base::internal::BindState<void (DomainReliabilityInternalsUI::*)(std::unique_ptr<base::Value,std::default_delete<base::Value> >) const,base::internal::UnretainedWrapper<const DomainReliabilityInternalsUI> >,void (std::unique_ptr<base::Value,std::default_delete<base::Value> >)>::Run 16 chrome!base::OnceCallback<void (std::unique_ptr<std::basic_string<char,std::char_traits<char>,std::allocator<char> >,std::default_delete<std::basic_string<char,std::char_traits<char>,std::allocator<char> > > >)>::Run 17 chrome!network::`anonymous namespace'::SaveToStringBodyHandler::NotifyConsumerOfCompletion 18 chrome!network::`anonymous namespace'::BodyReader::ReadData 19 chrome!base::RepeatingCallback<void (unsigned int, const mojo::HandleSignalsState &)>::Run 1a chrome!mojo::SimpleWatcher::OnHandleReady 1b chrome!base::OnceCallback<void ()>::Run 1c chrome!base::debug::TaskAnnotator::RunTask 1d chrome!base::MessageLoop::RunTask 1e chrome!base::MessageLoop::DeferOrRunPendingTask 1f chrome!base::MessageLoop::DoWork 20 chrome!base::MessagePumpForIO::DoRunLoop 21 chrome!base::MessagePumpWin::Run 22 chrome!base::RunLoop::Run 23 chrome!content::BrowserThreadImpl::IOThreadRun 24 chrome!content::BrowserThreadImpl::Run 25 chrome!base::Thread::ThreadMain 26 chrome!base::`anonymous namespace'::ThreadFunc 27 kernel32!BaseThreadInitThunk 28 ntdll!RtlUserThreadStart
,
Mar 16 2018
Could this be duplicate of https://bugs.chromium.org/p/chromium/issues/detail?id=816628#c4
,
Mar 16 2018
Looks like it. I got confused briefly because the reporting user agent was 65.0.3325.146, but that's because Chrome 66 was crashing. I compared the stack to that from bug 818168 (a duplicate of 816628) and I'm reasonably confident that this is a duplicate. Thanks for finding it!
,
Jun 23 2018
This bug has been closed for more than 14 weeks. Removing security view restrictions. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by gov...@chromium.org
, Mar 10 2018Labels: Needs-Triage-M65