Crash while tracing |
||
Issue descriptionI started recording. Then closed the tab using ctrl + W. Browser crashed. crash/848052ac336a6781 Stack trace: 0x00005586a9a9ea14 (chrome -trace_config.h:200 ) content::TracingControllerImpl::OnMetadataAvailable(std::__1::unique_ptr<base::DictionaryValue, std::__1::default_delete<base::DictionaryValue> >) 0x00005586a9a9f505 (chrome -bind_internal.h:226 ) base::internal::Invoker<base::internal::BindState<void (DomainReliabilityInternalsUI::*)(std::__1::unique_ptr<base::Value, std::__1::default_delete<base::Value> >) const, base::internal::UnretainedWrapper<DomainReliabilityInternalsUI const> >, void (std::__1::unique_ptr<base::Value, std::__1::default_delete<base::Value> >)>::Run(base::internal::BindStateBase*, std::__1::unique_ptr<base::Value, std::__1::default_delete<base::Value> >&&) 0x00005586a95198d4 (chrome -callback.h:105 ) tracing::mojom::Coordinator_StopAndFlush_ForwardToCallback::Accept(mojo::Message*) 0x00005586aab1fe4e (chrome -interface_endpoint_client.cc:413 ) mojo::InterfaceEndpointClient::HandleIncomingMessageThunk::Accept(mojo::Message*) 0x00005586aab28268 (chrome -multiplex_router.cc:880 ) mojo::internal::MultiplexRouter::ProcessIncomingMessage(mojo::internal::MultiplexRouter::MessageWrapper*, mojo::internal::MultiplexRouter::ClientCallBehavior, base::SequencedTaskRunner*) 0x00005586aab27af0 (chrome -multiplex_router.cc:604 ) mojo::internal::MultiplexRouter::Accept(mojo::Message*) 0x00005586aab1f28c (chrome -connector.cc:445 ) mojo::Connector::ReadSingleMessage(unsigned int*) 0x00005586aab1f930 (chrome -connector.cc:474 ) mojo::Connector::ReadAllAvailableMessages() 0x00005586aab31be9 (chrome -callback.h:94 ) mojo::SimpleWatcher::OnHandleReady(int, unsigned int, mojo::HandleSignalsState const&) 0x00005586aaa481ae (chrome -callback.h:65 ) base::debug::TaskAnnotator::RunTask(char const*, base::PendingTask*) 0x00005586aaa615aa (chrome -message_loop.cc:391 ) base::MessageLoop::RunTask(base::PendingTask*) 0x00005586aaa61c04 (chrome -message_loop.cc:403 ) base::MessageLoop::DoWork() 0x00005586aaa6452d (chrome -message_pump_glib.cc:263 ) base::(anonymous namespace)::WorkSourceDispatch(_GSource*, int (*)(void*), void*)
,
Jan 23 2018
I am not completely sure. I accidentally started tracing on my browser, with ~50 tabs (2 profiles). I was not shutting down Chrome, I remember closing chrome://tracing tab, but not sure if I stopped recording or not. Chrome crashed after 10 seconds. Only "memory-infra" category was enabled. I might have had some other category ticked by mistake. Happened just once, didn't try again.
,
Jan 29 2018
I had this too, with a different stack trace but the crash occurred in the same function (TracingControllerImpl::OnMetadataAvailable): crash/c6556784059ef948 In my case, I put the laptop to sleep after starting the trace, and woke it up several hours later. Clicking "Stop" triggered the crash.
,
Jan 29 2018
Thanks for the info. I suspect in both cases we have a full trace buffer which somehow causes the crash. I'll take a look.
,
Mar 19 2018
I guess that this is fixed by crrev.com/c/913572. Crashes that have "OnMetadataAvailable" in their magic signature stopped in versions that have the fix: https://crash.corp.google.com/browse?q=product.name%3D%27Chrome%27%20AND%20STRPOS(expanded_custom_data.ChromeCrashProto.magic_signature_1.name%2C%20%27OnMetadataAvailable%27)%20%3E%200 |
||
►
Sign in to add a comment |
||
Comment 1 by chiniforooshan@chromium.org
, Jan 23 2018